2024 03 29 - v24.3

This release fix several issues reported since last release and adds some
new features and improvements.  


  * Add option control the "prefetch" used by oracle_fdw COPY/INSERT
    Prior to this change Ora2Pg uses the default "prefetch" of oracle_fdw,
    which at the time of writing is 50. Allowing this to be controlled by an
    Ora2Pg configuration/option gives the option of increased performance at
    the cost of some additional memory on the PostgreSQL side. Thanks to Martin
    Nash for the patch.

  * Modify the behavior of triggers export with EXPORT_INVALID. It used to
    apply to ENABLED or DISABLED triggers instead of real VALID or INVALID
    triggers. Export of INVALID triggers will be controlled by EXPORT_INVALID
    like others objects like functions, packages, etc. This mean that disabled
    triggers that are valid will be exported by default now, this was not
    the case before. Thanks to dcgadmin for the feature request.

  * Add new configuration directive PGTT_NOSUPERUSER. By default the pgtt
    extension is loaded using the superuser privilege when EXPORT_GTT is
    activated. Enabled it if you run the SQL scripts generated using a non
    superuser user. It will use:
    
        LOAD '$libdir/plugins/pgtt';
    
    instead of default:
    
        LOAD 'pgtt';
    
    Thanks to Simon Martin for the feature request.

Here is the full list of changes and acknowledgements:

  - Fix Inf replacement that must only be done with numeric datatype. Thanks to
    gael-efluid for the report.
  - Fix some replacement of OUTER JOIN (+). Thanks to Carens Kurniawan Wijaya
    for the report.
  - Fix schema filter with test function count. Thanks to dcgadmin for the
    report.
  - Fix TEST count objects when a table name is modified. Thanks to korolan
    for the report.
  - Fix issue with multi style comments. Thanks to newtglobal.com for the patch.
  - Fix documentation about EXPORT_INVALID to precise that it also concern
    disabled triggers. Thanks to dcgadmin for the report.
  - Fix missing FOR EACH clause in trigger export after a regression introduced
    by commit fb6b0ad. Thanks to Carens Kurniawan Wijaya for the report.
  - Fix data export for table with a geometry column. Thanks to ruralqiu for the
    report.
  - Revert changes introduced by commit fc7008c, for some obscurs Oracle reasons
    the DBA_SDO_GEOM_METADATA doesn't always exist.
    Always uses ALL_SDO_GEOM_METADATA instead. Thanks to ruralqiu and
    Pierre3939 for the report.

2024 03 07 - v24.2

This release fix several issues reported since last release and adds some
new features and improvements.

  * Allow DATA_EXPORT_ORDER to take a filename at value to be able to
    give a custom table order. The file must contain the ordered list
    of the tables. One table per line in upper case for Oracle. Thanks
    to DataCloudGaze for the feature request.
  * Add progress bar when --oracle_speed is used to avoid waiting for
    the whole data export ends.
  * Add replacement of the BITAND function by the & operator
  * Add option -f, --format to set the output format for the reports.
    It can be html or json. Default to html. Thanks to mgole001 for the
    feature request.
  * Add automatic addition of the partition key to the primary key. Prefix
    all columns with the alias in the query to get data from Oracle.
  * Add information about MSSQL masked columns in SHOW_COLUMN.
  * Add information about columnstore and compression on MSCSQL tables with
    action SHOW_TABLE.
  * Add new configuration directive PARTITION_BY_REFERENCE to defined
    how to export Oracle partition by reference. Possible values are none,
    duplicate or the number of hash partitions to create.
    Value 'none' mean no translation and export of partition by reference
    like before. Value 'duplicate' will duplicate the referenced column
    in the partitioned table and apply the same partitioning from the
    referenced table to the partitioned table. If the value is a number,
    the table will be partitioned with the HASH method using the value
    as the modulo. For example if you set it to 4 it will create 4 HASH
    partitions.
    Default is none to not export the partitions by reference definition.

Here is the full list of changes and acknowledgements:

  - Fix default values in MSSQL function declaration and missing END keyword.
  - Fix parsing of MSSQL function with a single query. Thanks to Saravanan Newt
    for the report.
  - Fix negative Oracle decimal when comparing data. Thanks to es99-24 for the
    report.
  - Fix typos in documentation. Thanks to Simon Martin for the report.
  - Avoid doubling the NAME keyword in the XMLELEMENT() function.
  - Remove clause "REFERENCING OLD AS OLD NEW AS NEW" that is useless and throw
    an error in PostgreSQL
  - Fix DSN for MSSQL in autogenerated configuration file using --init_project.
  - Fix parsing of REFERENCING clause in trigger. Thanks to Carens Kurniawan
    Wijaya for the report.
  - Fix case where ALL_TAB_COLUMNS and ALL_SDO_GEOM_METADATA was used instead
    of USER_TAB_COLUMNS. Thanks to rvanouter for the report.
  - Fix double quoting in partition by reference where clause.
  - Disallow setting of PARTITION_BY_REFERENCE to duplicate when FDW_SERVER
    is set.
  - Fix trigger export with duplicate FOR EACH clause when a REFERENCING clause
    is present. Thanks to Carens Kurniawan Wijaya for the report.
  - Fix typo in variable name in ora2pg_scanner
  - Fix export of default partition for MySQL. Thanks to Priyanshi Gupta for
    the report.
  - Fix TEST_DATA action with issues on foreign server and import foreign schema
    declaration. Thanks to Florent Jardin for the report.
  - Fix export of NOT NULL constraint with column replacement. Thanks to Florent
    Jardin for the report.
  - Prevent reading file ora2pg_stdout_locker when it does not exist. Thanks to
    Florent Jardin for the report.
  - Fix translation of timestamp with precision < 6. Thanks to Andrei Briukhov
    for the report.
  - Fix json_arrayagg with returning clause, type was not translated.
  - Fix data export for table partitioned by reference with duplicate method.
    The query to extract data on Oracle side add the duplicated column and
    perform the join with the referenced table using the FK definition.
  - Fix partition by LIST export with a useless cast to text. Thanks to
    Priyanshi Gupta for the report.
  - Added closing curly bracket for object details. Thanks to andreas42 for
    the patch.
  - Added quotes around value of "human days cost" and "migration level".
    Thanks to andreas42 for the patch.
  - Format object type detail output as JSON array. Thanks to andreas42 for
    the patch.
  - Fix variable declaration in previous commit
  - Fix MSSQL table export with nonexistent column auto_created in version
    bellow 2017. Thanks to Florent Jardin for the report.
  - Use ADD CONSTRAINT syntax to specify name of primary key. Thanks to Martin
    Karlgren for the patch.
  - Skip unwanted work on PG database when option --oracle_speed is enabled,
    especially drop of constraints. Thanks to John Tian for the report.
  - Remove trailing ); from primary key statements when reading from file.
    Thanks to Martin Karlgren for the patch.
  - Fix option in CREATE USER MAPPING for export of DBLINK.
  - Fix translation of MSSQL floating point datatype to use float(n) notation.
  - Exclude from assessment objects in Oracle bin.
  - Add Oracle package HTP and HTF to migration assessment.
  - Fix MSSQL foreign key export with multiple columns referenced.
  - Fix MSSQL export of unique constraints that was merging columns of several
    unique constraints for the same table into a single constraint.
  - Fix case of MSSQL datetime default value 0 that must be converted to
    '1900-01-01 00:00:00'
  - Add rewrite of MSSQL getutcdate() function.
  - Fix MSSQL index type and add compression + columnstore information
  - Add clause IF NOT EXIST to all CREATE EXTENSION calls
  - Fix MSSQL bit data migration through tds_fdw, it is already exported
    as boolean by the fdw.
  - Fix duplicated indexes in MSSQL export.
  - Add export of MSSQL indexes with columns included (CREATE INDEX+INCLUDE).

2023 09 08 - v24.1

This release fix several issues reported since last release and adds some
new features and improvements.  

  - Replace "set feedback off" by "\set QUIET on;" and "set pagesize 0" with
    "\pset pager off". Thanks to Martin Gerhardy for the suggestion.
  - Always add package name into search_path of packages functions. Thanks to
    janopha for the report.
  - Allow to specify a password file to set PG_PWD at PostgreSQL database
    connection.  If the specified file exists on the system, Ora2Pg will read
    the first line to get the password at each call of send_to_pgdb(). It can
    be useful in some situation where the password change during data migration.
    Thanks to Marius Hope for the feature request.
  - Added option --dump_as_json and fixed some json output errors. Thanks to
    Martin Gerhardy for the patch.

Here is the full list of changes and acknowledgements:

  - Fix schema prefixing of type created in packages stored procedures.
    Thanks to janopha for the report.
  - Fix perl function get_schema_condition() to use quote_ident() in generated
    SQL filters. Thanks to franxav06 for the report.
  - Attempt to better test the not null constraint count in Oracle.
  - Fix regression with PSQL_RELATIVE_PATH. Thanks to Ryan Taylor for the
    report.
  - Do not add the partition key to PK if DISABLE_PARTITION is enabled. Thanks
    to pavel-moskotin-db for the report.
  - Filter list of indexes on name instead of the GENERATED column because
    we are missing the ones that have been created automatically by the
    Automatic Indexing feature of Oracle 19c. Thanks to Franck Pachot for
    the report.
  - Fix incompleteness in GRANT action, add grant usage on schema to owner and
    users. Thanks to elexus for the report.
  - Apply missing --blob_to_lo to import_all.sh script.
  - Fix regression in MySQL hash partitions export. Thanks to Shubham Dabriwala
    for the report.
  - Fix Oracle INTERVAL data export with negative value. Thanks to shubham-yb
    for the report.
  - Fix double count of not null constraints for TEST action. Thanks to Simon
    Pane for the patch.
  - Fix replacement of Oracle sys_refcursor in function return type.
  - Fix replacement of SQL script setting from Oracle.
  - Make scripts executable. Thanks to Martin Gerhardy for the patch.
  - Exclude data pump SYS_EXPORT_SCHEMA_.* tables from Oracle export.
  - ora2pg: use env based shebang for perl. Thanks to Martin Gerhardy for
    the patch.
  - Add unsupported clause message for PRAGMA AUTONOMOUS_TRANSACTION when it is
    not rewritten by Ora2Pg. Thanks to Martin Gerhardy for the patch.
  - Allow schema specific definitions of partitioning columns. Thanks to Martin
    Gerhardy for the patch.
  - Fixed invalid variable name in read_grant_from_file. Thanks to Martin
    Gerhardy for the patch.
  - Fix not double quoted column in alter set not null column statement.
    Thanks to leonteq-reisg for the patch.
  - Bug fixes and special treatment for Types with body. Thanks to Martin
    Gerhardy for the patch.
  - Avoid redundant definition of the version. Thanks to Martin Gerhardy for
    the patch.
  - Fix a MySQL exception when the column type is ENUM for version < 5.7.
    Thanks to Code-UV for the report.

2023 07 05 - v24.0

This major release adds support to migration of SQL Server database to
PostgreSQL. It also fixes several issues reported since past height months
and adds some new features and improvements.

  * Enable the use of ALLOW/EXCLUDE directive with SHOW_* reports and throw
    a fatal error if global filters in ALLOW/EXCLUDE are set.
  * Add replacement of DBMS_LOCK.SLEEP with pg_sleep
  * Split estimate cost details per function/procedure/and package function.
  * Add cmin, cmax, ctid to reserved keywords list.
  * Add cost for presence of ADD CONSTRAINT in PLSQL code. It needs constraint
    name stability.
  * Add CLOB_AS_BLOB configuration directive to treat CLOB as BLOB when
    exporting data. When enabled Ora2Pg will apply same behavior on CLOB
    than BLOB with BLOB_LIMIT setting. This could be useful if you have
    large CLOB data. Enabled by default. Thanks to Omar Mebarki for the patch.
  * Allow COPY and TABLE type to use the NULLIF construct. Thanks to Luke Davies
    for the patch.
  * Add new SEQUENCE_VALUES export type to export DDL to set the last values
    of sequences from current Oracle database last values like the following
    statements: ALTER SEQUENCE departments_seq START WITH 290;
    Thanks to sergey grinko for the feature request.
  * Add replacement of Oracle variable : varname into PG :'varname'.
  * Add SQL Server migration to Ora2Pg. Most of the SQL Server objects are
    supported as well as data export. Translation of the TSQL stored
    procedures to plpgsql is complicated because of the lack of statement
    separator in TSQL but as usual Ora2Pg is doing is best to do as much
    work as possible. Migration assessment is also possible with SQL Server
    database. There is some dedicated configuration directives added to
    ora2Pg.conf.
  * Add support to MySQL PARTITION BY KEY() with a translation to HASH
    partitioned table using the PK/UK definition of the table or the
    columns specified in the KEY() clause. Thanks to Shubham Dabriwala
    for the report.
  * Make EXPORT_INVALID configuration directive works with TRIGGER export.
    Until now disabled triggers were not exported, setting EXPORT_INVALID
    to 1 will force the export of disabled triggers. Thanks to chetank-yb
    for the report.
  * Add support of MySQL generated default value on update. For example:
      CREATE TABLE t1 (
        dt DATETIME DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP
      );
    Ora2Pg will translate this syntax into a trigger on the table to force
    the value of the column on an update event.
    Thanks to heysky for the report.
  * Add translation of ST_GEOMETRY data type to PostGis geometry datatype.
  * Replace ROWNUM in target list with a "row_number() over ()" clause. Thanks
    to Rui Pereira for the report.

New configuration directives:

  * Add configuration directive ST_GEOMETRYTYPE_FUNCTION to be able to set the
    function to use to extract the geometry type from a ST_Geometry column.
    Default: ST_GeometryType, example it should be set to sde.ST_GeometryType
    for ArcSDE. Thanks to Albert88373 for the report.
  * Add four new configuration directive to be able to change or prefix the
    functions used to extract information from ST_Geometry object and values.
    - ST_SRID_FUNCTION: Oracle function to use to extract the srid from
      ST_Geometry meta information. Default: ST_SRID, for example it should be
      set to sde.st_srid for ArcSDE.
    - ST_DIMENSION_FUNCTION: Oracle function to use to extract the dimension
      from ST_Geometry meta information. Default: ST_DIMENSION, for example it
      should be set to sde.st_dimention for ArcSDE.
    - ST_ASBINARY_FUNCTION: Oracle function to used to convert an ST_Geometry
      value into WKB format. Default: ST_ASBINARY, for example it should be set
      to sde.st_asbinary for ArcSDE.
    - ST_ASTEXT_FUNCTION: Oracle function to used to convert an ST_Geometry
      value into WKT format. Default: ST_ASTEXT, for example it should be set
      to sde.st_astext for ArcSDE.
    Thanks to Albert88373 for the report.
  * Add INSERT_ON_CONFLICT configuration directive. When enabled this instruct
    Ora2Pg to add an ON CONFLICT DO NOTHING clause to all INSERT statements
    generated for this type of data export. Thanks to Clemens Rieder for the
    feature request.

Backward compatibility:

  * Change the behavior of CASE_INSENSITIVE_SEARCH to allow the use of a
    collation instead of the citext extension. To disable the feature the
    value none can be used. If the migration is not MSSQL this feature is
    disabled.
  * Remove PREFIX_PARTITION configuration directive, it is now replaced by
    the RENAME_PARTITION directive. Previous behavior was to construct the
    partition name from the table name, the partition name and the sub
    partition name if any. The problem is that we often reach the max length
    for an object name and this leads to duplicate partition name. Now, when
    RENAME_PARTITION is enabled the partition tables will be renamed
    following rules:
        <tablename>_part<pos>
    where "pos" is the partition number. For subpartition this is:
       <tablename>_part<pos>_subpart<pos>
    If this is partition/subpartition default:
        <tablename>_part_default
        <tablename>_part<pos>_subpart_default
    This change will break backward comaptibilty, if PREFIX_PARTITION is
    still set, it will simply enable RENAME_PARTITION.
  * Set START value to MINVALUE when a sequence is cycled and that the START
    value is upper that MAXVALUE. Thanks to Shane Borden for the report.

Here is the full list of changes and acknowledgements:

  - Fix MODIFY_STRUCT that was not working with MySQL. Thanks to Code-UV for
    the report.
  - Fix license string in Makefile.PL. Thanks to RodRaen for the report.
  - Do not remove non alphanumeric character in index name. Thanks to gwidt
    for the report.
  - Reorder trigger event when the update of column is not the last one. Thanks
    to tayalarun1 for the report.
  - Fix export of MySQL function containing special characters and white spaces
    in names. Thanks to Shubham Dabriwala for the report.
  - Fix grant export for partitions. Thanks to elexus for the report.
  - Add some other transformation for sqlplus/psql scripts.
  - Remove comma as possible separator for values in DEFINED_PK, it was
    preventing the use of a function with multiple parameters.
  - Fix export of geometry tables when PG_SCHEMA is set.
  - Add rewriting of some sqlplus settings to psql settings.
  - Fix TABLESPACE export for partitioned tables. Thanks to elexus for the
    report.
  - Fix for Issue #1637. Thanks to Simon Pane for the patch.
  - Fix typo in --init_project directories tree generation for sequences
    values.
  - Fix alias in view target list for function call without alias defined in
    MySQL export. Thanks to Shubham Dabriwala for the report.
  - Fix Mysql procedure export when a datatype with precision is used in
    parameter list. Thanks to Shubham Dabriwala for the report.
  - Fix collation on string default values. Thanks to Shubham Dabriwalafor
    the report.
  - Exclude recycle bin object from ALL_TAB_COLUMNS lookup. Thanks to Dave
    Betterton for the report.
  - Fix data types translation (TINYINT|SMALLINT|MEDIUMINT|INTEGER|BIGINT|INT)
    for MySQL table export. Thanks to Shubham Dabriwala for the report.
  - Do not export synonym destination table with table_owner when EXPORT_SCHEMA
    is disabled. Thanks to Priyanshi Gupta for the report.
  - Fully qualify calls to get_sequence_last_values() when PG_SCHEMA is set.
    Thanks to Marius Hope for the report.
  - Fix regression on exporting view as table when VIEW_AS_TABLE contains
    regexp. Thanks to Neil Bower for the report.
  - Fix missing execution of initial command statements at start of TEST_DATA
    action and on both side, those applying to source and destination. Thanks
    to Petter Jacobsen for the report.
  - Fix script to get sequence last value with TEST action. Thanks to franxav06
    for the patch.
  - Prepend PERFORM before call to DBMS_OUTPUT.* when USE_ORAFCE is enabled.
  - Disable USE_ORAFCE when export type is SHOW_REPORT.
  - Extending the enhancement in Pull Request #1621 to the Oracle_FDW user
    mapping. Thanks to Simon Pane for the patch.
  - Changed prefix string to "DIFF:" in test report. Thanks to Simon Pane for
    the patch.
  - Fix cases where %ROWCOUNT was not correctly replaced. Thanks to Rui Pereira
    for the report.
  - Fix parsing of ORACLE_DSN when creating foreign server in COPY mode. Thanks
    to Luke Davies for the report.
  - Fix for Issue #1622, #1627. Thanks to Simon Pane for the patch.
  - Fix index creation with DESC order in COPY action when DROP_INDEXES is
    enabled. Thanks to Luke Davies for the report.
  - Fix for Issue #1610, #1612, #1617 and #1381. Thanks to Simon Pane for the
    patch.
  - Fix typo in sqlnet.ora name (was sqlnet.or). Thanks to Martin Nash for the
    patch.
  - Fix data export, REPLACE_QUERY was not applied. Thanks to Bachev Constantin
    for the report.
  - Fix call to replace_sys_context().
  - Fix timestamp(n) data type translation.
  - Remove use of column GENERATION_EXPRESSION for MySQL version < 5.7.0. Thanks
    to Hans Choi for the report.
  - Fix conversion of DATE datatype to timestamp(0) instead of timestamp. Thanks
    to Akhil Reddy for the report.
  - Add NVARCHAR/NCHAR defaut convertion data types to DATA_TYPE configuration
    directive in ora2pg.conf comments. Thanks to Akhil Reddy for the report.
  - Rename method _get_partitions_type function into _get_partitions_list.
  - Fix synonym export when no schema information is available.
  - Fix support of REFERENCING clause in triggers.
  - Fix partition output file renaming with new RENAME_PARTITION directive.
    Thanks to Rahul Barigidad for the report.
  - Fix export of the ROWNUM clause when there is a variable.
  - Fix sprintf placeholders in geometry queries.
  - Fix some others issues with row count report.
  - Fix row count with destination schema and when the PostgreSQL table
    doesn't exist.
    Thanks to bizen-ya for the report.
  - Fix tests comparison with the different settings of EXPORT_SCHEMA,
    SCHEMA and PG_SCHEMA. Thanks to Marius Hope and bizen-ya for the
    report.
  - Fix St_AsText() call for MySQL data extraction.
  - Add column count comparison for MySQL
  - Export multi column partition by list as an expression with concat
    operator. Multi column partition by list is not supported by PostgreSQL.
  - Fix creation of non existant indexes on partition. Thanks to Shubham
    Dabriwala for the report.
  - Fix MySQL function export when there is no BEGIN clause. Thanks to
    Shubham Dabriwala for the report.
  - Fix MySQL export of unsigned numeric. Thanks to Shubham Dabriwala for
    the report.
  - Fix MySQL output with wrong synthax for JOIN without ON clause. Thanks
    to Shubham Dabriwala for the report.
  - Fix virtual column export. Thanks to Rafal Hollins for the report.
  - Fix index creation on partition with no columns for MySQL export.
    Thanks to Shubham Dabriwala for the report.
  - Fix export of MySQL auto_increment when PG_INTEGER_TYPE is disabled.
    Thanks to Shubham Dabriwala for the report.
  - Fix MySQL subpartition export. Thanks to Sanyam Singhal for the report.
  - Move any INTO clause in CONNECT BY query to the final SELECT on the
    resulting CTE. Thanks to taptarap for the report.
  - Fix translation of MySQL curtime() function in default values. Thanks
    to Shubham Dabriwala for the report.
  - Fix possible "Nested quantifiers in regex" error when exporting package
    with package name containing regex special characters. Thanks to durandm70
    for the report.
  - Fix documentation about use of unique key for ORACLE_COPY.
  - Fix extra comma at end of a CHECK contraint. Thanks to Shubham Dabriwala
    for the report.
  - Always add DROP TYPE statements with package export even if DROP_IF_EXISTS
    is not enabled. Thanks to Rui Pereira for the report.
  - Fix default value of simple dot in MySQL export. Thanks to Shubham
    Dabriwala for the report.
  - Fix regression in data type translation after fix on unsigned numeric type.
    Thanks to Shubham Dabriwala for the report.

2022 10 08 - v23.2

This release fix several issues reported since past height months and
adds some new features and improvements.  

  * Add export of MySQL KEY and LINEAR KEY partitioning, translated as HASH
    partitioning.
  * Allow export of object with dollar sign is his name.
  * Add export of CHECK constraints for MySQL >= 8.0.
  * Add Functional/Expression indexes export from MYSQL.
  * Add export of MySQL virtual column. Thanks to Shubham Dabriwala for the
  * Remove scale and precision of a numeric if the scale is higher than the
    precision. PostgreSQL does not support decimal/numeric attributes where
    the scale is explicitly higher than the precision.
  * Add command line option --drop_if_exists to add statement to drop objects
    before creation if it exists. It corresponds to the DROP_IF_EXISTS
  * Add option -C | --cdc_file to be able to change the name of the default file
    used to store/read SCN per table during export. Default is TABLES_SCN.log in
    the current directory. This is the file written by the --cdc_ready option.
  * Add multiprocess to count rows in PostgreSQL tables (TEST_COUNT) using -P
    command line option.
  * Add support to PostgreSQL 14 procedure with out parameters.
  * Set default PostgreSQL database target version to 14.

New configuration directives:

  * Add configuration directive MVIEW_AS_TABLE and command line option
    --mview_as_table to set which materialized view to export as table.
    By default none. Value must be a list of materialized view name or
    regexp separated by space or comma. If the object name is a materialized
    view and the export type is TABLE, the view will be exported as a create
    table statement. If export type is COPY or INSERT, the corresponding data
    will be exported.
  * Add configuration variable FDW_IMPORT_SCHEMA to rename the schema where
    foreign tables for data migration will be created. If you use several
    instances of ora2pg for data migration through the foreign data wrapper,
    you might need to change the name of the schema for each instance.
    Default: ora2pg_fdw_import
  * Add TRANSFORM_VALUE configuration directive to apply an expression when
    retrieving data from Oracle. For example:
        TRANSFORM_VALUE   CLOB_TABLE[CHARDATA:translate("CHARDATA", chr(0), ' ')]
    to replace all Oracle char(0) in a string by a space character.
  * Add EXCLUDE_COLUMNS configuration directive. Unlike MODIFY_STRUCT
    that is used to redefine a table structure, this directive allow
    to specify a list of columns per table that must be excluded from
    the export. For example:
        EXCLUDE_COLUMNS        T1(nocol1,nocol2) T2(nocol1,nocol2)
  * Add new configuration directive EXPORT_GTT to export Oracle Global Temporary
    Table using syntax recognized by the pgtt extension. For more information see
    https://github.com/darold/pgtt Default is to not export global temporary
    table as they are not supported natively by PostgreSQL.
  * Add new configuration option NO_EXCLUDED_TABLE. By default Ora2Pg exclude
    from export some Oracle "garbage" tables that should never be part of an
    export. This behavior generates a lot of REGEXP_LIKE expressions which are
    slowing down the export when looking at tables. To disable this behavior
    enable this directive, you will have to exclude or clean up later by
    yourself the unwanted tables. The regexp used to exclude the table are
    defined in the array @EXCLUDED_TABLES in lib/Ora2Pg.pm. Note this is behavior
    is independent to the EXCLUDE configuration directive.
   
Backward compatibility:

  * Force rewrite of all invalid date starting with zero year 0000 into 1970-01-01
    when it is a default value and NULL for data. Old behavior was to only replace
    0000-00-00 date.
  * Until now there was a lot of untranslated call to TRUNC(date) because
    Ora2Pg is unable to detect that the parameter is a date or a number.
    The problem is that Oracle has TRUNC(number) too and Ora2Pg try to not
    apply the transformation if there is a doubt. In most of the migration
     have met very few TRUNC(number) so now all call to TRUNC()
    will be converted to date_trunc(). There must be false positive rewrite
    but this should be far less work than the actual situation.

Here is the full list of changes and acknowledgements:

  - Fixed PostgreSQL "relation not found error" in _dump_fdw_table(), PostgreSQL
    search_path was not being used. Thanks to James Schriever for the patch.
  - Fix year and month quoting as reserved words when they are used as aliases.
    Thanks to duursma for the report.
  - Fix conversion of to_number(substr(...)) when PG substr() return empty
    string where Oracle return NULL which make the conversion to numeric fail.
    The fix using (nullif(substr(...), )::numeric) only concern TABLE export.
    Thanks to Menelaos Perdikeas for the report.
  - Add export of MySQL KEY and LINEAR KEY partitioning, now exported as HASH
    partitioning. Thanks to Sanyam Singhal for the report.
  - Remove $ sign from characters that require object name quoting.
  - Fix export of objects with the $ sign in the name. Thanks to yano-rxa and
    duursma for the report.
  - Prevent translation of EXEC when used as alias. Thanks to Rui Pereira for
    the report.
  - Fix MySQL enum data type export with regression introduced by commit 24a476.
    Thanks to Shivansh Gahlot for the report.
  - Rename ORACLE_FDW_TRANSFORM to TRANSFORM_VALUE to apply in all case an
    expression when retrieving data from Oracle. For example:
      TRANSFORM_VALUE   CLOB_TABLE[CHARDATA:translate("CHARDATA", chr(0), ' ')]
    to replace all Oracle char(0) in a string by a space character.
  - Fix add_month() translation failing with some use cases. Thanks to duursma
    for the report.
  - Add export of CHECK constraints for MySQL >= 8.0. Thanks to Rahul
    Barigidad for the report.
  - Fix MySQL unsigned numeric data type conversion. Thanks to Rahul Barigidad
    for the report.
  - Add Functional/Expression indexes fail while exporting from MYSQL. Thanks to
    Shubham Dabriwala for the report.
  - Fix export of descending indexes for MySQL. Thanks to Shubham Dabriwala for
    the report.
  - Force MySQL auto increment sequence to start at 1 when value is 0.
    Thanks to Rahul Barigidad for the report.
  - Fix conversion of MySQL decimal(p,s) to keep the original datatype
    instead of use of real or double.
  - Add export of MySQL virtual column. Thanks to Shubham Dabriwala for the
    report.
  - Fix export of MySQL function with return clause only. Thanks to Shubham
    Dabriwala for the report.
  - Remove scale and precision of a numeric if the scale is higher than the
    precision. PostgreSQL does not support decimal/numeric attributes where
    the scale is explicitly higher than the precision. Thanks to Rahul Barigidad
    for the report.
  - Fix export of comment for views. Thanks to gh-k-murata for the report.
  - Add command line option --drop_if_exists to add statement to drop objects
    before creation if tehy exists. It corresponds to the DROP_IF_EXISTS
    configuration directive. Thanks to Yoni Sade for the feature request.
  - Add option --mview_as_table to documentation.
  - Add configuration directive MVIEW_AS_TABLE and command line option
    --mview_as_table to set which materialized view to export as table.
    By default none. Value must be a list of materialized view name or
    regexp separated by space or comma. If the object name is a materialized
    view and the export type is TABLE, the view will be exported as a create
    table statement. If export type is COPY or INSERT, the corresponding data
    will be exported.
  - Disable EXPORT_GTT when export type is not TABLE. Thanks to gh-k-murata for
    the report.
  - Fix generated external servers wrongly placed in a schema. Thanks to duursma
    for the report.
  - Add configuration variable FDW_IMPORT_SCHEMA to rename the schema where
    foreign tables for data migration will be created. If you use several
    instances of ora2pg for data migration through the foreign data wrapper,
    you might need to change the name of the schema for each instance. Default
    to ora2pg_fdw_import. Thanks to James Schriever for the feature request.
  - Fix wrong conversion of rownum clause when a subquery is used. Thanks to
    Rui Pereira for the report.
  - Escape comma and backslashes in BFILE data export. Thanks to duursma for
    the patch.
  - Fix possible infinite loop in Oracle outer join parsing. Thanks a lot to
    yano-rxa for the report.
  - Remove privileges default settings on views exported from SYNONYMs
  - Add support for Rectangle geometry type. Thanks to duursma for the patch.
  - Fix double replacement of IS NULL/IS NOT NULL when NULL_EQUAL_EMPTY is
    enabled.
  - Add CHECK not null only constraints to not null constraint count.
  - Fix CHECK NOT NULL only constraints that was not exported by generating
    NOT NULL constraints instead. They are exclude from the count of CHECK
    constraint as suggested by Florent Jardin but a count difference persist
    for NOT NULL constraints.
  - Fix TYPE export when SCHEMA and PRESERVE_CASE have different values. Thanks
    to Florent Jardin for the report.
  - Fix custom exception replacement. Thanks to Rui Pereira for the report.
  - Fix Collection and Polygon geometry INTERNAL export.
    Thanks to duursma for the patch.
  - Fix export of efile with parenthesis. Thanks to duursma for the report.
  - Fix wrong column indices used in spatial index extraction. Thanks to duursma
    for the report.
  - Fix call of ST_GeomFromText() with WKT export. Thanks to duursma for the
    report.
  - Disable USE_LOB_LOCATOR with WKT geometry export type, ST_GeomFromText and
    SDO_UTIL.TO_WKTGEOMETRY functions return a CLOB instead of a geometry.
    Thanks to duursma for the report.
  - Fix INTERNAL conversion uses the srid from the object instead of the
    meta-data. Thanks to duursma for the report.
  - Fix regression in data export when REPLACE_AS_BOOLEAN is set. Thanks to
    Juri Berlanda for the report.
  - Fix call to procedure using dblink. Thanks to Rui Pereira for the report.
  - Keep untouched call to DBMS_OUTPUT functions if USE_ORAFCE is enabled.
    Thanks to Sanyam Singhal for the report.
  - Partial fix for MySQL subpartitioning export.
  - Fix partitions export for MySQL. Thanks to Sanyam Singhal for the report.
  - Fix generation of export_all.sh following the operating system.
  - Add information of use of PARALLEL_TABLES with COPY, INSERT and TEST_DATA
    actions. It is also useful with TEST, TEST_COUNT, and SHOW_TABLE if
    --count_rows is used for real row count.
  - Prevent calling real rows count twice with TEST action, and allow it for
    the SHOW_TABLE action.
  - Handle count errors when single process.
  - Move row count wait for all child die to the right place
  - Fix rewrite of nested replace() functions in CHECK constraint. Thanks to
    Menelaos Perdikeas for the report.
  - Fix call of procedures with out parameters when it is not declared in a
    package. Thanks to taptarap for the report.
  - Some minor code improvement. Thanks to Markus Elfring for the patch.
  - Set encoding to read configuration file to utf8.
  - Remove useless multiple semi-colon after END of a function.
  - Fix conversion of regexp_replace() by always appending the 'g' modifier.
    Thanks to Rui Pereira for the report.
  - Fix synonym detection to avoid listing public synonym when no schema is
    specified. Thanks to Dilan Salinda for the report.
  - Fix regexp error with multi-line comment in default value declaration. Thanks
    to taptarap for the report.
  - Add missing sub-partition key in partitioned table primary key. Thanks to
    downvoteit for the report.
  - Replace all invalid date starting with zero year 0000- to 1970-01-01 when it
    is a default value and NULL for data. Old behavior was to only replace
    0000-00-00 date. Thanks to duursma for the report.
  - Enclose \i path to data file beween quote to fix import of table with space
    in their name.
  - Add PARTITION to the list of reserved work and fix custom keywords list from
    ORA_RESERVED_WORDS that was not applied. Thanks to markhooper99 for the
    report.
  - Add LOAD of pgtt extension before creating global temporary table with TABLE
    :export. Thanks to duursma for the report.
  - Fix case where package names should be lower cased. Thanks to Sergey Petrov
    for the patch.
  - Cover more case where ALTER ... OWNER TO should not be generated.
  - Fix case where ALTER ... OWNER TO should not be generated when a view as
    table definition was not exported.
  - Fix sub-partition unique and primary keys that lacks columns part of the
    partition key.  Thanks to downvoteit for the report.
  - Path for function_per_file are mixed case enabled now. Thanks to Sergey
    Petrov for the patch.
  - Fix AUTOINCREMENT script to set last value to sequences for serial an
    identity column for PG version < 12. Thanks to Jaouad Bouras for the report.
  - Fix detection of ENUM data type for MySQL
  - Fix issue when exporting table with a geometry column. The search for
    the SDO_GTYPE need a FROM clause with a FQDN table when the connection
    user is not the same as the table schema. Thanks to Argo64 for the report.
  - Rewrite numeric operation with ADD_MONTH(), LAST_DAY() and TRUNC() to use
    interval. Thanks to duursma for the report.
  - Fix rewrite of CONNECT BY in cursors and just after a BEGIN. Thanks to
    taptarap for the report.
  - Add partition keys to unique index on partitioned table. Thanks to
    downvoteit for the report.
  - Fix case where global variable are tested against NULL. Thanks to duursma
    for the report.
  - Fix remove of %ROWTYPE in function argument and returned data type. Add
    regression test.  Thanks to Eric Delanoe for the report.
  - Fix case clause in autoincrement parameters. Thanks to jbouras for the
    report.
  - Fix typo in ORACLE_FDW_TRANSFORM example.
  - Fix progress bar output in quiet mode. Thanks to Sanyam Singhal for the
    report.
  - Fix error Can't locate object method "gzclose" via package "IO::File".
    Thanks to Sanyam Singhal for the report.
  - Fix cases where translation of function with out parameter was not done
    correctly.
  - Fix translation of function with out parameter that returns a value. For
    example the following Oracle function:
    
      CREATE FUNCTION foo(a int, OUT b int)
      RETURN boolean IS
      BEGIN
        b := a;
        RETURN true;
      END;
    
    is now translated by adding an extra out parameter for the return value:
    
      CREATE OR REPLACE FUNCTION foo (a integer, OUT b integer,
							OUT extra_param boolean)
      RETURNS record AS $body$
      BEGIN
        b := a;
        extra_param := true;
        RETURN;
      END;
      $body$ LANGUAGE PLPGSQL STABLE;
    
    Thanks to Akhil Reddy for the report.
  - Fix undefined database connection handle. Thanks to Alexander for the report
  - Fix case preservation for row count in Oracle side with TEST_COUNT action..
    Thanks to Veka for the report.
  - Only generate the Powershell script when we are running on a Windows
    operating system.
  - Fix #1400 and generate PowerShell script "export_schema.ps1". Thanks to
    moh-hassan for the report.
  - Fix rewriting assignment of a global variable using SELECT INTO. Thanks to
    duursma for the report.
  - Fix partition export for MySQL. Thanks to Sanyam Singhal for the report.
  - Apply WHERE clause to FDW data export.
  - Fix useless ST geometry parsing. Thanks to jieguolove for the report.
  - Replace backslash with slash in BFILE filename when destination data type is
    text or efile.
  - Fix RAW(16)/RAW(32) data export when MOFDIFY_TYPE is used on the column.
    Thanks to Sergey Evseev for the report.
  - Fix ST_SRID() call. Thanks to jieguolove for the report.
  - Skip table data export when the table has no column defined. This was
    generating a fatal error.
  - Fix untranslated function returned data type when there was a comment just
    after. The comment is removed. Thanks to taptarap for the report.
  - Fix other fetching all-column-all-table properties for every table.
    Thanks to Sergey Petrov for the report.
  - Fix fetching all-column-all-table properties for every table. Thanks to
    Sergey Petrov for the report.
  - Remove any comments between RETURN and returned type to not break parsing.
    Thanks to taptarap for the report.
  - Fix global variables in the DECLARE section are not replaced if used with
    a package name. Thanks to taptarap for the report.
  - Remove renaming of dist configuration file under Windows OS. Thanks to
    Julien Monticolo and ohamed Hassan for the report.
  - Fix remaining data export query failure. Thanks to Sung Woo Chang for the
    report.
  - Fix data export, query to retrieve data was broken since change for GTT.
  - Set function as VOLATILE when there is CALL in the body.
  - Add support to PG14 procedure out parameters. Thanks to Rui Pereira for the
    feature request.
  - Fix missing parenthesis in index column expression with input file.
  - Fix missing END keyword after embedded CASE clause in a package function.
    Thanks to taptarap for the report.
  - Fix conversion of dbms_lob.substr() where second and third parameters
    must be inverted. Thanks to taptarap for the report.
  - Fix an other case of wronf NOT NULL detection from input file.
  - Fix detection of NOT NULL constraint in input file.
  - Do not quit on error "Undefined subroutine &Ora2Pg::ReadLine", just
    continue to be able to leverage an Oracle Wallet (SEPS) when no Oracle
    user and password are provided. If you want to use the interactive mode
    to type the username and password at command line you must install the
    Perl package Term::ReadKey before. Thanks to Simon Pane for the report.
  - Fix partitioning by LIST, only the first value of a list was exported.
    Thanks to Sergey Grinko for the report.
  - Fix quoting of DEFAULT NULL. Thanks to Veka for the report.
  - Fix unwanted multiple CALL keywords. Thanks to taptarap for the report.
  - Add assessment counter for FND_* packages.
  - Fix LONG RAW export as bytea in COPY mode. Thanks to Helena Adiduyulmus for
    the report.
  - Add new configuration option NO_EXCLUDED_TABLE.    
    By default Ora2Pg exclude from export some Oracle "garbage" tables that
    should never be part of an export. This behavior generates a lot of
    REGEXP_LIKE expressions which are slowing down the export when looking at
    tables. To disable this behavior enable this directive, you will have to
    exclude or clean up later by yourself the unwanted tables. The regexp used
    to exclude tables are defined in the array @EXCLUDED_TABLES in lib/Ora2Pg.pm
    This behavior is independent to the EXCLUDE configuration directive. Thanks
    to Peter Humaj for the feature request.
  - Replace all remaining CURSORNAME%NOTFOUND with NOT FOUND
  - Change translation to SYSDATE from LOCALTIMESTAMP to statement_timestamp()
    in non PL/SQL code.
  - Prevent append of SECURITY DEFINER when a procedure execute transaction
    control statements (ex: COMMIT). When defined with this clause an error
    is thrown. Thanks to Suman Michael for the report.
2022 02 10 - v23.1

This release fix several issues reported since past four months and
adds some new major features and improvements.

  * Add use of greatest/least functions from new version of Orafce when
    required to return NULL on NULL input like Oracle.
  * ALLOW and EXCLUDE configuration values can now be read from a file.
    Use -a filename or -e filename to specify the list of tables that need
    to be filtered. This is useful if you have a lot of table to filter.
  * Add possibility to use of System Change Number (SCN) for data export or
    data validation by providing a specific SCN.  It can be set at command
    line using the -S or --scn option. You can give a specific SCN or if you
    want to use the current SCN at first connection time set the value to
    'current'. To use this last case the connection user must have the role
    "SELECT ANY DICTIONARY" or "SELECT_CATALOG_ROLE", the current SCN is
    looked at the v$database view.
    Example of use:
        ora2pg -c ora2pg.conf -t COPY --scn 16605281
    This adds the following clause to the query used to retrieve data for example:
        AS OF SCN 16605281
    You can also use th --scn option to use the Oracle flashback capability by
    specifying a timestamp expression instead of a SCN. For example:
        ora2pg -c ora2pg.conf -t COPY --scn "TO_TIMESTAMP('2021-12-01 00:00:00', 'YYYY-MM-DD HH:MI:SS')"
    This will add the following clause to the query used to retrieve data:
        AS OF TIMESTAMP TO_TIMESTAMP('2021-12-01 00:00:00', 'YYYY-MM-DD HH:MI:SS')
    or for example to only retrieve yesterday's data:
        ora2pg -c ora2pg.conf -t COPY --scn "SYSDATE - 1"
  * Add json output format to migration assessment. Thanks to Ted Yu for the patch.
  * Add new TO_CHAR_NOTIMEZONE configuration directive to remove any timezone
    information into the format part of the TO_CHAR() function. Disabled by default.
    Thanks to Eric Delanoe for the report.
    Note that the new default setting breaks backward compatibility, old behavior
    was to always remove the timezone part.
  * Add new configuration directive FORCE_IDENTITY_BIGINT. Usually identity
    column must be bigint to correspond to an auto increment sequence so
    Ora2Pg always force it to be a bigint. If, for any reason you want
    Ora2Pg to respect the DATA_TYPE you have set for identity column then
    disable this directive.
  * Add command line option --lo_import. By default Ora2Pg imports Oracle BLOB
    as bytea, the destination column is created
    using the bytea data type. If you want to use large object instead of bytea,
    just add the --blob_to_lo option to the ora2pg command. It will create the
    destination column as data type Oid and will save the BLOB as a large object
    using the lo_from_bytea() function. The Oid returned by the call to
    lo_from_bytea() is inserted in the destination column instead of a bytea.
    Because of the use of the function this option can only be used with actions
    SHOW_COLUMN, TABLE and INSERT. Action COPY is not allowed.
    If you want to use COPY or have huge size BLOB ( > 1GB) than can not be
    imported using lo_from_bytea() you can add option --lo_import to the
    ora2pg command. This will allow to import data in two passes:
    1) Export data using COPY or INSERT will set the Oid destination column
       for BLOB to value 0 and save the BLOB value into a dedicated file. It
       will also create a Shell script to import the BLOB files into the
       database using psql command \lo_import and to update the table Oid
       column to the returned large object Oid. The script is named
       lo_import-TABLENAME.sh
    2) Execute all scripts lo_import-TABLENAME.sh after setting the
       environment variables PGDATABASE and optionally PGHOST, PGPORT, PGUSER,
       etc. if they do not correspond to the default values for libpq.
       You might also execute manually a VACUUM FULL on the table to remove
       the bloat created by the table update.
    Limitation: the table must have a primary key, it is used to set the
    WHERE clause to update the Oid column after the large object import.
    Importing BLOB using this second method (--lo_import) is very slow so it
    should be reserved to rows where the BLOB > 1GB for all other rows use
    the option --blob_to_lo. To filter the rows you can use the WHERE
    configuration directive in ora2pg.conf.
  * Add command line option --cdc_ready to use current SCN per table when
    exporting data and register them into a file named TABLES_SCN.log This
    can be used for Change Data Capture (CDC) tools.
  * Allow to export only invalid objects when EXPORT_INVALID is set to 2
  * Disable per partition data export when a WHERE clause is define on the
    partitioned table or that a global WHERE clause is defined.

Backward compatibility:

Ora2Pg used to removr any timezone information from the TO_CHAR() format
function. To recover this behavior set TO_CHAR_NOTIMEZONE to 1 in ora2pg.conf

Complete list of changes:

  - Replace PERFORM by CALL when the stored procedure is a procedure. Thanks
    to Rui Pereira for the report.
  - Fix open cursor translation when using is in the query but not as keyword.
    Thanks to taptarap for the report.
  - Fix replacement of global variables in DECLARE section. Thanks to taptarap
    for the report.
  - Fix missing suffix in function name with autonomous transaction when export
    schema was enabled and fix revoke and owner to wrapper function. Thanks to
    Sergey Grinko for the report.
  - Fix export of type declaration in packages without body. Thanks to Sergey
    Grinko.
  - Fix column name duplicates when exporting data of partition. Thanks to
    Sergey Grinko for the report.
  - Fix BLOB export with INSERT mode, call decode() was missing.
  - Fix applying of DEFAULT_PARALLELISM_DEGREE hint that was not working
    anymore for a long time. Thanks to Marcel Pils for the patch.
  - Update documentation about PARALLEL_TABLES and view export. Thanks to
    xinferum for the report.
  - Fix unwanted quote escaping in global variable constant. Thanks to
    sergey grinko for the report.
  - Fix export of global variable when there is function in the default value.
  - Fix end of statements in last merged PR.
  - Add json output format to migration assessment. Thanks to Ted Yu for
    the patch.
  - Fix parsing of package when a comment follow the AS keyword. Thanks to
    Eric Delanoe for the report.
  - Adapt MAXVALUE for identity columns if the datatype has been changed to
    integer.
  - Fix a regression on data validation introduced with commit to fix data
    export of virtual column.
  - Fix Can't locate object method is_pk via package Ora2Pg error
  - Exclude unique keys using expression to validate data.
  - Fix ORDER BY clause for data validation.
  - Fix error on open pragma when encoding is not set.
  - Fix a regression in data export of virtual column. Thanks to Code-UV and
    IgorM12 for the report.
  - Fix a second regression with empty column name in target list to
    retrieve data.
  - Fix PG version to enable virtual column.
  - Fix binmode when it is set to raw or locale to not call encoding() in
    open pragma.
  - Fix regression in export view as table. Thanks to Sebastian Albert for
    the report.
  - Update Copyright year.
  - Quote tables names when necessary during TEST action.
  - Fix undefined call to auto_set_encoding().
  - Add test count of column per table and add output of the PG table struct
    modified to be used with MODIFY_STRUCT.
  - Fix handling of PRESERVE_CASE with update au sequences values
  - Fix handling of PRESERVE_CASE with TEST_DATA
  - Fix unwanted replacement of sysdate operation to epoch. Thanks to taptarap
    for the report.
  - Remove extra END clause at end of package function when a space or a
    comment was present. Thanks to taptarap for the report.
  - Fix missing import of module Encode. Thanks to Menelaos Perdikeas for
    the report.
  - Fix case where data type defined in function was not exported when
    EXPORT_SCHEMA was enabled. Thanks to Eric Bourlon for the report.
  - Fix missing EXECUTE on OPEN CURSOR statements. Thanks to taptarap for
    the report.
  - Fix missing declaration of min() function in Oracle.pm. Thanks to
    nicscanna for the report.
  - Fix SYSDATE subtract of seconds instead of days
  - Fix PERFORM replacement in CTE. Thanks to taptarap for the report.
  - Fix wrong stored procedure code conversion when use types named with
    "default" and broken decode to case translation. Thanks to taptarap
    for the report.
  - Add missing import of FTS indexes in script import_all.sh. Thanks to
    vijaynsheth for the report.
  - Fix another procedure parsing with return. Thanks to Eric Bourlon for
    the report.
  - Fix case where parenthesis are not added to index creation.
  - Add creation of the uuid extension when it is used.
  - Add HTML report of tables and columns with name > 63 characters.
  - Add report of DBMS_ERROR and Quartz Scheduler tables found.
  - Add mark (date?) on columns of DATE data type in Oracle to check if
    it should be translated into date instead of default timestamp.
  - SHOW_COLUMN: mark column data type with (numeric?) when it is a NUMBER
    without precision.
  - SHOW_TABLE+SHOW_COLUMN: Add mark of tables and columns name > 63
    characters
  - Fix translation of TYPE ... AS TABLE OF ...
  - Fix parsing of function call in check constraints. Thanks to Menelaos
    Perdikeas for the report.
  - Fix missing data export file for partitioned tables when TRUNCATE_TABLE
    was disabled. Thanks to Menelaos Perdikeas for the report.
  - Fix named parameter inserted in procedure call with inout parameters.
    Thanks to Rui Pereira for the report.
  - Fix unwanted quoting of index columns clause when there is an operation.
    Thanks to Menelaos Perdikeas for the report.
  - Move comment in procedure parameters before the BEGIN. Thanks to Eric
    Bourlon for the report.
  - Fix parsing of FOR CUSOR followed by a parenthesis. Thanks to Eric Bourlon
    for the report.
  - Fix parsing of TYPE ... IS REF CUSOR declaration in procedures. Thanks to
    Eric Bourlon for the report.
  - Add replacement of SDO_CS.TRANSFORM into ST_Transform. Thanks to mukesh3388
    for the report.
  - Add missing table namer to index renaming.
  - Create a function for index renaming for code reuse.
  - Fix support translation of type VARRAY from store procedure. Thanks to
    Eric Bourlon for the report.
  - Fix conversion of SQL%ROWCOUNT when part of a string concatenation. Thanks
    to boubou191911 for the report.
  - Remove other non alphanumeric character from index name. Thanks to Menelaos
    Perdikeas for the report.
  - Fix date formatting when error is logged with INSERT failure. Thanks to
    xinjirufen for the report.
  - Remove possible comma from index renaming.
  - Fix drop of indexes with renaming when there is a function call. Thanks to
    Menelaos Perdikeas for the report.
  - Fix empty geometry type since the move of ORA2PG_SDO_GTYPE into
    lib/Ora2Pg/Oracle.pm
  - Move most of the Oracle specific code to a dedicated Perl library
    lib/Ora2Pg/Oracle.pm with the same functions as lib/Ora2Pg/MySQL.pm
    This will help to maintain and extend Ora2Pg to other RDMS. There is
    still Oracle database related specific code in the main library but
    it will be also moved later. There should not be any regression or
    usage change with this huge patch.
  - Fix translation of type with not null clause. Thanks to Yasir1811 for
    the report.

2021 11 15 - v23.0

This release fix several issues reported since past five months and
adds some new major features and improvements.

  * Add new option --blob_to_lo that can be used to export BLOB as large
    objects. It can only be used with action SHOW_COLUMN, TABLE and INSERT.
    When used with TABLE action, the BLOB column will be translated into oid
    PostgreSQL data type. When used with the INSERT export action BLOB data
    will be store as large object in the pg_largeobjects table and the oid
    referencing this large object will be stored in the main table instead
    of a bytea.
    It is not possible to use oid with COPY because this feature use function
    lo_from_bytea() that stores the large object in the external table and
    returns the oid.
    This feature works with or without the use of oracle_fdw to import the
    data and option -J can be used to improve the speed of the INSERT import
    provide that there is a numeric unique key on the table.
    Thanks to rodiq for the feature request.
  * Add command line option -W | --where clause to set the WHERE clauses to
    apply to the Oracle query to retrieve data. It can be used multiple time.
    It will override the WHERE configuration directive if there is a global
    WHERE clause or the same table WHERE clause definition. Otherwise the
    clause will be appended.
  * Add data validation feature consisting in comparing data retrieved from a
    foreign table pointing to the source Oracle table and a local PostgreSQL
    table resulting from the data export. By default Ora2Pg will extract 10000
    rows from both side, you can change this value using DATA_VALIDATION_ROWS.
    When it is set to zero all rows of the tables will be compared.
    Data validation requires that the table has a primary key or unique index
    and that the key columns is not a LOB.
    Due to differences in sort behavior between Oracle and PostgreSQL, if the
    collation of unique key columns in PostgreSQL is not 'C', the sort order of
    is different compared to Oracle. In this case the data validation will fail.
    Ora2Pg will stop comparing two tables after 10 errors, result is dumped to
    an output file named data_validation.log.
  * Add DATA_VALIDATION_ORDERING configuration directive enabled by default.
    Order of rows between both sides are different once the data have been
    modified. In this case data must be ordered using a primary key or a
    unique index, that mean that a table without such object can not be
    compared. If the validation is done just after data import in mode single
    process and without any data modification the validation can be done on all
    tables without any ordering.
  * Add DATA_VALIDATION_ERROR to stop validating data from a table after a
    certain amount of row mismatch. Default is to stop after 10 rows
    validation errors.
  * Allow multiprocess for TEST_DATA action to validate data import. Use -P
    or PARALLEL_TABLES to set the number of parallel tables checked. Output
    is now done to a file named data_validation.log saved in the current
    directory.
  * Add replacement of UTL_RAW.CAST_TO_RAW with encode().
  * Add rewrite of XMLTYPE() with xmlparse(DOCUMENT convert_from(..., 'utf-8')).
  * Add VARCHAR_TO_TEXT configuration directive. By default VARCHAR2 without
    size constraint are tranlated into text PG data type. If you want to use
    varchar instead, disable this directive.
  * Add detection of XML function for migration assessment cost.
  * Add DBMS_RANDOM to the list of Oraclism handled by Orafce.
  * Add support to mysql_fdw foreign data wrapper to export data
    PostgreSQL tables. Thanks to Yoni Sade for the feature request.
  * Allow to transform all NUMBER(*,scale) to an other data type by a
    redefinition like NUMBER(*\,2):decimal in the DATA_TYPE configuration
    directive. Thanks to Florent Jardin for the patch.
  * Add information on how to use SSL encrypted connection to documentation.
  * Add TEST_COUNT action to just report the row count diff between Oracle and

Backward compatibility changes:

  - Add FORCE_PLSQL_ENCODING configuration directive. By default Ora2Pg
    encode all functions code to ut8, this sometime can generate double
    encoding. To change this behavior, disable this configuration directive.
    Thanks to rynerisraid and lee-jongbeom for the report.
  - Change behavior regarding RAW columns. Now RAW(16) and RAW(32) columns or
    RAW columns with "SYS_GUID()" as default value are now automatically
    translated into uuid. Data will be automatically migrated as PostgreSQL
    uuid data type provided by the "uuid-ossp" extension. To recover the old
    behavior to export data as bytea whatever is the precision, the following
    must be set with DATA_TYPE configuration:  RAW(16):bytea,RAW(32):bytea

Complete list of changes:

  - Fix USE_LOB_LOCATOR handling.
  - Fix data validation using oracle_fdw where zero after decimal is not strip
    unlike with PG.
  - Apply MODIFY_STRUCT redefinition to test actions
  - Fix PG filter when DATA_VALIDATION_ORDERING is disabled
  - Apply RAW to uuid transformation for data validation
  - Apply boolean transformation for data validation
  - Do not export data for virtual column for PG >= 13.
  - Fix wrong replacement function with name including a regexp_* function in
    his name. Thanks to Rui Pereira for the report.
  - Remove comments in the from clause before rewrite outer join (+), the entire
    FROM clause will be rewritten and we don't know where to restore.
  - Fix export of columns information for data verification.
  - Fix TEST_VIEW for row count returned by views to exclude views created in
    extensions.
  - Fix comment on procedures
  - Fix translation of MySQL type UNSIGNED
  - Fix test count of indexes for MySQL database.
  - Fix test MySQL sequence count.
  - Do not display error messages when user and db is first checked in the
    import_all.sh script
  - Fix ordering of check constraints
  - Fix mysql table scan when table name is using reserved word. Thanks to
    Stanley Sung for the report.
  - Fix double BOTH keyword in TRIM function. Thanks to Rui Pereira for the
    report.
  - Fix aliases placed in a wrong way. Thanks to Rui Pereira for the report.
  - Fix parsing of procedure broken on keyword RETURN. Thanks to Pavithra
    Jayasankar.
  - Fix case where default partition is taken as a value. Thanks to Karsten
    Lenz for the report.
  - Fix conversion of NUMBER without precision in PL/SQL code to respect
    settings PG_NUMERIC_TYPE, PG_INTEGER_TYPE and DEFAULT_NUMERIC. Fix
    translation of INTEGER/BINARY_INTEGER that was wrongly exported as
    numeric. Thanks to Philippe Beaudoin for the report.
  - Documentation fix. Thanks to mperdikeas for the patch.
  - Fix case where SQL%ROWCOUNT was not replaced by GET DIAGNOSTIC. Thanks to
    Awdotia Romanowna for the report.
  - Fix quote of unique constraints name. Thanks to Veka for the report.
  - Fix looking at package function metadata when there is a huge amount of
    package.
  - Fix error when trying to remove temporary files.
  - Fix wrong translation of a call to a procedure with PRAGMA AUTONOMOUS
    TRANSACTION through dblink. Thanks to Rui Pereira for the report.
  - Remove schema name in front of index name. Thanks to Menelaos Perdikeas
    for the report.
  - Fix virtual column generated from an other column of the table (supported
    in PG 12). Thanks to Veka for the report.
  - Fix case of columns names in boolean transformation when oracle_fdw is used
    to export data. Thanks to veka for the report.
  - Remove extra parenthesis with sub query and TABLE function. Thanks to Rui
    Pereira for the report.
  - Fix WHERE clause not removed in ROWNUM replacement. Thanks to Rui Pereira
    for the report.
  - Exclude extensions tables from table test count. Thanks to Yoni Sade for the
    report.
  - Fix pg_attribute column adsrc removed in PG 12. Thank to Thorsten Hochreuter
    for the patch.
  - Fix unwanted aliases after row_number() over(). Thanks to Rui Pereira for
    the report.
  - Fix several spelling issues. Thanks to Florian Eckert for the patch.
  - Fix wrong condition to import constraints in import_all.sh. Thanks to
    Thorsten Hochreuter for the report.
  - Fix BITMAP_AS_GIN detection. Thanks to Nishanth Bejgam for the patch.
  - Fix parsing of views from file and add PASSWORD, KEY and REF to the list
    of reserved keywords.
  - Fix replacement of CURSOR ... IS when there is comment after IS.
  - Fix comment in auto generated file global_variables.conf
  - Fix XML data export that was transformed by the call to function
    extract(/).getClobVal(), it is now replaced by a direct call to
    getClobVal().
  - Improve COPY FREEZE data export when FILE_PER_TABLE is enabled, the
    transactions are now managed per individual file and not following
    the main file. Thanks to Yoni Sade for the report.
  - Fix addition to UNLOGGED keyword on foreign table when exporting
    data using oracle_fdw. Thanks to Veka for the report.
  - Fix FK error when using TRUNCATE before data export with oracle_fdw.
  - Fix export of user defined type. Actually type definitions are extracted
    from ALL_SOURCE which contain the original CREATE TYPE and eventually
    all the ALTER TYPE commands. Previously those type as considered as not
    supported by Ora2Pg. 
  - replace date(n) by timestamp.
  - Always remove the fqdn SYS schema before functions call.
  - Add report of GTT in SHOW_TABLE action.
  - Fix empty partition values for Oracle 9i.
  - Add creation of schema in user defined type export when EXPORT_SCHEMA is
    enabled to fix an error when the schema has not already been created.
  - Fix some wordings and exclude from export user defined type starting
    with SYS_PLSQL_ found in a 9i export. It looks that they are internal
    to PL/SQL code.
  - Exclude DBMS_SQL from the DBMS count in migration assessment when
    USE_ORAFCE is enabled.
  - Handle case where indexes name include the schema at create time
  - Fix PL/SQL numeric datatype conversion


2021 07 02 - v22.1

This is a maintenance release to extend the feature of data export through the
oracle_fdw PostgreSQL extension to migration that use the public schema and
do not preserve case.

There is also some other fixes:

  - Fix compile_schema() call that breaks valid function based indexes by
    adding compile_all => FALSE to DBMS_UTILITY.compile_schema().
    Thanks to Pawel Fengler for the patch.
  - Force foreign table for data export as readonly to avoid accidental
    write if import schema is not cleaned.
  - Fix data export to file not possible since last changes for oracle_fdw
    export. Thanks to Niels Jespersen for the report.

2021 06 26 - v22.0

This release fix several issues reported since past three months and
adds some new features and improvements.  I must thanks MigOps Inc who
hire me to drive Oracle to PostgreSQL migrations and to develop Ora2Pg.
It's been a long time that I was looking for such a company and it is
an amazing gift for the 20 years of Ora2Pg. All improvements and new
new features developed during my work at MigOps will be available in
the public GitHub repository, here are the new ones.

  - Add export of data using oracle_fdw when FDW_SERVER is set and export
    type is COPY or INSERT. Multi-process using -P or -J is fully supported
    but option -j is useless in this case. Boolean transformation of some
    columns or data type is also supported. Actually, expect that it works
    just like data migration without oracle_fdw. This can improve the data
    migration speed from 30 to 40% especially for BLOB export.
 
  - Improve export performances with huge number of objects by avoiding join
    between Oracle catalog tables.

  - Set a maximum of assessment score for tables, indexes, sequences,
    partitions, global temporary table and synonym following the number of
    objects.

  - Add detection of XML functions to the assessment cost.

  - Allow to change the assessment cost unit value in the export_all.sh script
    when ora2pg is used with options --init_project and --cost_unit_value.

  - Remove pragma restrict_references from P/PSQL code, it is useless.

  - Add the oracle schema to search_path in SQL files generated and improve
    the migration assessment when USE_ORAFCE is enabled.

  - Apply ALLOW and EXCLUDED filtered stored procedures at package extraction
    level. Previous this patch there was no way to not export some package
    functions or to exclude them from assessment.

  - Add new tests to check sequences last values and number of identity columns
    in both side.

  - Apply ALLOW/EXCLUDE without object to table object by default in TEST
    action.

New configuration directives:

  - Add ORACLE_FDW_TRANSFORM configuration directive to apply a transformation
    to a column when exporting data. Value must be a semicolon separated list of
        TABLE[COLUMN_NAME, <replace code in SELECT target list>]
    For example to replace string 'Oracle' by 'PostgreSQL' in a varchar2 column
    use the following.
        ERROR_LOG[DBMS_TYPE:regexp_replace("DBMS_TYPE",'Oracle','PostgreSQL')]
    Thanks to MigOps for the patch.

  - Add DROP_IF_EXISTS configuration directive to add a statement
    "DROP <OBJECT> IF EXISTS" before creating the object. Can be
    useful in an iterative work. Default is disabled. Thanks to
    dherzhau for the feature request.

Backward compatibility:

    There is a backward compatibility issue with old configuration files
    where FDW_SERVER is set by default. This directive was not used when
    exporting data, this is not the case anymore as it instruct Ora2Pg to
    use the given foreign server to use oracle_fdw to migrate the data.

Here is the full list of changes and acknowledgements:

  - Fix replacement of TO_CLOB() function, now it is just removed and the
    parenthesis are kept. Thanks to Rui Pereira for the report.
  - Fix incorrect detection of cursor on dynamic query. Thanks to Rui
    Pereira for the report.
  - Fix quoting column names with spaces and dots. Dots are replaced by
    underscore. Thanks to Veka for the report.
  - Fix one case where DEFINED_PKEY with PRESERVE_CASE was not handled
    correctly. Thanks to Veka for the report.
  - Fix quoting of reserved keywords in CREATE INDEX columns names. Thanks
    to Veka for the report.
  - Fix column name starting with number not quoted in COMMENT. Thanks to
    Veka for the report.
  - Fix addition of PERFORM on call to stored procedures not prefixed by
    the package name. Thanks to Rui Pereira fo the report.
  - Fix search of ora2pg_conf.dist under Windows instead of ora2pg.conf.dist
    when --init_project is used. Thanks to Julien Monticolo for the report.
  - Fix translation from file of check constraint when created on same column,
    only the last one was exported. Also shortened the prefix for constraint
    naming, ora2pg_ckey becomes o2pc, ora2pg_ukey is now o2pu and ora2pg_fkey
    is renamed into o2pf. Thanks to anvithaprabhu8 for the report.
  - Replace wildcard precision * for numeric by 38.
  - Fix incomplete listagg() conversion. Thanks to avandras for the report.
  - Fix potential problem in last_day conversion when USE_ORAFCE is off and
    a number is added or subtracted to the last day. Thanks to atlterry for
    the report.
  - Do not apply utf8 conversion of comments to input files.
  - Fix termination of last writer process when parallel and quiet mode are
    used together. Thanks to David Harper for the patch.
  - Remove precision in number of digit in timestamp microseconds when
    setting NLS_TIMESTAMP_FORMAT at session startup. Now use:
      ALTER SESSION SET NLS_TIMESTAMP_FORMAT='YYYY-MM-DD HH24:MI:SS.FF'
    instead of '.FF6'. This was possibly the cause of sporadic error:
      ORA-03106: fatal two-task communication protocol error (DBD ERROR: OCIStmtFetch)
    Default is precision 6 so it may not change anything. Thanks to
    Abhijeet Bonde for the report.
  - Fix call to deprecated MySQL spatial function when version is after 5.7.6.
    Thanks to naveenjul29 for the report.
  - Fix false positive detection for nested table with MySQL export.
  - Fix sequence export read from file.
  - Add PG_VERSION to the documentation. Thanks to xinjirufen for the report.
  - Fix error report when preparing query to ALL_IND_COLUMNS IC. Thanks to
    ganeshakorde for the report.
  - ora2pg_scanner: Fix detection of service_name in DSN.
  - Fix error raise_application_error() with named parameters.
  - Fix replacement of out parameters in triggers.
  - Change sequence export result storage to hash instead of array.
  - Fix package export when there is a comment between PACKAGE BODY and the
    name of the package.
  - Fix Perl error "malformed utf-8 character in substitution" when there is
    character not in utf8 in the comment or constants.
  - Fix detection of MySQL FUNCTION vs PROCEDURE for version >= 5.5. Thanks
    to naveenjul29 for the report.
  - Exclude nested tables from the export as it is not supported and it
    always generate an error. A warning is raised.
  - Fix column case in check constraints when PRESERVE_CASE is enabled.
  - Fix search_path in direct PG data export when PG_SCHEMA is set.
  - Fix -Infinity insert for direct PG data export.
  - Fix drop indexes when PRESERVE_CASE is enabled.
  - Remove potential double affectation for function with out parameter.
  - Create immutable to_char function when used in an index.
  - Replace dmake by gmake on Windows installation instruction. Thanks to
    Julien Monticolo for the report.
  - Fix MySQL version conditions. Thanks to Christoph Berg for the report.
  - Fix HASH partitioning for duplicate WITH clause.
  - Fix tests when no schema are set to compare all objects in all schemes.
    Thanks to gp4git and dlc75 for the report.
  - Apply ALLOW/EXCLUDE without object to table object by default in TEST
    action. Thanks to Yony Sade for the feature request.
  - Add DROP_IF_EXISTS configuration directive. Thanks to dherzhau for the
    feature request.
  - Fix regression in removing %ROWTYPE from function parameters. Thanks to
    Eric Delanoe for the report.
  - Fix not adding default values to parameters when this is an OUT parameter.
    Thanks to Eric Delanoe for the report.
  - Fix ALL_DIRECTORIES call be using table name relative to USER_GRANTS.
    Thanks to Yoni Sade for the report.
  - Change all remaining call to static ALL_* tables to a call relative to
    USER_GRANTS. Thanks to Yoni Sade for the report.
  - Fix export or partitioned table with unsupported partitioning type
    like PARTITION BY REFERENCE. The table is created without partition
    and a warning it fired as well as the following message in the output
    file as a comment: -- Unsupported partition type, please check

2021 04 01 - v21.1

This release fix several issues reported since past six months and
as usual adds some new features and improvements.

  * Now that Orafce 3.15.0 has a definition for the REGEXP_* function,
    makes the translation optional to USE_ORAFCE directive.
  * Add set application name in connection to Oracle/MySql/PostgreSQL.
  * Add translation of REGEXP_COUNT() and change assessment cost.
  * Rewrite the way REGEXP_LIKE() is translated into regexp_match to
    support modifiers. This rewrite also fix default behavior between
    Oracle and PostgreSQL.
  * Replace DBMS_LOB.GETLENGTH() by PostgreSQL octet_length() function.
  * Add types correspondences for VARCHAR2 and NVARCHAR2 in DATA_TYPE
    configuration directive.
  * Add autodetection and support of geometry type, srid and dimension
    for ArcGis geometries.
  * Add conversion of default value in function parameters.
  * Add -u | --unit option to ora2pg_scanner to be able to set the
    migration cost unit value globally.
  * Replace DBMS_LOB.SUBSTR() by SUBSTR()
  * Remove TO_CLOB() it is useless, manual cast could be necessary.
  * Replace IS JSON validation clause in CHECK constraints by
        (CASE WHEN $1::json IS NULL THEN true ELSE true END)
    When the code is invalid an error is fired.
  * DISTINCT and UNIQUE are synonym on Oracle

Backward compatibility changes:

  - Force a column to be bigint if this is an identity column. Thanks
    to MigOps.com for the patch.
  - Fix EMPTY_LOB_NULL, enable/disable was inverted, keeping default
    to enabled. Take care that in old ora2pg.conf it is disabled so it
    will break backward compatibility with old configuration.
  - Replace NO_LOB_LOCATOR with USE_LOB_LOCATOR and NO_BLOB_EXPORT
    with ENABLE_BLOB_EXPORT to avoid confusion with double negative
    variable. Backward compatibility is preserved with a warning. 
  - SRID for SDO_GEOMETRY export is now taken from the value not forced
    from the metadata table. 

Here is the full list of changes and acknowledgements:

  - Take Geometry SRID from the data and fallback to SRID defined in
    metadata when not found. Thanks to Sebastian Albert for the report.
  - Fix case where Ora2Pg temporary substitution of '' by placeholder
    was not restored. Thanks to MigOps.com for the patch.
  - Fix identity column export on unsupported Oracle 18c options.
    Thanks to MigOps.com for the patch.
  - Fix export of columns indexes created with single quote.
    Thanks to MigOps.com for the patch.
  - Fix replacement of keyword PROCEDURE by FUNCTION in constraints
    constants definition. Thanks to marie-joechahine for the report.
  - Replace IS JSON validation clause in CHECK constraints. Thanks to
    marie-joechahine for the report and MigOps.com for the patch.
  - Add support to ON OVERFLOW clause in LISTAGG replacement.
    Thanks to MigOps.com for the patch.
  - Fix incorrect handling of HAVING+GROUP BY rewriting.
    Thanks to MigOps.com for the patch.
  - Add replacement of TO_NCHAR by a cast to varchar. Thanks to
    MigOps.com for the patch.
  - Fix replacement of NOTFOUND when there is extra space or new line
    in the WHEN EXIT clause. Thanks to MigOps.com for the patch.
  - Fix a regression in NO_VIEW_ORDERING, it was not taken in account
    anymore. Thanks to RonJojn2 for the report.
  - Replace DATA_TYPE with DTD_IDENTIFIER in MySQL catalog queries for
    version prior 5.5.0. Thanks to zejeanmi for the report.
  - Fix import script to import sequences before tables. Thanks to
    MigOps.com for the patch.
  - Fix detail report of custom type in migration assessment. Thanks
    to MigOps.com for the patch.
  - Fix duplicate schema prefixed to SYNONYM. Thanks to dlc75 for the
    reports.
  - Replace NO_LOB_LOCATOR with USE_LOB_LOCATOR and NO_BLOB_EXPORT with
    ENABLE_BLOB_EXPORT to avoid confusion with double negative variable.
    Thanks to Rob Johnson for the report.
  - Fix some missing replacements of NVL and rewrite !=-1 into != -1.
    Thanks to MigOps.com for the patch.
  - Fix ROWNUM followed by + or - operator and when no aliases are
    provided. Thanks to MigOps.com for the patch.
  - Add DBSFWUSER to the list of user/schema exclusion. Thanks to
    MigOps.com for the patch.
  - Fix regexp to not append subquery aliases on JOIN clause. Thanks
    to Rui Pereira for the report.
  - Handle PRESERVE_CASE and EXPORT_SCHEMA in sequence name. Thanks
    to marie-joechahine for the report.
  - Add CREATE SCHEMA statement to sequence export when EXPORT_SCHEMA
    is enabled. Thanks to marie-joechahine for the report.
  - Fix duplicate index name on subpartition. Thanks to Philippe
    Beaudoin for the report.
  - Exclude sequences used for IDENTITY column (ISEQ$$_). Thanks to
    marie-joechahine for the report.
  - Fix parsing from file of CREATE SEQUENCE. Thanks to Rui Pereira
    for the report.
  - In export_all.sh script use the database owner provided if it is a
    superuser instead of postgres user. Thanks to jjune235 for the
    feature request.
  - Fix parsing of triggers when there is a CASE inside the code.
    Thanks to Rui Pereira for the report.
  - Add set application name in connection to Oracle/MySql/PostgreSQL.
    Thanks to Yoni Sade for the patch.
  - Fix double column alias when replacing ROWNUM. Thanks to Rui
    Pereira for the report.
  - Add translation of the REGEXP_COUNT function and change assessment
    cost.
  - Rewrite the way REGEXP_LIKE is translated into regexp_match to
    support modifiers. This rewrite also fix default behavior between
    Oracle and PostgreSQL.  Thanks to otterrisk for the report.
  - Add IS JSON to assessment. Thanks to marie-joe Chahine for the
    report.
  - Fix multi-columns RANGE partitioning. Thanks to Philippe Beaudoin
    for the report.
  - Improve reordering columns. Sort by fieldsize first, if same size
    then it sorts by original position. Thanks to Sebastien Caunes for
    the patch.
  - Append partition's column to the primary key of the table as it
    must be part of the PK on PostgreSQL. Thanks to xinjirufen for the
    report.
  - Fix partition export where PRESERVE_CASE was applied to Oracle
    side. Thanks to schleb1309 for the report.
  - Fix trigger export with column restriction. Thanks to Sebastien
    Caunes for the report.
  - Update installation information.
  - Fix table reordering following data type. Thanks to Sebastien
    Caunes for the patch.
  - Fix incorrect variable name corresponding to DATA_EXPORT_ORDER
    making this directive inefficient. Thanks to Ron Johnson for the
    report.
  - Fix translation of check constraint when read from file
  - Fix EMPTY_LOB_NULL, enable/disable as inverted, keep default to
    enabled. Take care that in old ora2pg.conf it is disabled so it
    will break backward compatibility with old configuration.
  - Fix false positive detection of input filename is the same as
    output file.
  - Rename variables SCHEMA_ONLY, DATA_ONLY and CONSTRAINTS_ONLY in
    script import_all.sh to conform to their real use. Thanks to
    Sebastien Caunes for the report.
  - Fix comment detection breaking the package header parsing and
    global variable detection.
  - Fix ROWNUM detection for replacement by LIMIT
  - Fix escaping of psql command in configuration file comment and
    set default value for PG_VERSION to 12.
  - Replace precision by exactness in documentation. Thanks to
    Sebastien Caunes for the report.
  - Prevent reducing DATA_LIMIT when NO_BLOB_EXPORT is enabled.
   Thanks to Thomas Reiss for the report.
  - Fix geometry type detection.
  - Add autodetection of geometry type, srid and dimension for
    ArcGis geometries. Thanks to changmao01 for the feature request.
  - Fix call to ST_GeomFromText when no SRID is found.
  - Fix case where OVERRIDE SYSTEM VALUE clause could be added if PG
    version is < 10. Thanks to changmao01 for the report.
  - Fix unwanted call to internal GEOM library for ArcGis geometries.
    Thanks to changmao01 for the report.
  - Exclude schema SDE (ArGis) from export. Thanks to changmao01 for
    the report.
  - prevent looking twice to same custom data type definition.
  - Fix previous patch to catch SDO_GEOMETRY on lowercase regexp.
  - Limit detection of geometry data type to SDO_GEOMETRY.
  - Fix column name replacement in view definition. Thanks to Amit
    Sanghvi for the report.
  - Fix REPLACE_COLS parsing to allow space in column name. Thanks
    to Amit Sanghvi for the report.
  - Fix translation from file of triggers with WHEN clause. Thanks
    to Rui Pereira for the report.
  - Fix column name kept lowercase in the MOD() clause when -J is
    used. Thanks to Code-UV for the report.
  - Keep case of PG_SCHEMA definition when used in TEST action.
  - Fix data export for columns with custom data type. Thanks to
    Aymen Zaiter for the report.
  - Fix missing bracket with || operator in CREATE INDEX. Thanks to
    traxverlis for the report.
  - Fix export of single row unique function base index. Example:
      CREATE UNIQUE INDEX single_row_idx ON single_row ((1));
    Thanks to unrandom123 for the report.
  - Update documentation about schemas used in TEST action.
  - Disable materialized view export with MySQL export it is not
    supported. Thanks to naveenjul29 for the report.
  - Fix table alias detection in Oracle (+) join rewrite.
  - Fix an infinite loop in Oracle (+) join rewrite when there is no
    table aliases and the table is prefixed by its schema. Thanks to
    Olivier Picavet for the report.
  - Fix MODIFY_STRUCT when column name need to be escaped. Thanks to
    helmichamsi10 for the report.
  - Fix empty PARTITION BY () clause. Thanks to Aymen Zaiter.
  - Fix export of global variable from package description when there
    is no package body. Thanks to naveenjul29 for the report.
  - Add package description export when dumping package source,
    previously only the package body was dump. This will allow to
    check global variables export.
  - Whilst working on the Reproducible Builds effort (https//reproducible-builds.org/)
    it appears that ora2pg could not be built reproducibly. Thanks to
    Chris Lamb for the patch.
  - Fix case of NUMBER(*,10) declaration. Oracle has a precision of 1 to 38
    for numeric. Even if PostgreSQL allow a precision of 1000 use 38 to
    replace junk parameter. Thanks to xinjirufen for the report.
  - Add conversion of default value in function parameters, like syssdate
    rewriting for example. Thanks to unrandom123 for the report.
  - Fix a regression in data encoding when exporting data introduced in
    commit fa8e9de. Thanks to gp4git for the report.
  - Add debug information about the environment variables used before
    connecting to Oracle.
  - Fix case of duplicate between unique index and unique constraint with
    multiple columns. Thanks to gp4git.

2020 10 12 - v21.0

This release fix several issues reported since last release and adds
several new features and improvements.

  * Add clause OVERRIDING SYSTEM VALUE to INSERT statements when the
    table has an IDENTITY column.
  * Considerably increase the speed to generate the report about the
    migration assessment, especially for database with huge number of
    objects.
  * Reduce time passed in the progress bar. Following the number of
    database objects we were spending too much time in refreshing the
    progress bar.
  * Add number of identity columns in migration assessment report.
  * Make assessment details report initially hidden using HTML5 tags
    <details><summary>
  * Improve speed of BLOB/CLOB data export. Oracle recommends reading
    from and writing to a LOB in batches using a multiple of the LOB
    chunk size. This chunk size defaults to 8k (8192). Recent tests
    show that the best performances can be reach with higher value
    like 512K or 4Mb.
  * Add progress bar when --oracle_speed is use in single process mode.
  * Automatically activate USER_GRANTS when the connection user has no DBA
    privilege. A warning is displayed.
  * Complete port to Windows by using the Windows separator on stdout
    redirection into a file at ora2pg command line call and improve
    ora2pg_scanner port on Windows OS.
  * Add rewrite of MySQL JOIN with WHERE clause instead of ON.
  * Add MGDSYS (Oracle E-Business Suite) and APEX_040000 to the list
    of schemas excluded from the export.
  * Supply credentials interactively when a password is not defined in
    the configuration file. Need the installation of a new Perl module
    Term::ReadKey.
  * Add supports oracle connections "as sysdba" with username "/" and
    an empty password to connect to a local oracle instance.
  * Add translation of PRIVATE TEMPORARY TABLE from Oracle 18c into
    PostgreSQL basic temporary table, only the default behavior for
    on commit change.

New command line options:

  * Add new command line option to ora2pg_scanner: -b | --binpath DIR
    to set the full path to directory where the ora2pg binary stays.
    Might be useful only on Windows OS.
  * Add -r | --relative command line option and PSQL_RELATIVE_PATH
    configuration directive. By default Ora2Pg use \i psql command to
    execute generated SQL files if you want to use a relative path
    following the script execution file enabling this option will use
    \ir. See psql help for more information.

New configuration directives:

  * NO_VIEW_ORDERING:
      By default Ora2Pg try to order views to avoid error at import time
      with nested views. With a huge number of views this can take a very
      long time, you can bypass this ordering by enabling this directive.
  * NO_FUNCTION_METADATA
      Force Ora2Pg to not look for function declaration. Note that this
      will prevent Ora2Pg to rewrite function replacement call if needed.
      Do not enable it unless looking forward at function breaks other
      export.
  * LOB_CHUNK_SIZE
      See explanation in the new features and improvement list.
  * ALTERNATIVE_QUOTING_REGEXP
      To support the Alternative Quoting Mechanism ('Q' or 'q') for String
      Literals set the regexp with the text capture to use to extract the
      text part. For example with a variable declared as
            c_sample VARCHAR2(100 CHAR) := q'{This doesn't work.}';
      the regexp to use must be:
            ALTERNATIVE_QUOTING_REGEXP     q'{(.*)}'
      ora2pg will use the $$ delimiter, with the example the result will
      be:
            c_sample varchar(100) := $$This doesn't work.$$;
      The value of this configuration directive can be a list of regexp
      separated by a semi colon. The capture part (between parenthesis) is
      mandatory in each regexp if you want to restore the string constant.

Backward compatibility changes:

  - Default for NO_LOB_LOCATOR is now 1 to benefit from the LOB_CHUNK_SIZE
    performances gain.
  - Enable schema compilation (COMPILE_SCHEMA set to 1) by default to
    speed up DDL extraction.
  - Change the behavior of Ora2Pg with the parameters that follows a
    parameter with a default value. Ora2Pg used to change the order of the
    parameter's function to put all parameters with a default value at end
    of the list which need a function call rewrite. This have been abandoned
    now any parameter without default value after a parameter with a default
    value will be appended DEFAULT NULL.

Here is the full list of changes and acknowledgements:

  - Fix unwanted references to PK/UK when DROP_INDEXES is enabled.
  - Fix comparison between function name in TEST report.
  - Fix duplicates on retrieving partitions information.
  - Improve SHOW_TABLE report about partitioned tables information.
  - Drop code about removing DEFAULT NULL in functions parameters. Thanks to
    chaluvadi286 for the report.
  - Fix two other case where materialized view can be listed in the table list.
  - Fix case where materialized view can be listed in the table list. Thanks
    to Thomas Reiss for the report.
  - Fix %ROWTYPE removing to be restricted to REF CURSOR. Thanks to
    jagmohankaintura-tl for the report.
  - Fix PG functions count when comparing Oracle functions count in TEST action.
    Remove useless -l option to import_all.sh auto generated script.
  - Fix PRESERVE_CASE on schema name for functions extracted from a package.
  - Fix search_path adding public default schema.
  - Apply PRESERVE_CASE to partition by involved columns.
  - Add IF EXIXTS to create schema to avoid error when import_all.sh is run
    several time.
  - Fix sort order of comment on columns for tables and views.
  - Fix warning about data export from nonexistent table resulting of index
    lookup on nested table.
  - Fix infinite loop in global variables package extraction. Thanks to Thomas
    Reiss for the report.
  - Fix global variables and packages export when comments are present in the
    package description.
  - Add information about XML_PRETTY size limit to 4000
  - Fix column name in indexes when PRESERVE_CASE is enabled. Thanks 
    to Julien traxverlis for the report.
  - Fix Top 10 of largest tables sort order. Thanks to Tom Vanzieleghem
    for the patch.
  - Fix duplicates between indexes and constraints. Thanks to sdpdb and
    Jon Betts for the report.
  - Fix SYSDATE replacement and possible infinite loop in SYSDATE parsing.
    Thanks to pbidault for the report.
  - Fix export of Oracle TEXT indexes with USE_UNACCENT disabled. Thanks to
    Eric Delanoe for the report.
  - Add new configuration directive ALTERNATIVE_QUOTING_REGEXP to support
    the Alternative Quoting Mechanism ('Q' or 'q') for String Literals.
    Thanks to just-doit for the report.
  - Fix OF clause missing in update triggers. Thanks to just-doit for
    the report.
  - Fix IS NULL translation in WHERE clause of UPDATE statement. Thanks
    to Eric Delanoe for the report.
  - Remove DDL export of LOG indexes on materialized views.
  - Fix unexpected materialized view listed in table export. Thanks to
    jagmohankaintura-tl for the report.
  - Fix default values with single quote in create table DDL. Thanks to
    justdoit for the report.
  - Fix double quote in CREATE TRIGGER code and applying of preserve case
    on column name.
  - Supply credentials interactively when a password is not defined in
    configuration file. Thanks to rpeiremans for the patch.
  - Add supports oracle connections "as sysdba" with username "/" and
    an empty password to connect to a local oracle instance. Thanks to
    rpeiremans for the patch.
  - Fix documentation about materialized view export.
  - Fix export order of comments on columns.
  - Fix export of views comments when no schema is used for export and
    export schema is activated.
  - Fix cast in replacement with TO_NUMBER and TO_CHAR in indexes. Thanks
    to Kiran for the report.
  - Add MGDSYS (Oracle E-Business Suite) to the list of schemas excluded
    from the export. Thanks to naveenjul29 for the report.
  - Add more information about PG_DSN use. Thanks to Pepan7 for the report.
  - Update copyright year.
  - Fix regression where "SET client_encoding TO ..." was missing data file
    header. Thanks to Emmanuel Gaultier for the report.
  - Fix EDITABLE vs EDITIONABLE parsing. Thanks to Naveen Kumar for the report.
  - Fix typos in documentation. Thanks to swallow-life, ChrisYuan, Edward Betts,
    Jack Caperon and cavpollo for the patches.
  - Add OVERRIDING SYSTEM VALUE to INSERT statement when the table has an
    IDENTITY column. Thanks to Robin Windey for the report
  - Remove empty parenthesis of identity column options
  - Limit sequence/identity column value to bigint max
  - Add an example of DBD::Oracle DSN with 18c.
  - Fix parsing of identity column from file. Thanks to deepakp555 for the
    report.
  - Fix quoting of identifier when PRESERVE_CASE is enable and no particular
    schema is specified. Thanks to mkgrgis for the report.
  - Move setting of search_path before truncate table. Thanks to Michael Vitale
    for the report.
  - Add explanation about TEST and SIZE migration assessment values.
  - Mark XMLTYPE as having LOB locator.
  - Fix XMLTYPE columns that are exported as lob locator. Thanks to Tamas for
    the report.
  - Fix a problem of data export throughput that was slowing down all along
    the export when multiprocess for output was not used. Ora2Pg was forking
    a process for each chunk of data (see DATA_LIMIT) which is useless when
    write output is done on a single process (-j 1) and slow down the export.
    Thanks to markhooper99 and Tamas for reporting, testing and finding the
    source of the issue.
  - Fix progress bar in multiprocess mode, update was not displayed at each
    chunk of data processed.
  - Add internal debug information for progress bar.
  - Add debug information for SHOW_REPORT
  - Fix a long pending issue with custom data type export. Thanks to
    jhollandsworth for the patch.
  - Fix LOB data export with value changed to NULL when the CLOB value was 0.
    Thanks to jhollandsworth for the report.
  - Fix escape format issue with COPY and bytea. Thanks to Christoph Noel and
    dwbrock62 for the report.
  - Add LD_LIBRARY_PATH and PATH prerequisite to run ora2pg.
  - Fix use of the HIGH_VALUE column in partition listing with Oracle 9i. Thanks
    to Francisco Puga for the report.
  - Update the table row count logic to incorporate the PostgreSQL table FQN as
    established through the set_pg_relation_name routine. Thanks to Jacob
    Roberts for the patch.
  - Add the PostgreSQL FQN when printing the results in the TEST function. Thanks
    to Jacob Roberts for the patch.
  - Do not look forward function with the SHOW_* action
  - Fix BLOB export where \x was escaped. Thanks to Christophe Noel for the
    report.
  - Update Ora2Pg.pm to fix symbol in column name in create index statement.
    Thanks to kpoluektov for the patch.
  - Fix package function extraction when there is a start of comment (/*) in
    a constant string. Thanks to Tiago Anastacio for the report.
  - Fix type detection in package declaration. Thanks to Tiago Anastacio for
    the report.
  - Avoid displaying error ORA-22831 when exporting LOB. This error can
    appears when LOB chunk size is different from default 8192. The error
    has no incidence on the export so we can just ignore it. This patch
    also use DBD::Oracle ora_lob_chunk_size() method to gather chunk the
    chunk size of the LOB, fallback to 8192 if not available. Thanks to
    joedbadmin for the report.
  - Disable direct report of Oracle errors, all error should be handled at
    Ora2Pg level.
  - Fix MySQL data export with allow/exclude objects. Thanks to Manuel Pavy for
    the report.
  - Fix exclude/allow object feature in MySQL export that was not working since
    release 19.0. Thanks to Manuel Pavy for the report.
  - Add rewrite of MySQL JOIN with WHERE clause instead of ON. Thanks to Marc
    Rechte for the report.
  - Fix issue with custom type when multiprocess is used.
  - Fix progress bar on final total estimated data in multiprocess mode.
  - Fix ORACLE_HOME path in README.md. Thanks to Lubos Cisar for the patch.
  - Fix missing replacement with PERFORM in CASE ... WHEN statements. Thanks to
    Eric Delanoe for the report.
  - Fix duplicate ora2pg command in iteration.
  - Improve ora2pg_scanner port on Windows OS. Thanks to Marie Contencin for the
    report.
  - Add perl call to all ora2pg commands when the scanner is executed on
    Windows system as the shebang is not recognized. Thanks to Marie Contencin
    for the report.
  - Fix several issue with compressed output. Thanks to Bach Nga for the report.
  - Fix translation of CURSOR IS SELECT with a comment before the SELECT.
    Thanks to Izaak van Niekerk for the report.
  - Fix export of procedures as PostgreSQL procedures with version 11.
  - Add APEX_040000 to the schemas exclusion list.  Thanks to Don Seiler for the
    report.
  - Fix possible unquoted default values. Thanks to Marc Rechte for the report.
  - Fix MySQL SET TRANSACTION clause when TRANSACTION is set to readonly or
    readwrite this is not supported so fall back in READ COMMITTED isolation
    level in this case. Thanks to Marc Rechte for the report.
  - Fix export of functions, column DATA_TYPE does not exists in table
    INFORMATION_SCHEMA.ROUTINES before MySQL 5.5.0. Replace it with column
    DTD_IDENTIFIER for prior version. Thanks to Marc Rechte for the report.
  - Fix double quote in CREATE TRIGGER code and applying of preserve case on
    column name.

2019 01 18 - v20.0

This release fix several issues reported during the last three months
and adds several new features and improvement. The change of major
version is related to backward compatibility break with the removed of
most PG_SUPPORTS_* configuration directives and their replacement with
the new PG_VERSION directive.

New features and configuration directives in this release:

  * Add PG_VERSION configuration directive to set the PostgreSQL major
    version number of the target database. Ex: 9.6 or 10. Default is
    current major version at time of a new release. This replace the
    old PG_SUPPORTS_* configuration directives.
  * Removed all PG_SUPPORTS_* configuration directives minus
    PG_SUPPORTS_SUBSTR that is related to Redshift engine.
  * Export of BFILE as bytea is now done through a PL/SQL function to
    extract the content of a BFILE and generate a bytea data suitable
    for insert or copy into PostgreSQL.
  * Foreign keys that reference a partitioned table are no more
    exported.
  * Show table name on Oracle side during export using at connection
    time: DBMS_APPLICATION_INFO.SET_ACTION(table_name);
  * When the date format is ISO and the value is a constant the call
    to to_date() is removed and only the constant is preserved. For
    example: to_date(' 2013-04-01 00:00:00','SYYYY-MM-DD HH24:MI:SS')
    is replaced by a simple call to: ' 2013-04-01 00:00:00'.
    This rewrite is limited to PARTITION export type when directive
    PG_SUPPORTS_PARTITION is enabled.
  * Add DATA_EXPORT_ORDER configuration directive. By default data
    export order will be done by sorting on table name. If you have
    huge tables at end of alphabetic order and are using multiprocess,
    it can be better to set the sort order on size so that multiple
    small tables can be processed before the largest tables finish.
    In this case set this directive to size. Possible values are name
    and size. Note that export type SHOW_TABLE and SHOW_COLUMN will
    use this sort order too, not only COPY or INSERT export type.
  * Add NO_BLOB_EXPORT configuration directive. Exporting BLOB could
    take time and you may want to export all data except the BLOB
    columns. In this case enable this directive and the BLOB columns
    will not be included into data export. The BLOB column must not
    have a NOT NULL constraint. Thanks to Ilya Vladimirovich for the
  * Add PREFIX_SUB_PARTITION to enable/disable sub-partitioning table
    prefixing in case of the partition names are a part of the sub-
    partition names.
  * Add special replacement for case of epoch syntax in Oracle:
        (sysdate - to_date('01-Jan-1970', 'dd-Mon-yyyy'))*24*60*60
    is replaced by the PostgreSQL equivalent:
        (extract(epoch from now()))

Here is the full list of changes and acknowledgements:

  - Export indexes and constraints on partitioned table with pg >= 11.
  - Fix incorrect replacement of NLS_SORT in indexes.
  - Bring back DISABLE_UNLOGGED feature. Thanks to Jean-Christophe
    Arnu for the patch
  - Fix CREATE SCHEMA statement that was not written to dump file.
  - Fix DBMS_APPLICATION_INFO.set_action() call, old Oracle version
    do not support named parameters.
  - Fix duplicate index name on partition. Thanks to buragaddapavan
    for the report.
  - Add support to new configuration directive PG_VERSION to control
    the behavior of Ora2Pg following PostgreSQL version.
  - Fix error in creation of default partition with PostgreSQL 10.
    Thanks to buragaddapavan for the report.
  - Fix missing export of single MAXVALUE partition, this will produce
    the following range partition: ... FOR VALUES FROM (MINVALUE) TO
    (MAXVALUE) Previous behavior was to not export partition as it is
    better to not partition the table at all. However it is declared
    in Oracle so it is better to export it to see what can be done.
    Thanks to buragaddapavan for the report.
  - Do not export foreign keys that reference a partitioned table.
    Remove NOT VALID on foreign keys defined on a partitioned
    table if present. Thanks to Denis Oleynikov for the report.
  - Fix export of BFILE as bytea. Ora2Pg now use a PL/SQL function to
    extract the content of a BFILE and generate a bytea data suitable
    for insert or copy into PostgreSQL. Thanks to RickyTR for the
    report.
  - Add TIMEZONE_REGION and TIMEZONE_ABBR to migration assessment, no
    direct equivalent in PostgreSQL. Remove NSLSORT not used in
    migration assessment. Thanks to buragaddapavan for the report.
  - Fix output of multiple export type specified in TYPE directive.
  - Rewrite and renaming of _get_sql_data() function into
    _get_sql_statements().
  - Limit CURSOR weight in migration assessment to REF CURSOR only,
    other case are all covered. REF CURSOR might need a review to see
    if they need to be replaced with a SET OF RECORD.
  - Fix replacement of EMPTY_CLOB() or EMPTY_BLOB() with empty string
    when EMPTY_LOB_NULL is disabled and NULL when it is enabled.
  - Prefix output file with the export type in multiple export type
    mode, ex: sequence_output.sql or table_output.sql. Thanks to
    buragaddapavan for the report.
  - Fix export of data from an Oracle nested table. Thanks to rejo
    oommen for the report.
  - Removed cast to timestamp from partition range. Thanks to
    buragaddapavan and rejo-oommen for the report.
  - Fix partition default syntax. Thanks to rejo-oommen for the
    report.
  - Apply missing SYSUSERS schemas exclusion on columns and partition
    listing. Thanks to rejo-oommen for the report.
  - Add warning about parameter order change in output file.
  - Show table name on Oracle side during export using at connection
    time: DBMS_APPLICATION_INFO.SET_ACTION(table_name);
    Thanks to Denis Oleynikov for the feature request.
  - Report change in ORA_RESERVED_WORDS into documentation.
  - Add references in the keyword list of ORA_RESERVED_WORDS.
  - Fix the missing white space in some lines while creating
    import_all.sh file. Thanks to Fabiano for the patch.
  - Fix translation of infinity value for float. Thanks to Damien
    Trecu for the report.
  - Fix default value in timestamp column definition when a timezone
    is given. Thanks to buragaddapavan for the report.
  - Fix missing export of index and constraint in a partitioned
    table when DISABLE_PARTITION is enabled. Thanks to Denis Oleynikov
    for the report.
  - Prevent PARTITION BY when DISABLE_PARTITION is enabled. Thanks to
    Denis Oleynikov for the report.
  - Add DATA_EXPORT_ORDER configuration directive. By default data
    export order will be done by sorting on table name. If you have
    huge tables at end of alphabetic order and are using multiprocess,
    it can be better to set the sort order on size so that multiple
    small tables can be processed before the largest tables finish.
    In this case set this directive to size. Possible values are name
    and size. Note that export type SHOW_TABLE and SHOW_COLUMN will
    use this sort order too, not only COPY or INSERT export type.
    Thanks to Guy Browne for the feature request.
  - Fix remove leading ':' on Oracle variable taking care of regex
    character class. Thanks to jselbach for the report.
  - Add NO_BLOB_EXPORT configuration directive. Exporting BLOB could
    take time and you may want to export all data except the BLOB
    columns. In this case enable this directive and the BLOB columns
    will not be included into data export. The BLOB column must not
    have a NOT NULL constraint. Thanks to Ilya Vladimirovich for the
    feature request.
  - Fix incorrect rewrote of the first custom type in a row. Thanks
    to Francesco Loreti for the patch.
  - Remove double quote in type definition en set type name in lower
    case when PRESERVE_CASE is disabled.
  - Add PREFIX_SUB_PARTITION to enable/disable sub-partitioning table
    prefixing in case of the partition names are a part of the sub-
    partition names.
  - Fix epoch replacement case in CREATE TABLE statements.
  - Apply epoch replacement to default value in table declaration.
  - Add special replacement for case of epoch syntax in Oracle:
        (sysdate - to_date('01-Jan-1970', 'dd-Mon-yyyy'))*24*60*60
    is replaced by the PostgreSQL equivalent:
        (extract(epoch from now()))
    Thanks to rejo-oommen for the feature request.
  - A few typos in --help sections. Thanks to Christophe Courtois
    for the report.
  - Fix export of primary key on partition table. Thanks to chmanu
    for the patch.
  - Fix malformed user defined type export. Thanks to Francesco Loreti
    for the report.


2018 09 27 - v19.1

This release fix several issues reported during the last month and
add support to PostgreSQL 11 HASH partitioning.

It also adds some new features and configuration directives:

  * Add export of default partition and default sub partition. 
  * Add export of HASH partition type.
  * Add support of stored procedure object.
  * Add replacement of NLSORT in indexes or queries. For example:
        CREATE INDEX test_idx ON emp
                          (NLSSORT(emp_name, 'NLS_SORT=GERMAN'));
    is translated into
        CREATE INDEX test_idx ON emp
                    ((emp_name collate "german"));
    The collation still need to be adapted, here probably "de_DE".
    NLSSORT() in ORDER BY clause are also translated.
  * Prevent duplicate index with primary key on partition to be
    exported.
  * PostgreSQL native partitioning does not allow direct import of
    data into already attached partitions. We now force direct import
    into main table but we keep Oracle export of data from individual

This release also adds two new command line options:

   --oracle_speed: use to know at which speed Oracle is able to send
                   data. No data will be processed or written
   --ora2pg_speed: use to know at which speed Ora2Pg is able to send
                   transformed data. Nothing will be written

Use it for debugging purpose. They are useful to see Oracle speed to
send data and at what speed Ora2Pg is processing the data without
reaching disk or direct import into PostgreSQL. 

Two new configuration directive has been added:

  * PG_SUPPORTS_PROCEDURE : PostgreSQL v11 adds support to stored
    procedure objects. Disabled by default.
  - PARALLEL_MIN_ROWS: set the minimum number of tuples in a table
    before calling Oracle's parallel mode during data export. 
    Default to 100000 rows.

Note that PG_SUPPORTS_PARTITION and PG_SUPPORTS_IDENTITY are now
enabled by default to use PostgreSQL declarative partionning and
identity column instead of serial data type.

Here is the full list of changes and acknowledgements:

  - Fix automatic quoting of table or partition name starting with
    a number. Thanks to Barzaqh for the report.
  - Add information about custom directory installation. Thanks to
    joguess for the report.
  - Update list of action in documentation.
  - Fix export of spatial geometries. Thanks to burak yurdakul for
    the report.
  - Fix translation of default value in CREATE TABLE DDL when using
    a function. Thanks to Denis Oleynikov for the report.
  - Prevent moving index on partition during tablespace export.
    Thanks to Maxim Zakharov for the report.
  - Fix upper case of partition name in triggers.
  - Enforce KEEP_PKEY_NAMES when USE_TABLESPACE is enabled. Thanks
    to Maxim Zakharov for the patch.
  - Fix parsing of Oracle user login in dblink input from a file.
  - Fix multiple duplication of range clause in partition export.
  - Add bench of total time and rows to migrate data from Oracle
    in debug mode with speed average.
  - Fix sub partition prefix name.
  - Fix unset oracle username when exporting DBLINK from database.
    Thanks to Denis Oleynikov for the report.
  - Remove NO VALID to foreign keys on partitioned table. Thanks to
    Denis Oleynikov for the report.
  - Fix crash of Ora2Pg on regexp with dynamic pattern base on package
    code. Thank to Alain Debie and MikeCaliffCBORD for the report.
  - PostgreSQL native partitioning does not allow direct import of
    data into already attached partitions. When PG_SUPPORTS_PARTITION
    is enable we now force direct import into main single table but
    we keep Oracle export of data from individual partition. Previous
    behavior was to use main table from both side. Thanks to Denis
    Oleynikov for the report.
  - Add the PARALLEL_MIN_ROWS configuration directive to prevent
    Oracle's parallel mode to be activated during data export if the
    table have less than a certain amount of rows. Default is 100000
    rows. This prevent unnecessary fork of Oracle process. Thanks to
    Denis Oleynikov for the feature request.
  - Fix composite partition MODULUS value. Thanks to Denis Oleynikov
    for the report.
  - Fix count of partitions that was not including subpartition count.
  - Force PostgreSQL user in FDW user mapping to be PG_USER when it is
    defined.
  - Sometimes Oracle indexes can be defined as follow:
        CREATE INDEX idx_err_status_id
                                   ON err_status (status_id, 1);
    which generate errors on PostgreSQL. Remove column names composed
    of digit only from the translation. Thanks to Denis Oleynikov for
    the report.
  - Move Oracle indexes or PK defined on partitioned tables to each
    partition as PostgreSQL do not support UNIQUE, PRIMARY KEY,
    EXCLUDE, or FOREIGN KEY constraints on partitioned tables.
    Definition are created in file PARTITION_INDEXES_output.sql
    generated with the PARTITION export type. Thanks to Denis
    Oleynikov for the feature request.
  - Fix parallel data load from Oracle partitioned tables by using
    a unique alias. Thanks to Denis Oleynikov for the report.
  - Fix export of composite partitioned (range/hash) table when
    PG_SUPPORTS_PARTITION is disabled. Thanks to Denis Oleynikov
    for the report.
  - Remove composite sub partition from the list of partition, this
    return a wrong partition count.
  - Fix MODULUS value in hash sub partitioning.
  - Index and table partitions could be on separate tablespaces.
    Thanks to Maxim Zakharov for the patch.
  - Fix case where procedure object name is wrongly double quoted.
    Thanks to danghb for the report.
  - Fix parser to support comment between procedure|function name
    and IS|AS keyword. Thanks to danghb for the report.
  - Remove dependency to List::Util for the min() function.


2018 08 18 - v19.0

This major release fix several issues reported by users during last
year. It also adds several new features and configuration directives.

New features:

  - Add export of Oracle HASH partitioning when PG_SUPPORTS_PARTITION
    is enabled. This is a PostgreSQL 11 feature.
  - Add SUBTYPE translation into DOMAIN with TYPE and PACKAGE export.
  - Add automatic translation of
      KEEP (DENSE_RANK FIRST|LAST ORDER BY ...) OVER (PARTITION BY ...)
    into
      FIRST|LAST_VALUE(...) OVER (PARTITION BY ... ORDER BY ...).
   - Add PCTFREE to FILLFACTOR conversion when PCTFREE is upper than
     the default value: 10. 
  - Replace DELETE clause not followed with FROM (optional in Oracle).
  - Remove Oracle extra clauses in TRUNCATE command.
  - Allow use of NUMBER(*) in DATA_TYPE directive to convert all
    NUMBER(*) into the given type whatever is the length. Ex:
        DATA_TYPE    NUMBER(*):bigint.
  - Add a PARALLEL hint to all Oracle queries used to migrate data.
  - Add export of Identity Columns from Oracle Database 12c.
  - Add translation of UROWID datatype and information in documentation
    about why default corresponding type OID will fail at data import.
  - Remove unwanted and unused keywords from CREATE TABLE statements:
    PARALLEL and COMPRESS.
  - Remove TEMPORARY in DROP statements.
  - Improve speed of escape_copy() function used for data export.
  - Add translation of Oracle functions NUMTOYMINTERVAL() and
    NUMTODSINTERVAL().
  - Add counting of jobs defined in Oracle scheduler in the migration
    assessment feature. 
  - Add CSMIG in the list of Oracle default system schema
  - Fully rewrite data export for table with nested user defined types
    DBD::Oracle fetchall_arrayref() is not able to associate complex
    custom types to the returned arrays, changed this call to use
    fetchrow_array() also used to export BLOB.
  - QUERY export will now output translated queries as well as
    untranslated ones. This break backward compatibility, previously
    only translated query was dumped.
  - Auto detect UTF-8 input files to automatically use utf8 encoding.
  - Support translation of MySQL global variables.
  - Add translation of preprocessor in Oracle external table into
    program in foreign table definition. Allow translation of external
    table from file.
  - Add translation to NVL2() Oracle function.
  - Translate CONVERT() MySQL function.
  - Translate some form of GROUP_CONCAT() that was not translated.
  - Remove call to CHARSET in cast() function, replace it by COLLATE
    every where else. This must cover most of the cases but some
    specials use might not, so please reports any issue with this
    behavior.
  - Add -c | --config command line option to ora2pg_scanner to set
    custom configuration file to be used instead of ora2pg default:
    /etc/ora2pg/ora2pg.conf
  - Improve CONNECT BY and OUTER JOIN translation.
  - And lot of MySQL to PostgreSQL improvements.

Several new configuration directives have been added:

  - Add DEFAULT_PARALLELISM_DEGREE to control PARALLEL hint use
    when exporting data from Oracle. Default is disabled.
  - Make documentation about KEEP_PKEY_NAMES more explicit about
    kind of constraints affected by this directive.
  - Add PG_SUPPORTS_IDENTITY configuration directive to enable
    export of Oracle identity columns into PostgreSQL 10 feature.
    If PG_SUPPORTS_IDENTITY is disabled and there is IDENTITY column
    in the Oracle table, they are exported as serial or bigserial
    columns. When it is enabled they are exported as IDENTITY columns
    like:
    
          CREATE TABLE identity_test_tab (
                  id bigint GENERATED ALWAYS AS IDENTITY,
                  description varchar(30)
          ) ;
    
    If there is non default sequence option set in Oracle, they will
    be appended after the IDENTITY keyword. Additionally in both cases
    Ora2Pg will create a file AUTOINCREMENT_output.sql with a function
    to update the associated sequences with the restart value set to 
    "SELECT max(colname)+1 FROM tablename". Of course this file must
    be imported after data import otherwise sequence will be kept to
    start value.
  - Add DISABLE_UNLOGGED configuration directive. By default Ora2Pg
    export Oracle tables with the NOLOGGING attribute into UNLOGGED
    tables. You may want to fully disable this feature because you
    will lost all data from unlogged table in case of PostgreSQL crash.
    Set it to 1 to export all tables as normal table. When creating a
    new migration project using --init_project, this directive is
    activated by default. This is not the case in the default
    configuration file for backward compatibility.
  - Add FORCE_SECURITY_INVOKER configuration directive. Ora2Pg use
    the function's security privileges set in Oracle and it is often
    defined as SECURITY DEFINER. To override those security privileges
    for all functions and use SECURITY DEFINER instead, enable this
    directive.
  - Add AUTONOMOUS_TRANSACTION in configuration to enable translation
    of autonomous transactions into a wrapper function using dblink
    or pg_background extension. If you don't want to use this feature
    and just want to export the function as a normal one without the
    pragma call, disable this directive.
  - Add documentation about COMMENT_SAVEPOINT configuration directive.
  - Major rewrite in PACKAGE parser to better support global variables
    detection. Global variable that have no default values are now
    always initialized to empty string in file global_variables.conf
    so that we see that they exists. This might not change the global
    behavior.

I especially want to thank Pavel Stehule and Eric Delanoe who spent
lot of time this year to help me to improve the PL/SQL to plpgsql
translation and also Krasiyan Andreev who help a lot to finalize
the MySQL to PostgreSQL migration features.

Here is a complete list of changes and acknowledgments:

  - Fix translation of "varname cursor%ROWTYPE;". Thanks to Philippe
    Beaudoin for the report.
  - Fix return of autonomous transaction dblink call when function has
    OUT parameter. Thanks to Pavel Stehule for the report.
  - Add Oracle to PostgreSQL translation of windows functions 
    KEEP (DENSE_RANK FIRST|LAST ORDER BY ...) OVER (PARTITION BY ...)
    Thanks to Swapnil bhoot929 for the feature request.
  - Fix "ORA-03113: end-of-file on communication channel" that what
    generated by a too long query send to Oracle. The size of queries
    sent to Oracle to retrieve object information depend of the ALLOW
    and EXCLUDE directives. If you have lot of objects to filter you
    can experience this kind of non explicit error. Now Ora2pg use
    bind parameter to pass the filters values to reduce the size of
    the prepared query. Thanks to Stephane Tachoire for the report.
  - Add SUBTYPE translation into DOMAIN with TYPE and PACKAGE export.
    Thanks to Francesco Loreti for the feature request.
  - Fix PLS_INTEGER replacement.
  - Remove precision for RAW|BLOB as type modifier is not allowed for
    type "bytea".
  - Fix call of schema.pckg.function() in indexes with a replacement
    with pckg.function(). Thanks to w0pr for the report.
  - Fix translation of UPDATE trigger based on columns:
        "BEFORE UPDATE OF col1,col2 ON table".
    Thanks to Eric Delanoe for the report.
  - Remove single / from input file that was causing a double END in
    some case. Thanks to Philippe Beaudoin for the report.
  - Limit translation of PCTFREE into FILLFACTOR when PCTFREE is upper
    than the Oracle default value: 10. With PostgreSQL 100 (complete
    packing) is the default.
  - Add PCTFREE to FILLFACTOR conversion. Thanks to Maxim Zakharov
    for the patch.
  - Remove TRUNCATE extra clauses. Thanks to e7e6 for the patch.
  - Fix type conversion when extra \n added after ;. Thanks to
    Maxim Zakharov for the patch.
  - Fix DELETE clause not followed with FROM (optional in Oracle).
    Thanks to Philippe Beaudoin for the patch.
  - Limit call to ALL_TAB_IDENTITY_COLS to version 12+. Thanks to
    Andy Garfield for the report.
  - Fix comment parsing. Thanks to Philippe Beaudoin for the report.
  - Allow use of NUMBER(*) in DATA_TYPE directive to convert all
    NUMBER(*) into the given type whatever is the length.
    Thanks to lingeshpes for the feature request.
  - Fix bug in function-based index export. Thanks to apol1234 for
    the report.
  - Add PARALLEL hint to all data export queries. Thanks to jacks33
    for the report.
  - Make documentation about KEEP_PKEY_NAMES more explicit about kind
    of constraints affected by this directive.
  - Fix export of identity columns by enclosing options between
    parenthesis and replacing CACHE 0 by CACHE 1. Thanks to swmcguffin
    devtech for the report.
  - Add parsing of identity columns from file.
  - Fix unwanted replacement of IF () in MySQL code. Thanks to
    Krasiyan Andreev for the report.
  - Fix to_char() translation, thanks to Eric Delanoe for the report.
  - Fix untranslated PERFORM into exception. Thanks to Pavel Stehule
    for the report.
  - Add _get_entities() function to MySQL export. It returns nothing,
    AUTO_INCREMENT column are translated with corresponding types,
    smallserial/serial/bigserial.
  - Fix look at encrypted column on Oracle prior to 10. Thanks to
    Stephane Tachoires for the patch.
  - Add export of Identity Columns from Oracle Database 12c. Thanks
    to swmcguffin-devtech for the feature request.
  - Prevent Ora2Pg to scan ALL_SCHEDULER_JOBS for version prior to 10
    Thanks to Stephane Tachoires for the patch.
  - Fix pull request #648 to log date only when debug is enabled and
    use POSIX strftime instead of custom gettime function.
  - Add system time to debug log info. Thanks to danghb for the patch.
  - Fix parsing of trigger from file and exception.
  - Fix very slow export of mysql tablespace when number of table is
    large. Thanks to yafeishi for the report.
  - Fix translation of CAST( AS unsigned). Thanks to Krasiyan Andreev.
  - Fix MySQL character length to use character_maximum_length
    instead of equal character_octet_length. Thanks to yafeishi for
    the report.
  - Fix custom replacement of MySQL data type. Thanks to Krasiyan
    Andreev for the report.
  - Fix replacement of call to open cursor with empty parenthesis.
    Thanks to Philippe Beaudoin for the report.
  - Fix MySQL data type conversion in function declaration. Thanks to
    Krasiyan Andreev for the report.
  - Fix error with -INFINITY as default value for date or timestamp
    columns.
  - Fix procedure call rewrite with unwanted comma on begin of
    parameter list. Thanks to Pavel Stehule for the report.
  - Fix handling of foreign keys when exporting data and DROP_FKEYS
    is enabled and ALLOW/EXCLUDE directive is set. Now Ora2Pg will
    first drop all foreign keys of a table in the export list and all
    foreign keys of other tables pointing to the table. After data
    import, it will recreate all of these foreign keys. Thanks to
    Eric Delanoe for the report.
  - Fix broken transformation of procedure call with default parameter
    Thanks to Pavel Stehule for the report.
  - Translate call to TIMESTAMP in partition range values into a cast.
    Thanks to markiech for the report.
  - Fix CONNECT BY translation when the query contain an UNION. Thanks
    to mohammed-a-wadod for the report.
  - Fix CONNECT BY with PRIOR on the right side of the predicat.
  - Fix outer join translation when the (+) was in a function, ex:
        WHERE UPPER(trim(VW.FRIDAY))= UPPER(trim(FRIDAY.NAME(+))).
  - Order outer join pending tables in from clause.
  - Order by object name comments and indexes export.
  - Fix outer join translation when the table is not in the from
    clause. Thanks to Cyrille Lintz for the report.
  - Try to fix potential Oracle schema prefixing PostgreSQL schema
    name in CREATE SCHEMA. Thanks to Cyrille Lintz for the report.
  - Fix error in TRIM() translation. Thanks to Cyrille Lintz for the
    report.
  - Add translation of UROWID datatype and information in documentation
    about why default corresponding type OID will fail at data import.
    Thanks to Cyrille Lintz for the report.
  - Fix bug in exporting boolean default values in column definition.
  - Fix bug in column parsing in CREATE TABLE.
  - Adapt default value for data type changed to boolean.
  - Fix bad handling of -D (data_type) option.
  - Change behavior in the attempt to set MySQL global variable type.
    Now variable type will be timestamp if the variable name contains
    datetime, time if the name contains only time and date for date.
    Thanks to Krasiyan Andreev for the report.
  - Fix function replacement in MySQL declare section. Thanks to
    Krasiyan Andreev fr the report.
  - Apply REPLACE_ZERO_DATE to default value in table declaration.
    Thanks to Krasiyan Andreev for the report.
  - Add support to embedded comment in table DDL.
  - Fix replacement of data type for MySQL code. Thanks to Krasiyan
    Andreev for the report.
  - Fix MySQL type replacement in function. Thanks to Krasiyan Andreev
    for the report.
  - Improve speed of escape_copy() function used for data export.
    Thanks to pgnickb for the profiling.
  - Add translation of Oracle functions NUMTOYMINTERVAL() and
    NUMTODSINTERVAL(). Thanks to Pavel Stehule for the report.
  - Counting jobs defined in Oracle scheduler. Thanks to slfbovey
    for the patch.
  - Fix several issue in create table DDL parser:
      - remove double quote of object name when a list of column is
        entered
      - split of table definition to extract column and constraint
        parts is now more efficient
      - remove dot in auto generated constraint name when a schema
        is given in table name
      - fix default values with space that was breaking the parser
  - Remove use of bignum perl module that reports error on some
    installation. Thanks to Cyrille Lintz for the report.
  - Fix a typo preventing perldoc to complete. Thanks to slfbovey
    for the patch.
  - Fully rewrite data export for table with nested user defined types
    DBD::Oracle fetchall_arrayref() is not able to associate complex
    custom types to the returned arrays, changed this call to use
    fetchrow_array() also used to export BLOB. Thanks to lupynos for
    the report.
  - Fix renaming of temporary files during partitions data export.
  - Fix Oracle use of empty string as default value for integers.
    Oracle allow such declaration: SOP NUMBER(5) DEFAULT '' which
    PostgreSQL does not support. Ora2Pg now detect this syntax and
    replace empty string with NULL. Thanks to ricdba for the report.
  - Add detection of Oracle version before setting datetime format,
    needed for Oracle 8i compatibility.
  - Export of tables from Oracle database are now ordered by name by
    default. Thanks to Markus Roth for the report.
  - Fix an other case of missing translation of UNSIGNED into bigint.
    Thanks to Krasiyan Andreev for the report.
  - Force replacement of double quote into single quote for MySQL view
    and function code.
  - Fix case when SET @varname := ... is used multiple time in the
    same function. Thanks to Krasiyan Andreev for the report.
  - Fix case where SET @varname := ... was not translated. Thanks to
    Krasiyan Andreev for the report.
  - Adjust the regex pattern of last patch.
  - Fix unwanted newline after hint replacement that could break
    comments. Thanks to Pavel Stehule for the report.
  - Fix if() replacement in query. Thanks to Krasiyan Andreev for the
    report.
  - Remove extra parenthesis in some form of JOIN. Thanks to Krasiyan
    Andreev for the report.
  - Fix untranslated call to UNSIGNED, now translated as bigint.
  - Thanks to Krasiyan Andreev for the report.
  - Fix translation of double(p,s) into decimal(p,s).
  - Remove use of SET when an assignment is done through a SELECT
    statement. Thanks to Krasiyan Andreev for the report.
  - Fix non-quoted reserved keywords in INSERT / COPY statements when
    exporting data. Thanks to Pavel Stehule for the report.
  - Fix partition data export to file, temporary files for partition
    output was not renamed at export end then data was not loaded.
  - Fix double operator := during function with out param rewrite.
  - Fix commit f1166e5 to apply changes when FILE_PER_FUNCTION is
    disable or when an input file is given.
  - Fix translation of LOCATE(). Thanks to Krasiyan Andreev for the
    report.
  - Fix case where MySQL GROUP_CONCAT() function was not translated.
    Thanks to Krasiyan Andreev for the report.
  - Fix :new and :old translation in triggers.
  - Fully rewrite function call qualification process, the second pass
    now is only use to requalify call to pkg.fct into pkg_ftc when
    PACKAGE_AS_SCHEMA is disable. The replacement of all function
    calls using double quote when a non supported character is used or
    when PRESERVE_CASE is enabled has been completely removed as this
    takes too much time to process for just very few case. So by
    default now Ora2Pg will not go through the second pass. This can
    change in the future especially if this is more performant to
    process PERFORM replacement. Thanks a lot to Eric Delanoe for his
    help on this part.
  - Exclude function and procedure not from package to be used in
    requalify call. Thanks to Eric Delanoe for the report.
  - Fix function name qualification in multiprocess mode.
  - Fix unqualified function call due to unclose file handle.
  - Prevent try to requalify function call if the function is
    not found in the file content.
  - Remove ALGORITHM=.*, DEFINER=.* and SQL SECURITY DEFINER from
    MySQL DDL code.
  - An other missing change to previous commit on qualifying function
    call.
  - Limit function requalification to export type: VIEW, TRIGGER,
    QUERY, FUNCTION, PROCEDURE and PACKAGE.
  - Auto detect UTF-8 input files to automatically use utf8 encoding.
  - Remove all SHOW ERRORS and other call to SHOW in Oracle package
    source as they was badly interpreted as global variable.
  - Fix MySQL CREATE TABLE ... SELECT statement.
  - Fix pending translation issue on some DATE_FORMAT() case.
    Thanks to Krasiyan Andreev for the report.
  - Fix translation of IN (..) in MySQL view. Thanks to Krasiyan
    Andreev for the report.
  - Fix MySQL date format with digit.
  - Fix DATE_FORMAT, WHILE and IFNULL translation issues.
  - Fix not translated MySQL IF() function.
  - Fix other MySQL translation issues for @variable. Thanks to
    Krasiyan Andreev for the report.
  - Fix issue in MySQL IF translation with IN clause. Thanks to
    Krasiyan Andreev for the report.
  - Clarify comment about XML_PRETTY directive. Thanks to TWAC
    for the report.
  - Fix remaining MySQL translation issues for @variable reported
    in issue #590.
  - Fix no translated := in SET statement.
  - Fix output order of translated function.
  - Fix non printable character or special characters that make
    file encoding to ISO-8859 instead of utf8. Thanks to twac for
    the report.
  - Prevent MySQL global variable to be declared twice. Thanks to
    Krasiyan Andreev for the report.
  - Support translation of MySQL global variables. Session variable
    @@varname are translated to PostgreSQL GUC variable and global
    variable @varname are translated to local variable defined in a
    DECLARE section. Ora2Pg tries to gather the data type by using
    integer by default, varchar if there is a constant string ('...')
    in the value and a timestamp if the variable name have the keyword
    date or time inside. Thanks to Krasiyan Andreev for the feature
    request.
  - Fix DATE_ADD() translation.
  - Add translation of preprocessor in Oracle external table into
    program in foreign table definition. Thanks to Thomas Reiss for
    the report. Allow translation of external table from file.
  - Fix case where IF EXISTS might not be append when it is not
    supported by PG.
  - Translate CONVERT() MySQL function. Thanks to Krasiyan Andreev
    for the report.
  - Translate some form of GROUP_CONCAT() that was not translated.
    Thanks to Krasiyan Andreev for the report.
  - Apply same principe with COMMIT in MySQL function code than in
    Oracle code. It is kept untouched to be able to detect a possible
    change of code logic. It can be automatically commented if
    COMMENT_COMMIT_ROLLBACK is enabled. Also I have kept the START
    TRANSACTION call but it is automatically commented.
  - Add mysql_enable_utf8 => 1 to MySQL connection to avoid issues
    with encoding. Thanks to Krasiyan Andreev for the report.
  - Prevent removing of comment on MySQL function and add a "COMMENT
    ON FUNCTION" statement at end of the function declaration. Thanks
    to Krasiyan Andreev for the report.
  - Fix translation of types in MySQL function parameter. Thanks to
    Krasiyan Andreev for the report.
  - Remove START TRANSACTION from MySQL function code. Thanks to
    Krasiyan Andreev for the report.
  - Fix previous patch, we do not need to look forward for function
    or procedure definition in VIEW export and there is no package
    with MySQL. Thanks to Krasiyan Andreev for the report.
  - Fix call to useless function for MySQL function.
  - Add rewrite of MySQL function call in function or procedure code
    translation and some other translation related to MySQL code.
  - Fix ora2pg_scanner when exporting schema with $ in its name.
    Thanks to Aurelien Robin for the report.
  - Disable number of microsecond digit for Oracle version 9. Thanks
    to Aurelien Robin for the report.
  - Do not look at encrypted column for DB version < 10. Thanks to
    Aurelien Robin for the report.
  - Fix MySQL call to charset in cast function. MySQL charset "utf8"
    is also set to COLLATE "C.UTF-8". Thanks to Krasiyan Andreev for
    the report.
  - Fix two bug in CONNECT BY and OUTER JOIN translation.
  - Forgot to handle exception to standard call to IF in MySQL IF()
    translation. Thanks to Krasiyan Andreev for the report.
  - Forgot to apply previous changes to procedure.
  - Fix IF() MySQL replacement when nested and when containing an
    IN (...) clause. Thanks to  Krasiyan Andreev for the report.
  - Fix double BEGIN on MySQL function export. Thanks to Krasiyan
    Andreev for the report.
  - Fix enum check constraint name when PRESERVE_CASE is enabled.
  - Fix case where object with LINESTRING and CIRCULARSTRING was
    exported as MULTILINESTRING instead of MULTICURVE.
  - Fix export of MULTICURVE with COMPOUNDCURVE. Thanks to Petr Silhak
    for the report.
  - Fix several issue in MySQL table DDL export. Thanks to Krasiyan
    Andreev for the report.
  - Fix MySQL auto_increment data type translation and columns export
    order.
  - Fix translation of MySQL function CURRENT_TIMESTAMP(). Thanks to
    Krasiyan Andreev for the report.
  - Fix export of MySQL alter sequence name when exporting auto
    increment column. Thanks to Krasiyan Andreev for the report.
  - Replace IF() call with CASE ... END in VIEW and QUERY export for
    MySQL. Thanks to Krasiyan Andreev for the feature request.
  - Replace backquote with double quote on mysql statements when read
    from file.
  - Fix bug in REGEXP_SUBSTR replacement.
  - Prevent replacement with same function name from an other package.
    Thanks to Eric Delanoe for the report.
  - Apply same STRICT rule for SELECT INTO to EXECUTE INTO. Thanks to
    Pavel Stehule for the report.
  - Fix extra parenthesis removing when a OR clause is present. Thanks
    to Pavel Stehule for the report.
  - Keep autonomous pragma commented when conversion is deactivated
    to be able to identify functions using this pragma.
  - Fix bug in replacement of package function in string constant.
  - Fix malformed replacement of array element calls. Thanks to Eric
    Delanoe for the report.
  - Fix unwanted replacement of TO_NUMBER function. Thanks to Torquem
    for the report.
  - Add an example of DSN for MySQL in ORACLE_DSN documentation.
    Thanks to François Honore for the report.
  - Fix typo in default dblink connection string. Thanks to Pavel
    Stehule for the report.
  - Add information about Oracle Instant Client installation. Thanks
    to Jan Birk for the report.
  - Replace Oracle array syntax arr(i).x into arr[i].x into PL/SQL
    code. Thanks to Eric Delanoe for the report.
  - Use a more generic connection string for DBLINK. It will use
    unix socket by default to connect and the password must be set
    in .pgpass. This will result in the following connection string:
        format('port=%s dbname=%s user=%', current_setting('port'),
		current_database(), current_user)
    If you want to redefine this connection string use DBLINK_CONN
    configuration directive. Thanks to Pavel Stehule for the feature
    request.
  - Fix missing RETURN NEW in some trigger translation. Thanks to
    Pavel Stehule for the report.
  - Fix a missing but non mandatory semi-comma.
  - Keep PKs/unique constraints which are deferrable in Oracle also
    deferrable in PostgreSQL. Thank to Sverre Boschman for the patch.
  - Fix parsing and translation of CONNECT BY. Thanks to bhoot929
    for the report.
  - Fix FDW export when exporting all schema. Thanks to Laurenz Albe
    for the report.
  - Add a note about multiple value in export type that can not
    include COPY or INSERT together with others export type.
  - Fix duplicate condition. Thanks to Eric Delanoe for the report.
  - Fix unwanted translation into PERFORM after INTERSECT.
  - Comment savepoint in code. Thanks to Pavel Stehule for the patch.
  - Fix "ROLLBACK TO" that was not commented. Thanks to Pavel Stehule
    for the report.
  - Fix restore of constant string when additional string constant
    regex are defined in configuration file.
  - Fix translation of nextval with sequence name prefixed with their
    schema.
  - Cast call to TO_DATE(LOCALTIMESTAMP,...) translated into
    TO_DATE(LOCALTIMESTAMP::text,...). Thanks to Keshav kumbham
    for the report.
  - Remove double quote added automatically by Oracle on view
    definition when PRESERVE_CASE is not enable. Thanks to JeeIPI for
    the report.
  - Fix translation of FROM_TZ with a call to function as first 
    parameter. Thanks to TrungPhan for the report.
  - Fix package export when FILE_PER_FUNCTION is set. Thanks to
    Julien Rouhaud for the report.
  - Add translation of REGEXP_SUBSTR() with the following rules:
    Translation of REGEX_SUBSTR( string, pattern, [pos], [nth])
    converted into
        SELECT array_to_string(a, '')
          FROM regexp_matches(substr(string, pos), pattern, 'g')
               AS foo(a)
          LIMIT 1 OFFSET (nth - 1);
    Optional fifth parameter of match_parameter is appended to 'g'
    when present.  Thanks to bhoot929 for the feature request.
  - Add count of REGEX_SUBSTR to migration assessment cost.
  - Add translation support of FROM_TZ() Oracle function. Thanks
    to trPhan for the feature request.
  - Forces ora2pg to output a message when a custom exception code
    has less than 5 digit.
  - Fix errcode when Oracle custom exception number have less than
    five digit. Thanks to Pavel Stehule for the report.
  - Fix case where custom errcode are not converted. Thanks to Pavel
    Stehule for the report.
  - Fix print of single semicolon with empty line in index export.
  - Fix problem with TO_TIMESTAMP_TZ conversion. Thanks to Keshav-
    kumbham for the report.
  - Fix unwanted double quote in index column with DESC sorting.
    Thanks to JeeIPI for the report.
  - Fix non detection case of tables in from clause for outer join
    translation. Thanks to Keshav for the report.
  - Fix unwanted replacement of = NULL into IS NULL in update
    statement. Thanks to Pavel Stehule for the report.
  - Force schema name used in TEST action to lowercase. Thanks to
    venkatabn for the report.
  - Fix export of spatial geometries with CURVEPOLYGON + COMPOUNDCURVE
    Thanks to kabog for the report.

2017 09 01 - v18.2

This release fix several issues reported during the last six months.
It also adds several new features and configuration directives:

  - Add translation of SUBSTRB into substr.
  - Allow use of array in MODIFY_TYPE to export Oracle user defined
    type that are just array of some data type. For example:
      CREATE OR REPLACE TYPE mem_type IS VARRAY(10) of VARCHAR2(15);
    can be directly translated into text[] or varchar[]. In this case
    use the directive as follow: MODIFY_TYPE	CLUB:MEMBERS:text[]
    Ora2Pg will take care to transform all data of this column into
    the correct format. Only arrays of characters and numerics types
    are supported.
  - Add translation of Oracle function LISTAGG() into string_agg().
  - Add TEST_VIEW action to perform a simple count of rows returned by
    views on both database.
  - Translate SQL%ROWCOUNT into GET DIAGNOSTICS rowcount = ROW_COUNT
    and add translation of SQL%FOUND.
  - Add translation of column in trigger event test with IS DISTINCT,
    for example: IF updating('ID') THEN ... will be translated into:
    IF TG_OP = 'UPDATE' AND NEW.'ID' IS DISTINCT FROM OLD.'ID' then...
  - Replace UTL_MATH.EDIT_DISTANCE function by fuzzymatch levenshtein.
  - Allow use of MODIFY_STRUCT with TABLE export. Table creation DDL
    will respect the new list of columns and all indexes or foreign
    key pointing to or from a column removed will not be exported.
  - Add export of partition and subpartition using PostgreSQL native
    partitioning.
  - Auto detect encrypted columns and report them into the assessment.
    SHOW_COLUMN will also mark columns as encrypted.
  - Add information to global temporary tables in migration assessment.
  - Add experimental DATADIFF functionality.
  - Allow use of multiprocess with -j option or JOBS to FUNCTION and
    PROCEDURE export. Useful if you have thousands of these objects.
  - Force RAW(N) type with default value set to sys_guid() as UUID
    on PostgreSQL.
  - Replace function with out parameter using select into. For example
    a call to:       get_item_attr( attr_name, p_value );
    where p_value is an INOUT parameter, will be rewritten as
    
    	p_value := get_item_attr( attr_name, p_value );
    
    If there is multiple OUT parameters, Ora2Pg will use syntax:
    
    	SELECT get_item_attr( attr_name, p_value )
		INTO (attr_name, p_value);

  - Add translation of CONNECT BY using PostgreSQL CTE equivalent.
    This translation also include a replacement of LEVEL and
    SYS_CONNECT_BY_PATH native Oracle features. On complex queries
    there could still be manual editing but all the main work is done.
  - Add support to user defined exception, errcode affected to each
    custom exception start from 50001.
  - Translate call to to_char() with a single parameter into a cast
    to varchar. Can be disabled using USE_ORAFCE directive.
  - Improve ora2pg_scanner to automatically generates migration
    assessment reports for all schema on an Oracle instance. Before
    the schema name to audit was mandatory, now, when the schema
    is not set Ora2Pg will scan all schema. The connexion user need
    to have DBA privilege. Ora2Pg will also add the hostname and SID
    as prefix in the filename of the report. This last change forbids
    ora2pg_scanner to overwrite a report if the same schema name is
    found in several databases.

Several new configuration directives have been added:

  - Add USE_ORAFCE configuration directive that can be enabled if you
    want to use functions defined in the Orafce library and prevent
    Ora2Pg to translate call to these functions. The Orafce library
    can be found here: https://github.com/orafce/orafce
    By default Ora2pg rewrite add_month(), add_year(), date_trunc()
    and to_char() functions, but you may prefer to use the Orafce
    functions that do not need any code transformation. Directive
    DATE_FUNCTION_REWRITE has been removed as it was also used to
    disable replacement of add_month(), add_year() and date_trunc()
    when Orafce is used, useless now.
  - Add FILE_PER_FKEYS configuration directive to allow foreign key
    declaration to be saved in a separate file during schema export.
    By default foreign keys are exported into the main output file or
    in the CONSTRAINT_output.sql file. If enabled foreign keys will be
    exported into a file named FKEYS_output.sql
  - Add new COMMENT_COMMIT_ROLLBACK configuration directive. Call to
    COMMIT/ROLLBACK in PL/SQL code are kept untouched by Ora2Pg to
    force the user to review the logic of the function. Once it is
    fixed in Oracle source code or you want to comment this calls
    enable the directive.
  - Add CREATE_OR_REPLACE configuration directive. By default Ora2Pg
    use CREATE OR REPLACE in function DDL, if you need not to override
    existing functions disable this configuration directive, DDL will
    not include OR REPLACE.
  - Add FUNCTION_CHECK configuration directive. Disable this directive
    if you want to disable check_function_bodies.
    
    	SET check_function_bodies = false;
    
    It disables validation of the function body string during CREATE
    FUNCTION. Default is to use de postgresql.conf setting that enable
    it by default.
  - Add PG_SUPPORTS_PARTITION directive, disabled by default.
    PostgreSQL version prior to 10.0 do not have native partitioning.
    Enable this directive if you want to use PostgreSQL declarative
    partitioning instead of the old style check constraint and trigger.
  - Add PG_SUPPORTS_SUBSTR configuration directive to replace substr()
    call with substring() on old PostgreSQL versions or some fork
    like Redshift.
  - Add PG_INITIAL_COMMAND to send some statements at session startup.
    This directive is the equivalent used for Oracle connection,
    ORA_INITIAL_COMMAND. Both can now be used multiple time now.
  - Add DBLINK_CONN configuration directive. By default if you have
    an autonomous transaction translated using dblink extension the
    connection is defined using the values set with PG_DSN, PG_USER
    and PG_PWD. If you want to fully override the connection string
    use this directive to set the connection in the autonomous
    transaction wrapper function. For example:
    
        DBLINK_CONN    port=5432 dbname=pgdb host=localhost user=pguser password=pgpass
    
  - Add STRING_CONSTANT_REGEXP configuration directive. Ora2Pg replace
    all string constant during the pl/sql to plpgsql translation,
    string constant are all text include between single quote. If you
    have some string placeholder used in dynamic call to queries you
    can set a list of regexp to be temporary replaced to not break the
    parser. For example:
    
    	STRING_CONSTANT_REGEXP         <cfqueryparam value=".*">
    
    The list of regexp must use the semi colon as separator.
  - Add FUNCTION_STABLE configuration directive. By default Oracle
    functions are marked as STABLE as they can not modify data unless
    when used in PL/SQL with variable assignment or as conditional
    expression. You can force Ora2Pg to create these function as
    VOLATILE by disabling this configuration directive.
  - Add new TO_NUMBER_CONVERSION configuration directive to control
    TO_NUMBER translation behavior. By default Oracle call to function
    TO_NUMBER will be translated as a cast into numeric. For example,
    TO_NUMBER('10.1234') is converted into PostgreSQL call:
        to_number('10.1234')::numeric.
    If you want you can cast the call to integer or bigint by changing
    the value of the configuration directive. If you need better
    control of the format, just set it as value, for example:
	TO_NUMBER_CONVERSION    99999999999999999999D9999999999
    will convert the code above as:
          TO_NUMBER('10.1234', '99999999999999999999D9999999999')
    Any value of the directive that it is not numeric, integer or
    bigint will be taken as a mask format. If set to none, then no
    conversion will be done.
  - Add LOOK_FORWARD_FUNCTION configuration directive which takes a
    list of schema to get functions/procedures meta information that
    are used in the current schema export. When replacing call to
    function with OUT or INOUT parameters, if a function is declared
    in an other package then the function call rewriting can not be
    done because Ora2Pg only knows about functions declared in the
    current schema. By setting a comma separated list of schema as
    value of the directive, Ora2Pg will look forward in these packages
    for all functions, procedures and packages declaration before
    proceeding to current schema export.
  - Add PG_SUPPORTS_NAMED_OPERATOR to control the replacement of the
    PL/SQL operator used in named parameter => with the PostgreSQL
    proprietary operator := Disable this directive if you are using
    PG < 9.5
  - Add a warning when Ora2Pg reorder the parameters of a function
    following the PostgreSQL rule that all input parameters following
    a parameter with a default value must have default values as well.
    In this case, Ora2Pg extracts all parameters with default values
    and put them at end of the parameter list. This is to warn you
    that a manual rewrite is required on calls to this function.

New command line options have been added:

  - Add -N | --pg_schema command line option to be able to override
    the PG_SCHEMA configuration directive. When this option is set
    at command line, EXPORT_SCHEMA is automatically activated.
  - Add --no_header option with equivalent NO_HEADER configuration
    directive to output the Ora2Pg header but just the translated
    code.

There is also some behavior changes from previous release:

  - Remove SysTimestamp() from the list of not translated function,
    it is replaced with CURRENT_TIMESTAMP for a long time now.
  - Change migration assessment cost to 84 units (1 day) for type
    TABLE, INDEX and SYNONYM and to 168 units (2 days) for TABLE
    PARTITION and GLOBAL TEMPORARY TABLE, this is more realistic.
  - Set minimum assessment unit to 1 when an object exists.
    Improve PL/SQL code translation speed.
  - Change behavior of COMPILE_SCHEMA directive used to force Oracle
    to compile schema before exporting code. When this directive is
    enabled and SCHEMA is set to a specific schema name, only invalid
    objects in this schema will be recompiled. When SCHEMA is not set
    then all schema will be recompiled. To force recompile invalid
    object in a specific schema, set COMPILE_SCHEMA to the schema name
    you want to recompile. This will ask to Oracle to validate the
    PL/SQL that could have been invalidate after a export/import for
    example. The 'VALID' or 'INVALID' status applies to functions,
    procedures, packages and user defined types.
  - Default transaction isolation level is now set to READ COMMITTED
    for all action excluding data export.
  - Oracle doesn't allow the use of lookahead expression but you may
    want to exclude some objects that match the ALLOW regexp you have
    defined. For example if you want to export all table starting
    with E but not those starting with EXP it is not possible to do
    that in a single expression.
    Now you can start a regular expression with the ! character to
    exclude all objects matching the regexp given just after. Our
    previous example can be written as follow:   ALLOW   E.* !EXP.*
    it will be translated into
    
    	 REGEXP_LIKE(..., '^E.*$') AND NOT REGEXP_LIKE(..., '^EXP.*$')
    
    in the object search expression.
  - Fix quoting of PG_SCHEMA with multiple schema in search path. The
    definition of the search path now follow the following behavior:
      * when PG_SCHEMA is define, always set search_path to its value.
      * when EXPORT_SCHEMA is enabled and SCHEMA is set, the search_path
        is set the name of the schema.
  - Remove forcing of export_schema when pg_schema is set at command
    line. This could change the behavior of some previous use of these
    variables and the resulting value of the search_path but it seems
    much better understandable.
  - Rewrite translation of raise_application_error to use RAISE
    EXCEPTION with a message and the SQLSTATE code. Oracle user
    defined code -20000 to -20999 are translated to PostgreSQL 
    user define code from 45000 to 45999. Call to
  	raise_application_error(mySQLCODE, myErrmsg);
    will be translated into
   	RAISE EXCEPTION '%', myErrmsg USING ERRCODE = mySQLCODE;
  - Remove migration assessment cost for TG_OP and NOT_FOUND they
    might be fully covered now.

Here is the complete list of changes:

  - Fix bad inversion of HAVING/GROUP BY clauses. Thanks to bhoot929
    for the report.
  - Fix case of non translation of type in CAST() function. Thanks to
    Keshavkumbham for the report.
  - Fix spatial data export when using direct import into PostgreSQL
    and WKT or INTERNAL format. This can still be improved. Thanks to
    Valeria El-Samra for the report.
  - Improve translation of trunc() into date_trunc. Thanks to bhoot929
    for the report.
  - Translate to_char() without format into a simple cast to varchar.
    Thanks to bhoot929 for the report.
  - Fix line comment which does not disable multi-line comment.
    Thanks to Pavel Stehule for the report.
  - Fix overridden of output file global_variables.conf with
    multiple packages. Thanks to Oliver Del Rosario for the report.
  - Fix export of data stored in a nested user defined type. Thanks
    to lupynos for the report.
  - Fix data export from Oracle user defined types, where output of
    ROW(...) does not distinguish numeric from string or other types
    that need to be formatted. Thanks to Petr Silhak for the report.
  - Fix broken replacement of package procedure name. Thanks to Pavel
    Stehule for the report.
  - Add FLOWS_010600 to the objects exclusion listr.
  - Improve view/trigger migration assessment accuracy.
  - Fix OUTER JOIN (+) translation, all join filters with constant
    was written into the WHERE clause by default. Write them into the
    JOIN clause.
  - Fix weight of the number of triggers and views in the report with
    a limit of 2 man-days, of course SQL difficulties are still add
    after this limit.
  - Fix alias added to condition which is not a sub query. Thanks to
    nitinmverma for the report.
  - Fix wrong translation of OUTER JOIN with subquery in FROM clause.
    Thanks to nitinmverma for the report.
  - Fix typo preventing exclusion of system SYNONYM.
  - Fix an other case of bad translation of END fct_name. Thanks to
    nitinmverma for the report.
  - Fix unwanted translation of REGEXP_SUBSTR() in REGEXP_SUBSTRING().
    Thanks to nitinmverma for the report.
  - Fix broken translation of decode(). Thanks to nitinmverma for the
    report.
  - Fix error "Malformed UTF-8 character (unexpected continuation byte
    0xbf, with no preceding start byte) in pattern match" by including
    an arbitrary non-byte character into the pattern. Thanks to Bob
    Sislow for the report.
  - Fix missing translation of trunc() with date_trunc(). Thanks to
    nitinmverma for the report.
  - Add migration assessment weight to concatenation.
  - Fix space between operator, ex: a < = 15 must be translated as
    a <= 15. Thanks to nitinmverma for the report.
  - Handle translation of named parameters in function calls. Thanks
    to Julien Rouhaud for the patch.
  - Fix missing renaming of _initial_command() method.
  - Fix right outer join translation by converting them to left outer
    join first. Thanks to Julien Rouhaud for the hint.
  - Fix TEST on default value count and functions belonging to others
    than public schema, especially functions of packages.
  - Fix default number of man-days in migration assessment. Thanks to
    Nate Fitzgerald for the report.
  - Add host information into the filename of the report to prevent
    some additional case of report overriding. Thanks to Aurelien
    Robin for the report.
  - Prevent ora2pg script to complain if no ora2pg.conf file is found
    when a DSN is passed at command line and that user connection is
    set in the environment variables.
  - Do not declare a function stable if there is update/insert/delete
    statement inside.
  - Improve ora2pg_scanner to generate report of each instance schema
    when the schema to audit is not set. Thanks to Thomas Reiss for
    the patch.
  - Fix parser failure with quote in comments. Thanks to Eric Delanoe
    for the report.
  - Fix case where NVL is not replaced by COALESCE.
  - Add parenthesis to simple package.function call without parameter.
  - Fix replacement of INSTR() with optional parameters. Thanks to
    Pavel Stehule for the report.
  - Translate SQL%ROWCOUNT to GET DIAGNOSTICS rowcount = ROW_COUNT.
    Thanks to Pavel Stehule for the report.
  - Add translation of SQL%FOUND. Thanks to Pavel Stehule.
  - Remove qualifier in create type, "CREATE TYPE fusion.mytable AS
    (fusion.mytable fusion.finalrecord[]);" becomes "CREATE TYPE
    fusion.mytable AS (mytable fusion.finalrecord[]);". Thanks to
    Julien Rouhaud for the report.
  - Fix extra comma in FROM clause of triggers in outer join
    translation. Thanks to Pavel Stehule fora the report.
  - Use record for out parameters replacement only where there is more
    than one out parameter. Thanks to Pavel Stehule for the patch.
  - Add date type to the inout type array. Thanks to Pavel Stehule for
    the report.
  - Remove possible last spaces in inout type detection.
  - Fix REGEXP_LIKE translation.
  - Fix count of default values during test action.
  - Fix removing of double quote over column name in view declaration.
  - Do not set default value if it is NULL, this is already the case.
  - Fix data export that was truncated to the last DATA_LIMIT lines.
    Thanks to Michael Vitale for the report.
  - Fix an other bug in rewriting call to function with OUT parameter.
    Thanks to Pavel Stehule for the report.
  - Fix autodetection of composite out parameters.
  - Merge typo on PLPGSQL
  - Fix typo to PLPGSQL keyword. Thanks to Vinayak Pokale.
  - Fix regexp failure in is_reserved_words() method. Thanks to
    Patrick Hajek for the patch.
  - Use only one declaration of ora2pg_r RECORD, it is reusable.
  - Fix transformation of procedure CALL with OUT parameter that can't
    works when procedure/function has minimally one OUT parameter is
    of composite type. Thanks to Pavel Stehule for the patch.
  - Second attempt to fix outer join translation in triggers. Thanks
    to Pavel Stehule for precious help.
  - Fix RAISE NOTICE replacement with double % placeholder. Thanks to
    Pavel Stehule for the report.
  - Fix call to function replacement for function registered with
    double quote. Thanks to Pavel Stehule for the report.
  - Change assessment score of TG_OP.
  - Fix replacement of outer join in triggers by adding pseudo tables
    NEW and OLD to the list of tables. Thanks to Pavel Stehule for the
    report.
  - Handle custom exception in declare section of triggers.
  - Fix FUNCTION_CHECK option, it will be set in all file header.
    Thanks to Pavel Stehule for the report.
  - Replace call to ALL_TABLES to USER_TABLES when USER_GRANTS is
    enabled. Thanks to Bob Sislow.
  - Removed PRAGMA EXCEPTION_INIT() from declare section. Thanks to
    Pavel Stehule for the report.
  - Fix constant string that was breaking the parser. Thanks to Pavel
    Stehule for the report.
  - Fix missing space between EXCEPTION and WHEN. Thanks to Pavel
    Stehule for the report.
  - Fix broken function header resulting in missing space before OUT
    keyword. Thanks to Pave Stehule for the report.
  - Fix invalid RAISE command. Thanks to Pavel Stehule for the report.
  - Add port information to ORACLE_DSN in documentation and config
    file. Thanks to Bob Sislow for the report.
  - Fix broken declaration in triggers related to FOR cycle control
    variable when there is no declare section. Thanks to Pavel
    Stehule for the report.
  - Fix broken declaration in triggers related to FOR cycle control
    variable. Thanks to Pavel Stehule for the report.
  - Fix unterminated C style comment in trigger. Thanks to Pavel
    Stehule for the report.
  - Fix bug in package+function precedence replacement. Thanks to
    Eric Delanoe for the report.
  - Fix unwanted and broken export of tables created with CREATE TABLE
    tablename OF objType. Thanks to threenotrump for the report.
  - Add explanation on using REPLACE_AS_BOOLEAN when REPLACE_TABLES or
    REPLACE_COLS is also used on the same object. Thanks to Brendan
    Le Ny for the report.
  - Fix unwanted data export of materialized view. Thanks to Michael
    Vitale for the report. Fix ORA-00942 with table that are not yet
    physically created and has no data.
  - Fix calling functions with same name declared in several packages.
    The one declared in current package now takes precedence. Thanks
    to Eric Delanoe for the report.
  - Change zero-length lob/long to undef workaround for a bug in
    DBD::Oracle with the ora_piece_lob option (used when no_lob_locator
    is enabled) where null values fetch as empty string for certain
    types. Thanks to Alice Maz for the patch.
  - Fix createPoint() spatial method issue. Thanks to jwl920919.
  - Fix comment on REPLACE_COLS directive.
  - Fix an other issue in transformation of TYPE x IS REF CURSOR.
    Thanks to Pavel Stehule for the report.
  - Fix an other case of broken declaration in triggers related to FOR
    cycle control variables. Thanks to Pavel Stehule for the report.
  - Fix broken declaration in triggers related to FOR cycle control
    variables with empty DECLARE section.
  - Fix other case of replacement by EXIT WHEN NOT FOUND.
  - Fix output of global_variables.conf file when OUTPUT_DIR is not
    set. Fix non replacement of global variables.
  - Remove some AUTHID garbage in procedure declaration generated by
    a migration.
  - Fix trigger name quoting with non supported character. Thanks to
    Michael Vitale for the report.
  - Fix use of nextval() in default value.
  - Fix alias append in from clause of the extract() function. Thanks
    to Narayanamoorthys for the report.
  - Disable direct export to partition for PostgreSQL 10 if directive
    PG_SUPPORTS_PARTITION is enabled, import must be done in the main
    table.
  - Do not export partitions of a materialized view. Thanks to Michael
    Vitale for the report.
  - Fix wrong replacement of keyword after END. Thanks to Pavel
    Stehule for the report.
  - Remove Oracle hints from queries, they are not supported and can
    break comments. Thanks to Pavel Stehule for the report.
  - Fix unstable transformation of TYPE x IS REF CURSOR. Thanks to
    Pavel Stehule for the report.
  - Fix data export failure when no table match the ALLOW/EXCLUDE
    filter. Thanks to threenotrump for the report.
  - Add missing search_path to FKEY dedicated file. Thanks to Michael
    Vitale for the report.
  - Apply default oject name exclusion to synonym.
  - Skip some PL/SQL translation in migration assessment mode.
  - Change default type for virtual column whit round() function.
    Thanks to Julien Rouhaud for the report.
  - Fix bug with EXIT WHEN command. Thanks to Pavel Stehule.
  - Fix an other wrong replacement of DECODE in UPDATE statement.
    Thanks to Pavel Stehule for the report.
  - Fix wrong replacement of DECODE. Thanks to Pavel Stehule.
  - Fix unwanted replacement of INTO STRICT when this is an INSERT
    statement. Thanks to Pavel Stehule for the report.
  - Fix potential regexp error with special character in outer join
    filters. Thanks to Adrian Boangiu for the report.
  - Fix parsing of PK from file. Thanks to Julien Rouhaud.
  - Fix parsing of FK from file. Thanks to Julien Rouhaud.
  - Fix count of unique and primary key in TEST export. Thanks to
    Liem for the report.
  - Fix reserved keyword rewrite using double quote when directive
    USE_RESERVED_WORDS is enabled. Thanks to Michael Vitale.
  - Remove EVALUATION CONTEXT object type from migration assessment.
  - Add QUEST_SL_% pattern to the list of table that must be excluded
    from export.
  - Fix data export when BFILE are translated as text. Thanks to
    Michael Vitale for the report.
  - Fix export of package when a comment is placed just before the
    AS/IS keyword. Thanks to Michael Vitale for the report.
  - Fix other cases of function call replacement when they are
    declared in different packages and one with OUT parameter and
    the other one with only IN parameters. Thanks to Eric Delanoe.
  - Fix inconsistent behavior of import_all script with -h and -d
    Under Linux: When -h not specified, script defaults to unix domain
    sockets for psql and localhost for perl (which may error depending
    on pg_hba.conf). Now defaults to more performing sockets. -d
    wasn't passing DB name to some psql calls where it's necessary.
    Thanks to BracketDevs for the patch.
  - Fix file handle close when compression is enabled. Thanks to
    Sebastian Albert for the report.
  - Add information about ora2pg behavior during data export to files
    when files already exists. Thanks to Michael Vitale.
  - Update readme to provide tar command for bzip2 file. Thanks to
    Tom Pollard for the patch
  - Fix unwanted FTS_INDEXES empty file and append unaccent extension
    creation if not exists. Thanks to Michael Vitale for the report.
  - Fix missing explicitly declared variable for cycle in trigger.
    Thanks to Pavel Stehule for the report.
  - Fix global type/cursor declaration doubled in package export.
  - Always translate Oracle SELECT ... INTO to SELECT ... INTO STRICT
    in plpgsql as Oracle seems to always throw an exception. Thanks
    to Pavel Stehule for the report.
  - Fix too much semicolons on end of function. Thanks to Pavel
    Stehule for the report.
  - Fix ALTER TABLE to set the owner when table is a foreign table.
    Thanks to Narayanamoorthys for the report.
  - Fix case of untranslated procedure call when there was parenthesis
    in the parameters clause. Thanks to Pavel Stehule for the report.
  - Fix broken variable declaration with name containing DEFAULT.
    Thanks to Pavel Stehule for the report.
  - Change query ORDER BY clause to view export query.
  - Fix missing replacement of quote_reserved_words function by new
    quote_object_name. Thanks to Sebastian Albert for the report.
  - Fix REPLACE_COLS replacement of column name in UNIQUE constraint
    definition. Thanks to Bernard Bielecki for the report.
  - Fix export of Oracle unlogged table that was exported as normal
    tables.
  - Fix regression in package function calls rewrite leading to append
    unwanted comma when replacing out parameters. Thanks to Pavel
    Stehule and Eric Delanoe for the report.
  - Fix removing of function name after END keyword. Thanks to Pavel
    Stehule for the report.
  - Fix bug in package function extraction.
  - Improve VIEW export by only looking for package function name and
    fix a bug that was including unwanted "system" package definition.
    Also fix a potential bad rewriting of function call. Thanks to
    Eric Delanoe for the report.
  - Fix an other case of missing PERFORM replacement. Thanks to Pavel
    Stehule for the report.
  - Fix remplacement of "EXIT WHEN cursor%NOTFOUND". Thanks to Pavel
    Stehule for the report.
  - Fix missing conversion of type in cast function. Thanks to Michael
    Vitale for the report.
  - Fix TO_NUMBER that is now translated as a cast to numeric to
    correspond to the default behavior in Oracle. Thanks to Pavel
    Stehule for the report.
  - Fix replacement of function call from different schema, especially
    in overloaded cases.
  - Remove OUT parameter from the argument list of function call.
    Thanks to Pavel Stehule for the report.
  - Fix wrong replacement in FOR ... IN loop inserting EXCLUDE in the
    statement. Thanks to Pavel Stehule for the report.
  - Translate Oracle proprietary VALUES without parenthesis with the
    proprietary syntax of POstgreSQL. Thanks to Pavel Stehule for the
    report.
  - Fix function header translation when a comment is present between
    closing parenthesis  and the IS keyword. Thanks to Pavel Stehule
    for the report.
  - Fix RETURNS in autonomous transaction call when there is OUT
    parameters. Thanks to Pavel Stehule for the report.
  - Fix call to BMS_UTILITY.compile_schema() when COMPILE_SCHEMA is
    enable. Thanks to PAvel Stehule for the report.
  - Fix export of function and procedure with same name in different
    schema. Thanks to Pavel Stehule for the report.
  - Fix detection and replacement of global variable in package that
    was producing invalid code export. Fix progress bar on procedure
    export.
  - Fix regression in global variables default value export.
  - Rewrite multiprocess for procedure and function export to solve
    some performances issues.
  - Do not waste time trying to replace function call when it is not
    found in the current code.
  - Fix default value for FILE_PER_FUNCTION when parallel mode is
    enabled.
  - Output a fatal error with export type TABLE and multiple schema set
    to PG_SCHEMA when EXPORT_SCHEMA is enabled.
  - Fix replacement of function name with package prefix.
  - Fix documentation of PG_SCHEMA directive, especially on the use of
    a schema list. Thanks to Michael Vitale for the report.
  - Fix translation of INSTR() function.
  - Improve speed in function translation by not calling twice
    Ora2Pg::PLSQL::convert_plsql_code() on declare and code section.
    Thanks to Pavel Stehule for the profiling.
  - Fix unwanted replacement of SYSDATE, SYSTIMESTAMP and some other
    when they are part of variable or object name. Add rewrite of
    REF CURSOR during type translation.
  - Require support of LATERAL keyword for DATADIFF (Pg >= 9.3).
    Patch from Sebastian Albert.
  - Do not call replace_sdo_function(), replace_sdo_operator() and
    replace_sys_context() if the string SDO_ or SYSCONTEXT is not
    found. This might save some performances.
  - Remove the RETURNS clause when there is an OUT parameter
    PostgreSQL choose correct type by self. Thanks to Pavel Stehule
    for the report.
  - Add a note about performance improvement by updating stats on
    Oracle. Thanks to Michael Vitale for the report.
  - Remove newline characters in REVOKE statement when embedded in
    a comment. Thanks to Pavel Stehule for the report.
  - Fix replacement with PERFORM into package extracted from an
    Oracle database. Thanks to Eric Delanoe for the report.
  - Fix translation of call to function with out parameters.
    Thanks to Pavel Stehule for the report.
  - Fix case where call to procedure without parameter was not
    prefixed by PERFORM or when called in a exception statement.
    Thanks to Eric Delanoe for the report.
  - Add function quote_object_name to handle all cases where object
    name need to be double quoted (PRESERVE_CASE to 1, PostgreSQL
    keyword, digit in front or digit only and non supported character.
    Thanks to liemdt1811 for the report.
  - Add a note about RAW(n) column with "SYS_GUID()" as default value
    that is automatically translated to type of the column 'uuid'
    by Ora2Pg.
  - Remove old column count check to use char_length. Thanks to
    Alice Maz for the patch.
  - Fix some raise_application_error that was not replaced with a
    global rewrite of remove comments and text constants to solve
    some other issues like rewriting of package function call in
    dynamic queries. Thanks to PAvel Stehule for the report.
  - Fix cycle variable not generated for LOOP IN SELECT in trigger.
    Thanks to Pavel Stehule for the report.
  - Fix procedures with OUT parameters not processed in triggers.
    Thanks to Pavel Stehule for the report.
  - Remove other case where PERFORM must be or must not be inserted.
  - Remove case where PERFORM can be inserted. Thanks to Pavel
    Stehule and Eric Delanoe for the report.
  - Fix missing ; in some raise_application_error translation. Thanks
    to Pavel Stehule for the report.
  - Fix missing PERFORM in front of direct call to function and the
    rewrite of direct call to function with out parameters. Thanks
    to Eric Delanoe for the report.
  - Fix translation of rownum when the value is not a number. Thanks
    to Pavel Stehule for the report.
  - Fix missing space between cast and AS keyword. Thanks to Pavel
    Stehule for the report.
  - Fix translation of views and add support to comment inside views.
    Thanks to Pavel Stehule for the report.
  - Fix removing of AS after END keyword. Thanks to Pavel Stehule for
    the report.
  - Fix type in CAST clause not translated to PostgreSQL type. Thanks
    to Pavel Stehule for the report.
  - Treat citext type as text. Thanks to Tomasz Wrobel for the patch.
  - Fix packages migration assessment that was broken with parser
    rewriting on package extraction.
  - Rewrite parsing of PL/SQL packages to better handle package
    specification and especially types and global variables from this
    section.
  - Fix raise_application_error translation by removing extra boolean
    parameter.
  - Improve comments processing.
  - Fix package function name replacement adding a dot before package
    name. Thanks to Eric Delanoe for the report.
  - Add collect of functions/procedures metadata when reading DDL
    from file.
  - Fix replacement of function prefixed with their schema name.
    Thanks to Eric Delanoe for the report.
  - Try to minimized comment placeholders by aggregating multiline
    comments.
  - Remove new line character from _get_version() output.
  - Fix ENABLE_MICROSECOND test condition on NLS_TIMESTAMP_TZ_FORMAT
    setting. Thanks to Didier Sterbecq for the report.
  - Fix another issue with Oracle 8i and table size extraction.
  - Fix query to show column information on Oracle 8i
  - Fix query to look for virtual column on Oracle 8i
  - Fix query to list all table by size on Oracle 8i
  - Prevent ora2pg to look for external table definition in Oracle 8i.
  - Fix a regression on timestamp format setting for Oracle 8i.
  - Fix some regression on queries with Oracle 8i. Thanks to Didier
    Sterbecq  for the report.
  - Add a function to collect metadata of all functions.
  - Don't create empty partition index file when there's no partition.
  - Fix wrong translation in OPEN ... FOR statement. Thanks to Eric
    Delanoe for the report.
  - Fix call of method close() on an undefined value. Thanks to Eric
    Delanoe for the report.
  - Fix partition data export issues introduced with previous patches.
  - Fix unterminated IF / ELSIF block in subpartition export.
  - Fix subpartition export. Thanks to Maurizio De Giorgi for the
    report.
  - Force DATA_LIMIT default value to 2000 on Windows OS instead of
    10000 to try to prevent constant OOM error. Thanks to Eric Delanoe
    for the report.
  - Fix default partition table that was not used PREFIX_PARTITION.
    Thanks to ssayyadi for the report.
  - Limit datetime microsecond format to micro second (.FF6) as the
    format can be FF[0..9] and PostgreSQL just have FF[0..6]
  - Add to_timestamp_tz Oracle function translation. Thanks to Eric
    Delanoe for the feature request.
  - Fix custom data type replacement in function code. Thanks to Pavel
    Stehule for the report.
  - Fix non working INPUT_FILE configuration directive when action is
    QUERY. Thanks to Eric Delanoe for the report.
  - Fix unwanted global variable implicit declaration to handle
    autonomous transaction parameters. Thanks to Eric Delanoe for the
    report.
  - Fix call to dblink in function with PRAGMA AUTONOMOUS_TRANSACTION
    and no arguments. Thanks to Eric Delanoe for the report.
  - Fix package constant translation. Thanks to Eric Delanoe.
  - Fix unwanted alias on join syntax. Thanks to Eric Delanoe
    for the report.
  - Fix regression on dbms_output.put* translation. Thanks to Eric
    Delanoe for the report.
  - Fix handling of comments in statements to try to preserve them at
    maximum in the outer join rewriting.
  - Do not declare variable when it is an implicit range cursor, it
    do not need to be declared.
  - Export implicit variable in FOR ... IN ... LOOP as an integer if
    it don't use a select statement and export it as a RECORD when a
    statement is found. Thanks to Eric Delanoe and Pavel Stehule for
    the report.
  - Reduce migration assessment weight for CONNECT BY.
  - Fix derived table pasted two times in from clause. Thanks to Pavel
    Stehule for the report.
  - Fix some other unexpected ";" in function code. Thanks to Pavel
    Stehule for the report.
  - Remove %ROWTYPE in return type of function. Thanks to Pavel
    Stehule for the report.
  - Fix doubled AND in expression when a parenthesis is in front after
    rewriting. Thanks to Eric Delanoe for the report.
  - Fix unexpected ";" in function after post-body END when a comment
    is present. Thanks to Eric Delanoe for the report.
  - Fix unexpected ";" in some function variable declaration when a
    comment is at end of the declare section. Thanks to Eric Delanoe
    for the report.
  - Remove %ROWTYPE in function that have not been replaced with RECORD
    for cursor declaration. Thanks to Eric Delanoe for the report.
  - Fix removing of WHEN keyword after END. Thanks to Pavel Stehule for
    the report.
  - Fix missing table name with alias in from clause due to comments in
    the clause. I have also merge right and left outer join translation
    function into a single one, most of the code was the same.
  - Fix output order of outer join. Thanks to Pavel Stehule for the
    report.
  - Fix untranslated outer join in nested sub query. Thanks to Pavel
    Stehule for the report.
  - Rewrite again the decode() translation as a single function call
    for all replacement before any other translation.
  - Append table filter to check constraints extraction. Thanks to
    danghb for the report.
  - Fix issue with parenthesis around outer join clause. Thanks to
    Pavel Stehule for the report.
  - Move remove_text_constant_part() and restore_text_constant_part()
    function into the main module.
  - Include decode() replacement in recursive function call. Thanks
    to Pavel Stehule for the report.
  - Prevent removing of parenthesis on a sub select. Thanks to Pavel
    Stehule for the report.
  - Fix missing table exclusion/inclusion in column constraint export.
    Thanks to danghb for the report.
  - Fix an alias issue in view parsed from file.
  - Fix parsing of view from file when no semi comma is found.
  - Remove FROM clause without alias from migration assessment.
  - Fix order of outer join during translation. Thanks to Pavel
    Stehule for the report.
  - Fix case of missing alias on subquery in FROM clause. Thanks to
    Pavel Stehule for the report.
  - Fix missing alias replacement in nested subqueries. Thanks to
    Pavel Stehule for the report.
  - Fix wrong addition of aliases to using() in join clause
  - Fix nested decode replacement producing invalid CASE expression.
    Thanks to Pavel Stehule for the report.
  - Append aliases to subqueries in the from clause that do not have
    one. Thanks to Pavel Stehule for the report.

2017 02 17 - v18.1

This release fix several issues reported on outer join translation
thanks to the help of Pavel Stehule and reapply the commit on virtual
column export that was accidentally removed from v18.0. It also adds
several new features:

  - Remove CHECK constraints for columns converted into boolean using
    REPLACE_AS_BOOLEAN column.
  - Oracle function are now marked as stable by default as they can
    not modify data.

Two new configuration directives have been added:

  - DATE_FUNCTION_REWRITE: by default Ora2pg rewrite add_month(),
    add_year() and date_trunc() functions set it to 0 to force Ora2Pg
    to not translate those functions if translated code is broken.
  - GRANT_OBJECT: when exporting GRANT you can now specify a comma
    separated list of objects which privileges must be exported.
    Default is to export privileges for all objects. For example
    set it to TABLE if you just want to export privilege on tables.

and a new command line option:

  - Add -g | --grant_object command line option to ora2pg to be able
    to extract privilege from the given object type. See possible values
    with GRANT_OBJECT configuration directive.

Here is the complete list of changes:

  - Remove empty output.sql file in current directory with direct data
    import. Thanks to kuzmaka for the report.
  - Fix shell replacement of $$ in function definition in Makefile.PL
    embedded configuration file. Thanks to kuzmaka for the report.
  - Fix shell replacement of backslash in Makefile.PL embedded
    configuration file. Thanks to kuzmaka for the report.
  - Add warning level to virtual column notice.
  - Fix comment in where clause breaking the outer join association.
    Thanks to Pavel Stehule for the report.
  - Add parsing and support of virtual column from DDL file.
  - Reapply commit on virtual column export that was accidentally
    removed in commit d5866c9. Thanks to Alexey for the report.
  - Fix mix of inner join and outer join not translated correctly.
    Thanks to Pavel Stehule for the help to solve this issue.
  - Fix additional comma in column DEFAULT value from DDL input file.
    Thanks to Cynthia Shang for the report.
  - Fix comments inside FROM clause breaking translation to ANSI outer
    joins. Thanks to Pavel Stehule for the report.
  - Fix replacement of sdo_geometry type into function. Thanks to
    Saber Chaabane for the report.
  - Fix subquery in outer join clause. Thanks to Saber Chaabane for
    the report.
  - Fix duplicated subqueries placeholder in the from clause.
    Thanks to Saber Chaabane for the report.
  - Fix replacement of subquery place older during outer join rewrite.
    Thanks to Saber Chaabane for the report.
  - Add DATE_FUNCTION_REWRITE configuration directive. By default
    Ora2pg rewrite add_month(), add_year() and date_trunc() functions
    set it to 0 to force Ora2Pg to not translate those functions if
    translated code is broken. Thanks to Pavel Stehule for the feature
    request.
  - Do not report error when -g is used but action is not GRANT.
    Thanks to Shane Jimmerson for the report.
  - Oracle function can not modify data, only procedure can do that,
    so mark them as stable. Thanks to Pavel Stehule for the report.
  - Missed some obvious combination like upper/lower case or no space
    after AND/OR on outer join parsing and some other issues.
  - Add missing call to extract_subqueries() recursively. Thanks to
    Pavel Stehule for the report.
  - Add full support of outer join translation in sub queries.
  - Add translation of mixed inner join and Oracle outer join. Thanks
    to Pavel Stehule for the report.
  - Fix missing space between keyword AS and END from the decode()
    transformation. Thanks to Pavel Stehule for the report.
  - Fix parsing of outer join with UNION and translation to left join.
    Thanks to Pavel Stehule for the report.
  - Remove CHECK constraints for columns converted into boolean using
    REPLACE_AS_BOOLEAN column. Thanks to Shane Jimmerson for the
    feature request.
  - Fix regression on SQL and PLSQL rewrite when a text constant
    contained a semi-comma.
  - Add the GRANT_OBJECT configuration directive. When exporting GRANT
    you can specify a comma separated list of objects for which the
    privileges will be exported. Default is export for all objects.
    Here are the possibles values TABLE, VIEW, MATERIALIZED VIEW,
    SEQUENCE, PROCEDURE, FUNCTION, PACKAGE BODY, TYPE, SYNONYM and
    DIRECTORY. Only one object type is allowed at a time. For example
    set it to TABLE if you just want to export privilege on tables.
    You can use the -g option to overwrite it.  
    When used this directive prevent the export of users unless it is
    set to USER. In this case only users definitions are exported.
  - Add the -g | --grant_object command line option to ora2pg to be able
    to extract privilege from the given object type. See possible values
    with GRANT_OBJECT configuration directive.
  - Improve replacement of ROWNUM by LIMIT+OFFSET clause.
  - Fix extra semi-colon at end of statement.
  - Override ora2pg.spec with Devrim's one but with String::Random
    removing as it is no more used.

2017 01 29 - v18.0

This new major release adds several new useful features and lot of
improvements.

  * Automatic rewrite of simple form of (+) outer join Oracle's
    syntax. This major feature makes Ora2Pg become the first free
    tool that is able to rewrite automatically (+) outer join in
    command line mode. This works with simple form of outer join
    but this is a beginning.
  * Add export of Oracle's virtual column using a real column and
    a trigger.
  * Allow conversion of RAW/CHAR/VARCHAR2 type with precision in
    DATA_TYPE directive. Useful for example to transform all RAW(32)
    or VARCHAR2(32) columns into PostgreSQL special type uuid.
  * Add export NOT VALIDATED state from Oracle foreign keys and check
    constraints into NOT VALID constraints in PostgreSQL.
  * Replace call to SYS_GUID() with uuid_generate_v4() by default.
  * Add "CREATE EXTENSION IF NOT EXISTS dblink;" before an autonomous
    transaction or "CREATE EXTENSION IF NOT EXISTS pg_background;".
  * Major rewrite of the way Ora2Pg parse PL/SQL to rewrite function
    calls and other PL/SQL to plpgsql replacement. There should not
    be any limitation in rewriting when a function contains a sub
    query or an other function call inside his parameters.
  * Refactoring of ora2pg to not requires any dependency other than
    the Perl DBI module by default. All DBD drivers are now optionals
    and ora2pg will expect an Oracle DDL file as input by default.
  * Add export of Oracle's global variables defined in package. They
    are exported as user defined custom variables and available in
    a session. If the variable is a constant or have a default value
    assigned at declaration, ora2pg will create a new file with the
    declaration (global_variables.conf) to be included in the main
    configuration file postgresql.conf file.
  * Create full text search configuration when USE_UNACCENT directive
    is enabled using the auto detected stemmer or the one defined in
    FTS_CONFIG. For example:
  	CREATE TEXT SEARCH CONFIGURATION fr (COPY = french);
    	ALTER TEXT SEARCH CONFIGURATION fr ALTER MAPPING FOR
    		hword, hword_part, word WITH unaccent, french_stem;
    	CREATE INDEX place_notes_cidx ON places
    			USING gin(to_tsvector('fr', place_notes));
    
Changes and incompatibilities from previous release:

  * FTS_INDEX_ONLY is now enabled by default because the addition of
    a column is not always possible and not always necessary where a
    simple function-based index is enough.
  * Remove use to setweigth() on single column FTS based indexes.
  * Change default behaviour of Ora2Pg in Full Text Search index
    export.

A new command line option and some configuration directive have
been added:

  * Option -D | --data_type to allow custom data type replacement
    at command line like in configuration file with DATA_TYPE.
  * UUID_FUNCTION to be able to redefined the function called to
    replace SYS_GUID() Oracle function. Default to uuid_generate_v4.
  * REWRITE_OUTER_JOIN to be able to disable the rewriting of Oracle
    native syntax (+) into OUTER JOIN if rewritten code is broken.
  * USE_UNACCENT and USE_LOWER_UNACCENT configuration directives to
    use the unaccent extension with pg_trgm with the FTS indexes.
  * FTS_INDEX_ONLY, by default Ora2Pg creates an extra tsvector column
    with a dedicated triggers for FTS indexes. Enable this directive
    if you just want a function-based index like:
	CREATE INDEX ON t_document USING
		gin(to_tsvector('pg_catalog.english', title));
  * FTS_CONFIG, use this directive to force the text search stemmer
    used with the to_tsvector() function. Default is to auto detect
    the Oracle FTS stemmer. For example, setting FTS_CONFIG to
    pg_catalog.english or pg_catalog.french will override the auto
    detected stemmer.

There's also lot fixes of issues reported by users from the past two
months, here is the complete list of changes:

  - Fix return type in function with a single inout parameter and a
    returned type.
  - Prevent wrong rewrite of empty as null when a function is used.
    Thanks to Pavel Stehule for the report.
  - Add the UUID_FUNCTION configuration directive. By default Ora2Pg
    will convert call to SYS_GUID() Oracle function with a call to
    uuid_generate_v4 from uuid-ossp extension. You can redefined it
    to use the gen_random_uuid function from pgcrypto extension by
    changing the function name. Default to uuid_generate_v4. Thanks
    to sjimmerson for the feature request.
  - Add rewrite of queries with simple form of left outer join syntax
    (+) into the ansi form.
  - Add new command line option -D | --data_type to allow custom data
    type replacement at command line like in configuration file with
    DATA_TYPE.
  - Fix type in ROWNUM replacement expression. Thanks to Pavel Stehule
    for the report.
  - Add replacement of SYS_GUID by uuid_generate_v4 and allow custom
    rewriting of RAW type. Thanks to Nicolas Martin for the report.
  - Fix missing WHERE clause in ROWNUM replacement with previous patch
    thanks to Pavel Stehule for the report.
  - Fix ROWNUM replacement when e sub select is used. Thanks to Pavel
    Stehule for the report.
  - Fix wrong syntax in index creation with DROP_INDEXES enabled.
    Thanks to Pave Stehule for the report.
  - Remove replacement of substr() by substring() as PostgreSQL have
    the substr() function too. Thanks to Pavel Stehule for the report.
  - Move LIMIT replacement for ROWNUM to the end of the query. Thanks
    to Pavel Stehule for the report.
  - Fix text default value between parenthesis in table declaration.
    Thanks to Pavel Stehule for the report.
  - Fix return type when a function have IN/OUT parameter. Thanks to
    Pavel Stehule for the report.
  - Mark uuid type to be exported as text. Thanks to sjimmerson for
    the report.
  - Add EXECUTE to open cursor with like "OPEN var1 FOR var2;". Thanks
    to Pavel Stehule for the report.
  - Fix replacement of local type ref cursor. Thanks to Pavel Stehule
    for the report.
  - Add EXECUTE keyword to OPEN CURSOR ... FOR with dynamic query.
    Thanks to Pavel Stehule for the report.
  - Fix case sensitivity issue in FOR .. IN variable declaration
    replacement. Thanks to Pavel Stehule for the report.
  - Fix wrong replacement of cast syntax ::. Thanks to Pavel Stehule
    for the report.
  - Reactivate numeric cast in call to round(value,n).
  - Close main output data file at end of export.
  - Add virtual column state in column info report, first stage to
    export those columns as columns with associated trigger.
  - Fix unwanted replacement of REGEXP_INSTR. Thanks to Bernard
    Bielecki for the report.
  - Allow rewrite of NUMBER(*, 0) into bigint or other type instead
    numeric(38), just set DATA_TYPE to NUMBER(*\,0):bigint. Thanks to
    kuzmaka for the feature request.
  - Export partitions indexes into PARTITION_INDEXES_....sql separate
    file named. Thanks to Nicolas Martin for the feature request.
  - Fix fatal error when schema CTXSYS does not exists. Thanks to
    Bernard Bielecki for the report.
  - Fix missing text value replacement. Thanks to Bernard Bielecki
    for the report.
  - Fix type replacement in declare section when the keyword END was
    present into a variable name.
  - Export NOT VALIDATED Oracle foreign key and check constraint as
    NOT VALID in PostgreSQL. Thanks to Alexey for the feature request.
  - Add object matching of regex 'SYS_.*\$' to the default exclusion
    list.
  - Fix UTF8 output to file as the open pragma "use open ':utf8';"
    doesn't works in a global context. binmode(':encoding(...)') is
    used on each file descriptor for data output.
  - Improve parsing of tables/indexes/constraints/tablespaces DDL from
    file.
  - Improve parsing of sequences DDL from file.
  - Improve parsing of user defined types DDL from file.
  - Export Oracle's TYPE REF CURSOR with a warning as not supported.
  - Replace call to plsql_to_plpgsql() in Ora2Pg.pm by a call to new
    function convert_plsql_code().
  - Move export of constraints after indexes to be able to use USING
    index in constraint creation without error complaining that index
    does not exists.
  - Add "CREATE EXTENSION IF NOT EXISTS dblink;" before an autonomous
    transaction or "CREATE EXTENSION IF NOT EXISTS pg_background;".
  - Improve parsing of packages DDL from file.
  - When a variable in "FOR varname IN" statement is not found in the
    DECLARE bloc, Ora2Pg will automatically add the variable to this
    bloc declared as a RECORD. Thanks to Pavel Stehule for the report.
  - Major rewrite of the way Ora2Pg parse PL/SQL to rewrite function
    calls and other PL/SQL to plpgsql replacement. There should not
    be limitation in rewriting when a function contains a sub query
    or an other function call inside his parameters.
  - Fix unwanted SELECT to PERFORM transformation inside literal
    strings. Thanks to Pavel Stehule for the report.
  - Fix bug in DEFAULT value rewriting. Thanks to Pavel Stehule for
    the report.
  - Fix replacement of DBMS_OUTPUT.put_line with RAISE NOTICE.
  - Reset global variable storage for each package.
  - Improve comment parsing in packages and prevent possible infinite
    loop in global variable replacement.
  - Add the REWRITE_OUTER_JOIN configuration directive to be able to
    disable the rewriting of Oracle native syntax (+) into OUTER JOIN
    if it is broken.  Default is to try to rewrite simple form of
    right outer join for the moment.
  - Export types and cursors declared as global objects in package
    spec header into the main output file for package export. Types
    and cursors declared into the package body are exported into the
    output file of the first function declared in this package.
  - Globals variables declared into the package spec header are now
    identified and replaced into the package code with the call to
    user defined custom variable. It works just like globals variables
    declared into the package body.
  - Add auto detection of Oracle FTS stemmer and disable FTS_CONFIG
    configuration directive per default. When FTS_CONFIG is set its
    value will overwrite the auto detected value.
  - Create full text search configuration when USE_UNACCENT directive
    is enabled using the auto detected stemmer or the one defined in
    FTS_CONFIG. For example:
  	CREATE TEXT SEARCH CONFIGURATION fr (COPY = french);
    	ALTER TEXT SEARCH CONFIGURATION fr ALTER MAPPING FOR
    		hword, hword_part, word WITH unaccent, french_stem;
    	CREATE INDEX place_notes_cidx ON places
    			USING gin(to_tsvector('fr', place_notes));
  - Remove CONTAINS(ABOUT()) from the migration assessment, there no
    additional difficulty to CONTAINS rewrite.
  - Add ANYDATA to the migration assessment keyword to detect.
  - Allow conversion of CHAR/VARCHAR2 type with precision in DATA_TYPE
    directive. For example it's possible to transform all VARCHAR2(32)
    columns only into PostgreSQL special type uuid by setting:
        DATA_TYPE	VARCHAR2(32):uuid
    Thanks to sjimmerson for the feature request.
  - Update year in copyrights
  - Fix creation of schema when CREATE_SCHEMA+PG_SCHEMA are defined.
  - Fix renaming of temporary file when exporting partitions.
  - Move MODIFY_TYPE to the type section
  - Update documentation about globals variables.
  - Add export of Oracle's global variables defined in package. They
    are exported as user defined custom variables and available in
    a session. Oracle variables assignment are exported as call to:
            PERFORM set_config('pkgname.varname', value, false);
    Use of these variable in the code is replaced by:
            current_setting('pkgname.varname')::global_variables_type;
    the variable type is extracted from the pacjkage definition. If
    the variable is a constant or have a default value assigned at
    declaration, ora2pg will create file global_variables.conf with
    the definition to include in postgresql.conf file so that their
    values will already be set at database connection. Note that the
    value can always modified by the user so you can not have exactly
    a constant.
  - Fix migration assessment of view.
  - Remove call to FROM SYS.DUAL, only FROM DUAL was replaced.
  - Replace call to trim into btrim.
  - Improve rewrite of DECODE when there is function call inside.
  - Add function replace_right_outer_join() to rewrite Oracle (+)
    right outer join.
  - Improve view migration assessment.
  - Create a FTS section in the configuration file dedicated to FTS
    control.
  - Add USE_UNACCENT and USE_LOWER_UNACCENT configuration directives
    to use the unaccent extension with pg_trgm.
  - Do not create FTS_INDEXES_* file when there is no Oracle Text
    indexes.
  - Update query test score when CONTAINS, SCORE, FUZZY, ABOUT, NEAR
    keyword are found.
  - Remove use to setweigth() on single column FTS based indexes.
    Thanks to Adrien Nayrat for the report.
  - Update documentation on FTS_INDEX_ONLY with full explanation on
    the Ora2Pg transformation.
  - Refactoring ora2pg to not requires any dependency other than the
    Perl DBI module by default. All DBD drivers are now optionals and
    ora2pg will expect to received an Oracle DDL file as input by
    default. This makes easiest packaging or for any distribution that
    can not build a package because of the DBD::Oracle requirement.
    DBD::Oracle, DBD::MySQL and DBD::Pg are still required if you want
    Ora2Pg to migrate your database "on-line" but they are optional
    because Ora2Pg can also convert input DDL file, this is the
    default now. Thanks to Gustavo Panizzo for the feature request and
    the work on Debian packaging.
  - Remove String::Random dependency in rpm spec file, it is no used
    even if it was mentioned into a comment.
  - Exclude internal Oracle Streams AQ JMS types from the export.
    Thanks to Joanna Xu for the report.
  - Fix some other spelling issues. Thanks to Gustavo Panizzo for the
    patch.
  - Fix some spelling errors. Thanks to Gustavo Panizzo for the patch.
  - Revert patch 697f09d that was breaking encoding with input file
    (-i). Thanks to Gary Evans for the report.
  - Add two new configuration directive to control FTS settings,
    FTS_INDEX_ONLY and FTS_CONFIG.
   
2016 11 17 - v17.6

This release adds several new features:

  * Adds export of Oracle Text Indexes into FTS or pg_trgm
    based indexes,
  * Add export of indexes defined on materialized views
  * Allow export of materialized views as foreign tables
    when export type is FDW.
  * Add replacement of trim() by btrim().

Two new configuration directives have been added:

  * USE_INDEX_OPCLASS: when value is set to 1, this will force
    Ora2Pg to export all indexes defined on varchar2() and char()
    columns using *_pattern_ops operators. If you set it to a value
    greater than 1 it will only change indexes on columns where the
    character limit is greater or equal than this value.

  * CONTEXT_AS_TRGM: when enabled it forces Ora2Pg to translate
    Oracle Text indexes into PostgreSQL indexes using pg_trgm
    extension. Default is to translate CONTEXT indexes into FTS
    indexes and CTXCAT indexes using pg_trgm. Some time using
    pg_trgm based indexes is enough.

There's also some fixes of issues reported by users, here is the
complete list of changes:

  - Fixed non-use of custom temp_dir (-T). Thanks to Sebastian
    Albert for the patch.
  - Make export of FTS indexes from materialized view work as
    for tables.
  - Fix drop of indexes during export of data when DROP_INDEXES
    is enabled.
  - Remove double quote in function and procedure name from an input
    file to avoid creating a file with double quote in its name.
  - Fix export of unique index associated to a primary key.
  - Move OPTION (key "yes") of FDW table before NOT NUL constraint
    and default clause.
  - Fix some encoding issue during data export into file.
  - Rename FTS indexes prefix output file into FTS_INDEXES and
    export CTXCAT Oracle indexes as GIN pg_trgm indexes instead of
    FTS indexes.
  - Add export of indexes of type CTXCAT as FTS indexes.
  - Export triggers and update order for FTS indexes to separate file
    prefixed with FTS_INDEXES.
  - Exclude from export synonyms starting with a slash that correspond
    to incomplete deleted synonyms. Thanks to Nouredine Mallem for the
    report.
  - Add export of indexes defined on materialized views. Thanks to
    Nouredine Mallem for the report.
  - Fix export of foreign key and FTS indexes when looking at dba_*
    tables and multiple different schemas have the same fk or context
    indexes definition. Thanks to Nouredine Mallemfor the patch.
  - Fix export of CONTEXT or FULLTEXT Oracle index into PostgreSQL
    FTS with trigger and initial update statement.
  - Add configuration directive USE_INDEX_OPCLASS to force Ora2Pg to
    export all indexes defined on varchar2() and char() columns using
    those operators. A value greater than 1 will only change indexes
    on columns where the character limit is greater or equal than
    this value.
  - Fix FDW export of mysql tables. Thanks to yafeishi for the report.
  - Fix decode() rewrite. Thanks to Jean-Yves Julliot for the report.
  - Fix regression introduced into the export of NUMBER to integer
    like PG types.
  - Show partition name in progress bar instead of main table name.

2016 10 20 - v17.5

This is a maintenance release to fix several issues reported by users.
There is also some major improvement and new feature.

There is a new configuration directive or change default behavior:

  * Fix export of CLOBs and NCLOB that was truncated to 64 Kb.
  * PG_BACKGROUND : when enabled autonomous transactions will be
    built using Robert Haas extension pg_background instead of dblink.
    Default is to still used dblink as pg_background is available
    only for PostgreSQL >= 9.5.
  * All Perl I/O now use the open pragma instead of calling method
    binmode(). This will force input and output to utf8 using the
    Perl pragma:
	use open ':encoding(utf8)';
    when configuration directive BINMODE is not set or NLS_LANG is
    set to UTF8.
  * Ora2Pg will now export empty lob as empty string instead of NULL
    when the source column has NOT NULL constraint and that directive
    EMPTY_LOB_NULL is not activated.
  * Improve and fix progress bar especially when using JOBS/-J option.
  * Allow LOAD action to apply all settings defined in the input file
    on each opened session, this allow to use LOAD with export schema
    enabled. If settings are not set in the input file encoding and
    search_path is set from the ora2pg configuration settings.
  * NUMBER(*,0) is now exported as numeric(38) as well as a NUMBER
    with DATA_SCALE set to 0, no DATA_PRECISION and a DATA_LENGTH
    of 22. The last correspond to Oracle type INTEGER or INT.
  * Allow conversion of type with precision in DATA_TYPE directive.
    For example it is possible to transform all NUMBER(12,2) only
    into numeric(12,2) by escaping the comma. Example:
    	DATA_TYPE	NUMBER(12\,2):numeric(12\,2);...
  * Write data exported into temporary files (prefixed by tmp_) and
    renamed them at end of the export to be able to detect incomplete
    export and override it at next export.
  * Add export of type created in package declaration.
  * Export foreign key when the referenced table is not in the
    same schema.
  * Enabled by default PG_SUPPORTS_CHECKOPTION assuming that your Pg
    destination database is at least a 9.4 version.
  * Add 12 units to migration assessment report per table/column
    conflicting with a reserved word in PostgreSQL to reflect the
    need of code rewriting.
  * Output a warning when a column has the same name than a system
    column (xmin,xmax,ctid,etc.)
  * Replace SYSDATE by a call to clock_timestamp() instead of a call
    to LOCALTIMESTAMP in plpgsql code.
  * Add missing documentation about DISABLE_PARTITION directive used
    to not reproduce partitioning into PostgreSQL and only export
    partitioned data into the main table.
  * Show partition name in progress bar instead of main table name.

Here is the complete list of other changes:

  - Fix broken parallel table export (-P option).
  - Fix export of CLOBs and NCLOB that was truncated to 64Kb. Thanks
    to Paul Mzko for the patch.
  - Fix database handle in error report.
  - Fix use of wrong database handle to set search_path. Thanks to
    Paul Mzko for the report.
  - Ora2pg doesn't export schema ForeignKey constraint when connected
    as different DBA user. Thanks to Paul Mzko for the patch.
  - Fix Perl I/O encoding using open pragma instead of calling method
    binmode(). Thanks to Gary Evans for the report.
  - Force input to utf8 using Perl pragma: use open ':encoding(utf8)';
    when BINMODE is not set or NLS_LANG is UTF8.
  - Force ora2pg to export empty lob as empty string instead of NULL
    when the source column has a NOT NULL constraint and directive
    EMPTY_LOB_NULL is not activated. Thanks to Valeriy for the report.
  - Fix missing CASCADE attribute on fkey creation during data export
    when DROP_FKEY was enabled. Thanks to ilya makarov for the report.
  - Fix issue on converting NUMBER(*,0) to numeric, should be ported
    to numeric(38). Thanks to ilya makarov for the report.
  - Correct query for ForeignKey export from oracle. Thanks to ilya
    makarov for the patch.
  - Fix schema change in direct import of data to PostgreSQL.
  - Change query for foreign key extraction to keep the column order.
    Thanks to ilya makarov for the report.
  - Write data exported into temporary files (prefixed by tmp_) and
    renamed them at end of the export to be able to detect incomplete
    export and override it at next export. Thanks to Paul Mkzo for
    the feature request.
  - Fix infinite loop in blob extraction when error ORA-25408 occurs
    during ora_lob_read() call. Thanks to Paul Mzko for the report.
  - Fix order of columns in foreign keys definition. Thanks to ilya
    makarov for the report.
  - Fix export of partition by range on multicolumn. Thanks to Rupesh
    Admane for the report.
  - Update reserved keywords list. Thanks to Nicolas Gollet for the
    report.
  - Add ON DELETE NO ACTION on foreign key creation (DROP_FKEY) to
    obtain the same output than during constraints export.
  - Fix export of foreign key that was duplicating the columns in both
    part, local and foreign. Thanks to ilya makarov for the report.
  - Remove call to to_char(datecol, format) when exporting date and
    timestamp. This formating is no more needed as we are now forcing
    NLS_DATE_FORMAT and NLS_TIMESTAMP_FORMAT when opening a connection
    to Oracle using:
    	ALTER SESSION SET NLS_DATE_FORMAT='YYYY-MM-DD HH24:MI:SS
    and
    	ALTER SESSION SET NLS_TIMESTAMP_FORMAT='YYYY-MM-DD HH24:MI:SS
    This may result on some speed improvment during data export.
  - Fix parsing of packages from input file.
  - Add export of type created in package declaration. Thanks to
    dezdechado  for the report.
  - Fix converting of procedures with out arguments. Thanks to
    dezdechado for the report.
  - Update documentation about project management.
  - Fix replacement of = NULL by IS NULL in update statement.
    Thanks to dezdechado for the report.
  - Fix parsing of trigger from file that was broken and new line
    removed. Thanks to dezdechado for the report.
  - Fix erasing of quotes from text in triggers. Thanks to dezdechado
    for the report.
  - Fix "return new" on trigger function when there is exception.
    Thanks to dezdechado for the report and solution.
  - Fix conversion of INTEGER and INT into numeric(38) instead of
    numeric without precision. Thanks to dezdechado for the report.
  - Fix export of foreign key when the referenced table is not in the
    same schema. Thanks to Juju for the report.
  - Fix ddl create schema when EXPORT_SCHEMA and CREATE_SCHEMA are
    enabled but no schema is specified.
  - Fix export of NCHAR that was converted as char but was loosing its
    length definition. Thanks to lgerlandsen for the report.
  - Fix parsing of views using WITH statements. Thank to dezdechado
    for the report.
  - Fix case that breaks views/triggers definition when a semicolon
    is encountered in a string value when reading definition from
    file. Thanks to dezdechado for the report.
  - Fix included/excluded of sequences when using ALLOW/EXCLUDE
    directives. Thanks to Roman Sindelar for the report.
  - prepare options modified with some escaping improvements. Thanks
    to ioxgrey for the patch.
  - It seems that for a NUMBER with a DATA_SCALE set to 0, no
    DATA_PRECISION and a DATA_LENGTH of 22 in ALL_TAB_COLUMNS, Oracle
    use a NUMBER(38) instead. This correspond to Oracle type INTEGER
    or INT. I don't really understand the reason of this behavior,
    why not just using a data length of 38? ALL_TAB_COLUMNS and Ora2Pg
    reports a data length of 22, now Ora2Pg will report NUMBER(38) like
    the resulting type of the DESC command.
    
    The following Oracle table:
    
    	CREATE TABLE TEST_TABLE ( FIELD_1 INTEGER, FIELD_2 NUMBER );
    
    will be exported as follow by Ora2Pg:
    
    [1] TABLE TEST_TABLE (owner: HR, 0 rows)
    	FIELD_1 : NUMBER(38) => numeric
    	FIELD_2 : NUMBER(22) => bigint
    
    Oracle data type INTEGER and INT will be exported as numeric by
    default instead of an integer.
  - Fix parsing of function/procedure from file with comments after
    BEGIN statement.
  - Remove DEFERRABLE INITIALLY DEFERRED from CHECK constraints when
    parsed from file. Thanks to Felipe Lavoura for the report.
  - Fix double parenthesis in index definition when parsing index
    creation from file. Thanks to Felipe Lavoura for the report.
  - Fix parsing of COMMENT from file.
  - Fix undetected native Oracle type bug. Thanks to kvnema for the
    report.
  - Fix unwanted text formatting with bind value in INSERT action
    with direct import to PostgreSQL. Thanks to Oleg for the report.
  - Fix inversion of UPDATE_RULE and DELETE_RULE in foreign key
    creation for MySQL export. Thanks to Sebastian Albert for the
    report.
  - Update documentation about DEFER_FKEY and DROP_FKEY to report
    the new behavior.
  - Remove call to SET CONSTRAINTS ALL DEFERRED with direct import.
  - Fix use of NULL value in bind parameter that should be undefined
    (INSERT export mode only). Thanks to Oleg barabaka for the report.
  - Remove replacement of direct call to functions with PERFORM, there
    is too much false positive. Thanks to dezdechado for the reports.
  - Fix a typo in SYSDATE replacement. Thank to dezdechado for report.
  - Remove rewrite of concatenation in RAISE EXCEPTION. Thanks to
    dezdechado for the report.
  - Fix replacement of raise_application_error() when first argument
    is a variable. Thanks to dezdechado for the report.
  - Fix wrong insertion of PERFORM before some function calls. Thanks
    to dezdechado for the report.
  - Replace SYSDATE by a call to clock_timestamp() instead of call to
    LOCALTIMESTAMP in plpgsql code. Thanks to aleksaan for the report.
  - Allow use of comma for object name list separator instead of space
    as workaround on Window OS.
  - Fix documentation about MODIFY_TYPE. Thanks to Nicolas Gollet for
    the report.
  - Add missing documentation about DISABLE_PARTITION directive used
    to not reproduce partitioning into PostgreSQL and only export
    partitioned data into the main table. Thanks to Nicolas Gollet
    for the report.
  - Add information about how to export LONG RAW data type. They need
    to be exported as BLOB before into Oracle to be exported as BYTEA.
  - Fix case where select was wrongly replaced by perform in INSERT
    INTO with SELECT statement. Thanks to dezdechado for the report.
  - Fix links to ora2pg presentation. Thanks to Daniel Lenski for the
    patch.
  - Fix input parameters after one with a default value must also have
    defaults. Thanks to v.agapov fot the patch.
  - Fix debug mode that was interromping the last running table dump.
    Thanks to calbiston for the report.

2016 04 21 - v17.4

Errata in first release attempt.

  - Fix previous patch that does not handle blob case but just clob
  - Forgot to change back the query when EMPTY_LOB_NULL is not activated.
  - Put parenthesis around AT TIME ZONE expression

This is a maintenance release to fix several issues reported by users.
There is also some major data export speed improvement thanks to the
work of PostgreSQL Pro and a new RPM spec file provided by Devrim
Gunduz to be able to build RPM package for Ora2Pg.

There is a new configuration directive:

  - EMPTY_LOB_NULL: when enabled force empty_clob() and empty_blob()
    to be exported as NULL instead as empty string.

Here is the complete list of other changes:

  - Add EMPTY_LOB_NULL directive to force empty_clob() and empty_blob()
    to be exported as NULL instead as empty string. This might improve
    data export speed if you have lot of empty lob. Thanks to Alex
    Ignatov for the report.
  - Fix import_all.sh script to import grant and tablespace separately
    as postgres user and just after indexes and constraints creation.
  - Add parsing of tablespace from "alter table ... add constraint"
    with DDL input file. Thanks to Felipe Lavoura.
  - Remove --single-transaction in import_all.sh script with TABLESPACE
    import. Thanks to Guillaume Lelarge for the report.
  - Fix Makefile.PL to used with latest spec file from Devrim Gunduz
    and following the advice of calbiston.
  - Update spec file to v17.6 and latest change to Makefile.PL
  - Replace ora2pg.spec by postgressql.org spec file by Devrim Gunduz.
  - Generate man page to avoids rpmbuild error.
  - Fix Windows install. Thanks to Lorena Figueredo for the report.
  - Remove "deferrability" call for mysql foreign keys. Thanks to
    Jean-Eric Cuendet for the report.
  - Fix issue in restoring foreign key for mysql data export. Thanks
    to Jean-Eric Cuendet for the report.
  - Remove connection test to PostgreSQL instance as postgres or any
    superuser in import_all.sh
  - Fix creation of configuration directory.
  - Fix Makefile to dissociate CONFDIR and DOCDIR from PREFIX or
    DESTDIR. Thanks to Stephane Schildknecht for the report.
  - Fix date_trunc+add_month replacement issue. Thanks to Lorena
    Figueredo for the report.
  - Do not replace configuration directory in scripts/ora2pg if this
    is a RPM build. Thanks to calbiston for the report.
  - Return empty bytea when a LOB is empty and not NULL.
  - Regular expressions and conditions checks improvement in method
    format_data_type() to make it a bit faster on huge tables. Thanks
    to Svetlana Shorina for the patch.
  - Fix INSERT using on the fly data import with boolean values.
    Thanks to jecuendet for the report.
  - Allow MySQL data type to be converted into boolean. Thanks to
    jecuendet for the report.
  - Fix export of BIT mysql data type into bit bit varying. Thanks
    to jecuendet for the report.
  - Fix call to escape_copy/escape_insert function call.

2016 03 26 - v17.3

This release fix two regressions introduced in latest release.

  * Fix major bug in data export. Thanks to Frederic Guiet for the report.
  * Fix another regression with character data that was not escaped. Thanks
    to Frederic Guiet for the report.

2016 03 24 - v17.2

This is a maintenance release to fix several issues reported in new
LOB extraction method. There is also some feature improvement:

  * Allow NUMBER with precision to be replaced as boolean.
  * Allow full relocation of Ora2Pg installation using for
    example: perl Makefile.PL  DESTDIR=/opt/ora2pg

Here is the complete list of other changes:

  - Allow NUMBER with precision to be replaced as boolean. Thanks
    to Silvan Auer for the report.
  - Force empty LOB to be exported as NULL when NO_LOB_LOCATOR is
    activated to have the same behavior.
  - Fix case where a LOB is NULL and ora2pg reports error :
        DBD::Oracle::db::ora_lob_read: locator is not of type OCILobLocatorPtr
    LOB initialised with EMPTY_CLOB() are also exported as NULL
    instead of \\x
  - Fix replacement with PERFORM after MINUS. Thanks to Stephane
    Tachoires for the report.
  - Comment DBMS_OUTPUT.ENABLE calls. Thanks to Stephane Tachoire for
    the report.
  - Fix wrong replacement of SELECT by PERFORM after EXCEPT. Thanks
    to Stephane Tachoire for the report.
  - Apply ORACLE_COPIES automatic predicate on custom queries set with
    REPLACE_QUERY if possible. Thanks to pawelbs for the report.
  - Fix install of ora2pg.conf file in /etc/ instead of /etc/ora2pg/.
    Thanks to pawelbs for the report.
  - Add debug information before searching for custom type.
  - Attempt to fix error "ORA-01002: fetch out of sequence" when exporting
    data from a table with user defined types and ORACLE_COPIES. Thanks to
    pawelbs and Alex Ignatov fir the report.
  - Fix replacement of path to configuration file in scripts/ora2pg
  - Remove report sample from documentation about migration assessment
    report and replace it with a href link. Fix comment about export of
    domain index.
  - Always prefix table name with schema in Oracle row count, to prevent
    failure when the schema is not the connexion default.
  - Add pattern TOAD_PLAN_.* to the internal table exclusion list.
  - Fix modification of database owner search_path in import_all.sh auto
    generated script. Thanks to Stephane Tachoire for the report.

2016 02 29 - v17.1

This is a maintenance release to fix several issues reported in new
TEST action. There is also some feature improvement:

  * Add OPTIONS (key 'true') on table FDW export when a column is
    detected as a primary key.
  * Add DELETE configuration directive that allow a similar feature
    than the WHERE clause to replace TRUNCATE call by a "DELETE FROM
    table WHERE condition". This feature can be useful with regular
    "updates". Thanks to Sebastien Albert for the feature request.

Here is the complete list of other changes:

  - Fix the counter of user defined types and sequences in TEST action
  - Fix COPY import of data from column with user defined type with
    NULL value.
  - Fix DBD::Pg segmentation fault with direct INSERT import from
    column with user defined type.
  - Fix TEST action with multiple PG_SCHEMA export. Thanks to Michael
    Vitale for the report.
  - Fix documentation about PG_SCHEMA

2016 02 22 - v17.0

This new major release adds a new action type TEST to obtain a count
of all objects at both sides, Oracle and PostgreSQL, to perform a
diff between the two database and verify that everything have been
well imported. It also fixes several issues reported by users.

A new ora2pg command line option have been added to ora2pg script:

  * Add --count_rows command line option to perform a real row count
    on both side, Oracle and PostgreSQL, in TEST report.

Here is the complete list of changes and bugfixes:

  - Prefix direct call to function with a call to PERFORM. Thanks to
    Michael Vitale for the feature request.
  - Fix revoke call on function with multiline parameters declaration.
  - Fix auto setting of internal schema variable with mysql.
  - Define ORACLE_HOME with the corresponding environment variable in
    generic configuration when available and --init_project is used.
    Thanks to Stephane Tachoires for the report.
  - Fix documentation about exporting view as table.
  - Remove some obsolete code and display information when a view is
    exported as table.
  - Fix empty LOB data export with Oracle Lob locator (NO_LOB_LOCATOR
    set to 0).
  - Fix data export of partitions with single process mode and when
    FILE_PER_TABLE is enabled.
  - Fix export of RAW data type.
  - Fix missing $ to call to self variable. Thanks to NTLIS and Sirko
    for the report.
  - Force FKey to be initially immediate when deferred is not set.
    Thanks to Stephane Tachoire for the report.
  - Fix count of check constraint when a schema is forced.
  - Allow TEST action on mysql database too with some improvements
    and bug fix on the feature.
  - Fix index column renaming in mysql export.
  - Fix dblink extraction query when an exclusion is set.
  - Fix sequence name auto generation for mysql serial number.
  - Add --count_rows command line option to make optional the real
    row count in TEST report. This is useful when you have lot of
    data and do not want to loose time in call to count(*).
  - Update documentation about the TEST action and usage, see
    chapter "Test the migration".
  - Apply schema context on PostgreSQL side with TEST action.
  - Add TEST action type to ask Ora2Pg to count rows and all objects
    at both sides, Oracle and PostgreSQL, to verify that everything
    have been well imported.
  - Fix missing export of foreign keys on multiple columns, ex:
	ALTER TABLE products ADD CONSTRAINT fk_supplier_comp
		FOREIGN KEY (supplier_id,supplier_name)
		REFERENCES supplier(supplier_id,supplier_name)...
  - Fix import of BLOB data using INSERT statements into the bytea.
    Thanks to rballer for the patch.
  - Fix missing export of FK when no schema is provided.

2016 01 13 - v16.2

This release fixes several issues, is more accurates on migration
assessment report and adds some new ora2pg command line options:

  * Add --pg_dsn, --pg_user and --pg_pwd to be able to set the
    connection to import directly into PostgreSQL at command line.
  * Add -f option to script import_all.sh to force to not check
    user/database existing and skip their creation.

Potential backward compatibility issues:

  * PG_SUPPORTS_CHECKOPTION is now enabled by default, you may want
    to migrate to PostgreSQL 9.4 or above.
  * Remove modification of CLIENT_ENCODING in generic configuration
    file with --init_project, use the default instead.
  * Remove modification of directive NLS_LANG to AMERICAN_AMERICA.UTF8
    in generic configuration file with --init_project, use the default
    instead.

Here is the complete list of other changes:

  - Adjust DBMS_OUTPUT calls to the migration assessment count.
  - Fix migration assessment count of call to cursor%ISOPEN and
    cursor%ROWCOUNT.
  - Replace zero date also with prepared statement with online
    PostgreSQL import and INSERT action. Thanks to Sebastian Albert
    for the report.
  - Remove REFERENCING clause in conditional triggers. Thanks to Raqua
    for the report.
  - Fix position of TG_OP condition when an exception is defined.
    Thanks to Raqua for the report.
  - Fix wrong replacement of SELECT with PERFORM when a comment was
    found between an open parenthesis and the select statement.
    Thanks to Raqua for the report.
  - Fix procedure return type with OUT and INOUT parameter. Thanks to
    Raqua for the report.
  - Fix rewrite of triggers with referencing clause. Thanks to Raqua
    for the report.
  - Fix default number of --human_days_limit in usage.
  - Fix replacement of placeholder %TEXTVALUE-d% to hide text string
    in query during function call rewrite. Thanks to Lorena Figueredo
    for the report.
  - Fix progress bar when a WHERE clause is used to limit the number
    of row to export.
  - Fix error "DBD::Pg::db do failed: SSL error: decryption failed or
    bad record mac" with pararellel table export (-P) and direct
    import to PostgreSQL via a ssl connection. Thanks to pbe-axelor
    for the report.
  - Fix missing index name in indexes creation. Thanks to Raqua for
    the report.
  - Fix pg DSN in import_all.sh autogenerated script.
  - Fix extraction of trigger. When the name of a column or something
    contained INSERTING, DELETING or UPDATING was converted to TG_OP
    = 'INSERT' or corresponding event. Thanks to Stanislaw Jankowski
    for the patch.
  - Fix multiple use of same column in check constraint and indexes
    of partitions when there was several schema with the same objects.
  - Fix default value for HUMAN_DAY_LIMIT to 5 when it is not defined
    in ora2pg.conf.
  - Fix double quote on column name in COPY export of partition tables
    Thanks to Chris Brick for the report.
  - Prevent case with several time same column in multicolumns unique
    constraints. Fix typo in previous patch.
  - Fix double quoted name with auto incremented sequence exported as
    serial.
  - Fix syntax error with MySQL data export with a WHERE clause using
    LIMIT.

2015 11 30 - v16.1

This release fixes several issues and adds some very useful features:

    * Generate automatically a new import_all.sh shell script when using option
      --init_project to help automate all import into PostgreSQL.

	See sh import_all.sh -? for more information.

    * Export Oracle bitmap index as PostgreSQL btree_gin index. This require the
      btree_gin extension and PostgreSQL >= 9.4. This is the default.

    * Auto set DEFINED_PK to the first column of a table that have a unique key
      defined that is a NUMBER. This allow data of any table with a numeric
      unique key to be extracted using multiple connexions to Oracle using -J
      option. Tables with no numeric unique key will be exported with a single
      process.

    * Improve BLOB export speed by using hex encoding instead of escape. This
      might speed up be BLOB export by 10.

    * Allow use of LOB locator to retrieve BLOB and CLOB data to prevent having
      to set LONGREADLEN. Now LONGREADLEN is set to 8KB. Old behavior using
      LONGREADLEN can still be enabled by setting NO_LOB_LOCATOR to 0, given
      for backward compatibility. Default is to use LOB locator.

    * Ora2Pg will also auto detect table with BLOB and automatically decrease
      DATA_LIMIT to a value lower or equal to 1000. This is to prevent OOM.

    * Improving indexes and constraints creation speed by using the LOAD action
      and a file containing SQL orders to perform. It is possible to dispatch
      those orders over multiple PostgreSQL connections. To be able to use this
      feature, PG_DSN, PG_USER and PG_PWD must be set. Then:

	ora2pg -t LOAD -c config/ora2pg.conf -i schema/tables/INDEXES_table.sql -j 4
    
      will dispatch indexes creation over 4 simultaneous PostgreSQL connections.
    
      This will considerably accelerate this part of the migration process with
      huge data size.

    * Domain indexes are now exported as b-tree but commented to let you know
      where possible FTS are required.

    * Add number of refresh ON COMMIT materialized view in detailed report.

    * Allow redefinition of numeric type, ex: NUMBER(3)::bigint to fix wrong
      original definition in Oracle.

    * Allow export of all schemas from an Oracle Instance when SCHEMA directive
      is empty and EXPORT_SCHEMA is enabled. All exported objects will be
      prefixed with the name of their original Oracle schema or search_path will
      be set to that schema name. Thanks to Magnus Hagander for the feature
      request.
 
    * Allow use of COPY FREEZE to export data when COPY_FREEZE is enabled. This
      will only works with export to file and when -J or ORACLE_COPIES is not
      set or default to 1. It can be used with direct import into PostgreSQL
      under the same condition but -j or JOBS must be unset or default to 1.
      Thanks to Magnus Hagander for the feature request.

Some new configuration directives:
 
    * BITMAP_AS_GIN: enable it to use btree_gin extension to create bitmap
      like index with pg >= 9.4. You will need to create the extension by
      yourself: "create extension btree_gin;". Default is to create GIN index,
      when disabled, a btree index will be created.
    * NO_LOB_LOCATOR: to disable use of LOB locator and extract BLOB "inline"
      using a less or more high value in LONGREADLEN.
    * BLOB_LIMIT: to force the value of DATA_LIMIT for tables with BLOB. Default
      is to automatically set this limit using the following code:
	BLOB_LIMIT=DATA_LIMIT; while (BLOB_LIMIT > 1000) BLOB_LIMIT /= 10
    * COPY_FREEZE: use it to use COPY FREEZE instead of simple COPY to speedup
      import into PostgreSQL.

Here is the complete list of other changes:

  - Limite package function name rewrite to call with parenthesis after the
    function name to avoid rewriting names of other objects.
  - Fix extra replacement of function name with package prefix. On some
    condition it was done multiple time.
  - Set REPLACE_ZERO_DATE to -INFINITY in generic configuration when --mysql
    is enabled.
  - Fix extraction of partition with MySQL that was not limited to a single
    database.
  - Do some replacement on ORACLE_DNS and SCHEMA into generic configuration
    when --mysql is used for better understanding.
  - Add call to round() on -J parallelization when the auto detected column
    is a numeric with scale.
  - Add COMMIT to the difficulties migration assessment keywords as it need
    context analyzing.
  - Add call to cursor's %ROWCOUNT, %ISOPEN and %NOTFOUND to difficulties
    migration assessment keywords.
  - Replace call to CURSOR%ROWTYPE by RECORD. Thanks to Marc Cousin for the
    report.
  - Fix ALTER FUNCTION ... OWNER TO ... and REVOKE statement where functions
    parameters were missing.
  - Add Get_Env to the Oracle functions list for migration assessment.
  - Disable variable NO_LOB_LOCATOR and set LONGREADLEN to 8192 bytes to use
    LOB locators to extract BLOB in generic configuration file.
  - Fix call method "disconnect" on unblessed reference at line 9998. Thanks
    to Stephane Tachoires for the report.
  - Exclude from export objects name matching /.*\$JAVA\$.*/ and /^PROF\$.*/.
  - Fix migration assessment report when created during the package export.
  - Force writing Oracle package body in separate files when FILE_PER_FUNCTION
    is enabled and PLSQL_PGSQL disable to obtain package source code.
  - Fix case where sequence max value is lower than start value, in this case,
    set max value = start value.
  - Fix missing newline after each package file to import in global package.sql
    file when FILE_PER_FUNCTION is enabled.
  - Remove export of user PUBLIC in GRANT export.
  - Set DISABLE_TRIGGERS to 1 in generic configuration file auto generated when
    ora2pg option --init_project is used.
  - Remove call to quote_reserved_words() with index column when it we detect
    a function based index, too much false positive are rewritten with SQL code
    like CASE...WHEN.
  - Update export_schema.sh to remove .sql files when there is not such object
    leaving export directory empty.
  - Prevent creating TBSP_INDEXES_tablespace.sql when no tablespaces are found
  - Update documentation on WHERE clause on how to limit the number of tuples
    exported for Oracle and MySQL to test data import.
  - Fix unlisted spatial indexes in assessment report.
  - Fix double quote on index name with index renaming and reserved keyword.
  - Do not try to export tablespaces, privileges and audited queries as non DBA
    user when USER_GRANT is enabled.
  - Remove carriage return from list file.
  - Force SCHEMA to database name with MySQL migration.
  - Fix missing declaration of _extract_sequence_info(). Thank to Yannick DEVOS
    for the report.
  - Add documentation about COPY_FREEZE directive and add a note about export
    of all schema.
  - Remove systematic schema name appended to table name on KETTLE export, this
    must only be true when EXPORT_SCHEMA is enabled.
  - Fix TO_NUMBER() wrong replacement when a function is called as a parameter.
  - Fix non converted DECODE() when they was called in an XMLELEMENT function.
  - Suppress MDSYS.SDO_* from MDSYS call in migration assessment cost.
  - Remove use of DBMS_STANDARD called with raise_application_error function
  - Fix STRING type replacement
  - Recreate README as a text file, not a man page.
  - Reformat changelog to 80 characters.
  - Add -t | --test command line option to ora2pg_scanner to be able to test
    all connections defined into the CVS list file.

2015 10 15 - v16.0

This major release improve PL/SQL code replacement, fixes several bugs and
adds some major new features:

  * Full migration of MySQL database, it just work like with Oracle database.
  * Full migration assessment report for MySQL database.
  * New script, ora2pg_scanner, to perform a migration assessment of all
    Oracle and MySQL instances on a network.
  * Add technical difficulty level in migration assessment.
  * Allow migration assessment on client queries extracted from AUDIT_TRAIL
    (oracle) or general_log table (mysql).
  * Ora2Pg has a "made in one night" brand new Web site (still need some work).
    See http://ora2pg.darold.net/

Example of technical difficulty level assessment output for the sakila database
with some more difficulties:

	Total	83.90 cost migration units means approximatively 1 man-day(s).
	Migration level: B-5

Here are the explanation of the migration level code:

    Migration levels:
	A - Migration that might be run automatically
	B - Migration with code rewrite and a human-days cost up to 10 days
	C - Migration with code rewrite and a human-days cost above 10 days
    Technical levels:
	1 = trivial: no stored functions and no triggers
	2 = easy: no stored functions but with triggers, no manual rewriting
	3 = simple: stored functions and/or triggers, no manual rewriting
	4 = manual: no stored functions but with triggers or views with code
	            rewriting
	5 = difficult: stored functions and/or triggers with code rewriting

This is to help you to find the database that can be migrated first with small
efforts (A and B) and those who need to conduct a full migration project (C).

This release has also some new useful features:

  * Export type SHOW_TABLE now shows additional information about table type
    (FOREIGN, EXTERNAL or PARTITIONED with the number of partition).
  * Connection's user and password can be passed through environment variables
    ORA2PG_USER and ORA2PG_PASSWD to avoid setting them at ora2pg command line.
  * Improve PL/SQL replacement on ADD_MONTH(), ADD_YEAR(), TRUNC(), INSTR() and
    remove the replacement limitation on DECODE().
  * Add detection of migration difficulties in views, was previously reserved
    to functions, procedures, packages and triggers.
  * Replace values in auto generated configuration file from command line
    options -s, -n, -u and -p when --init_project is used.
  * Adjust lot of scores following new functionalities in Ora2Pg, ex: dblink or
    synomyms are now easy to migrate.

There is some new command line options to ora2pg script:

  * -m | --mysql : to be used with --init_project and -i option to inform
    ora2pg that we work with a MySQL format
  * -T | --temp_dir : option to be able to set a distinct temporary directory
    to run ora2pg in parallel.
  * --audit_user : option to set the user used in audit filter and enable
    migration assessment report on queries from AUDIT_TRAIL (oracle) or
    general_log table (mysql).
  * --dump_as_sheet and --print_header options to be able to compute a CSV
    file with all migration assessment from a list of oracle database.
  * --dump_as_csv option to report assessments into a csv file. It will not
    include comments or details, just objects names, numbers and cost.

Backward compatibility:

  - Change NULL_EQUAL_EMPTY to be disabled by default to force change in the
    application instead of transforming the PL/SQL.

This release adds some new configuration directives:

  * MYSQL_PIPES_AS_CONCAT: Enable it if double pipe and double ampersand
    (|| and &&) should not be taken as equivalent to OR and AND.
  * MYSQL_INTERNAL_EXTRACT_FORMAT: Enable it if you want EXTRACT() replacement
    to use the internal format returned as an integer.
  * AUDIT_USER: Set the comma separated list of user name that must be used
    to filter from the DBA_AUDIT_TRAIL or general_log tables.
  * REPLACE_ZERO_DATE: "zero" date: 0000-00-00 00:00:00 it is replaced by a
    NULL by default, use it to use the date of your choice.
  * INDEXES_RENAMING: force renaming of all indexes using tablename_columnsname
    Very useful for database that have multiple time the same index name or
    that use the same name than a table.
  * HUMAN_DAYS_LIMIT: default to 5 days, used to set the number of human-days
    limit for migration of type C.

Here is the full list of other changes:

  - Remove list of acknowledgment that was not maintained anymore and some
    person may feel injured. Acknowledgment for patches or bug reports are
    always written to changelog, so this part reports to it now.
  - Fix bad trigger export when objects was enclosed in double quote and fix
    an additional bug in WHEN clause export. Thanks to Cyrille for the report.
  - Update documentation.
  - Update Makefile.PL with new script to install and new configuration
    directives in auto generated configuration file.
  - Update with new and missing files.
  - Add a Perl Module dedicated to MySQL database object discovery and export,
    lib/Ora2Pg/MySQL.pm.
  - Fix function based index type replacement in previous commit.
  - Do not report indexes with just DESC as function based index like Oracle
    report it. Thanks to Marc Cousin for the report.
  - Some excluded table was missing in the previous patch.
  - Remove use of DBI InactiveDestroy call when a fork is done and replace it
    to a single use AutoInactiveDestroy at connection. This require DBI>=1.614.
  - Add SDO_* tables and OLS_DIR_BUSINESSES in table exclusion list to fix issue
    #124 when no schema is provided. Thanks to Kenny Joseph for the report.
  - Fix partition prefix.
  - Remove UNIQUE keyword from spatial index.
  - Fix alter triggers function with missing parenthesis. Thanks to Spike Hodge
    MWEB for the report.
  - Fix export of foreign keys when they was defined in lowercase. Thanks to
    Spike for the report.
  - Fix wrong offset when rewriting ROWNUM with LIMIT+OFFSET. Thanks to Marc
    Cousin for the report.
  - Allow -INFINITY to be used to replace zero date.
  - Migration assessment in hour-day are now set to 1 man-day, we do not need
    such a precision and it is easier to process csv report. Thanks to Stephane
    Tachoire for the report.
  - Fix some issue with FDW and WKT spatial export. Add migration assessment
    of queries from the AUDIT_TRAIL table.
  - Adjust assessment units of some objects and add QUERY migration weight.
  - Rewrite information about migration levels.
  - Fix speedometer in progress bar, it will now shows the current speed in
    tuples/sec and the speed and time related to a table when export ended for
    the object. Thanks to Alex Ignatov for the report.
  - Fix break line when export data using INSERT mode. Thanks to Vu Bui for
    the report.
  - Do not display line about non existent objects in migration assessment
    reports.
  - Fix date default value for date when value is 0000-00-00 00:00:00
  - Suppress display of title for function and trigger details when there is
    no details.
  - Remove INSTR() from the list of Oracle function that are not supported.
    It is now replaced by position().
  - Fix condition to call _get_largest_tables().
  - Fix some minor issues in OUT/INOUT type returned by a function.
  - Fix default value that may appears unquoted.
  - Fix several issues on partition export: column with function, index on
    default partition table and plsql to plpgsql translation in check condition.
  - Fix some minor issues.
  - Replace values from command line options -s, -n, -u and -p in --init_project
    auto generated configuration file. Thanks to Stephane Tachoire for the
    feature request.
  - Fix wrong object count in SHOW_REPORT. Thanks to Stephane Tachoire for
    the report.
  - Use DBA_SEGMENTS to find database size when USER_GRANT is disable, aka user
    is a DBA
  - Remove report of Migration Level when --estimate_cost is not enabled.
  - Add missing BINARY_INTEGER for type replacement.
  - Always exclude function squirrel_get_error_offset() that is created by the
    Squirrel Oracle plug-in.
  - Adjust assessment scores following new functionalities in Ora2Pg, ex:
    autonomous transaction, dblink or synomyms are now easy to migrate.
  - Remove man page from source, it is auto generated by Makefile.PL and make.
  - Fix unterminated DECODE replacement when there was more than 5 parameters
    to DECODE() and remove the limitation to 10 parameters. There is no more
    limit in the number of decode parameters. Thanks to Mael Rimbault for the
    report.
  - Remove inclusion of unwanted object when exporting a limited list of view
    with ALLOW.
  - Disable unsupported recursive query used to reorder views when Oracle
    version is 11gR1. Thanks to Mael Rimbault for the patch.
  - Add PLPGSQL replacement of INSTR() by POSITION(). Thanks to Mael Rimbault
    for the report.
  - Add difficulty level information in migration assessment, this include a
    new configuration directive HUMAN_DAYS_LIMIT (default to 5 days) to set
    the number of human-days limit for migration of type C.
  - Add MERGE with a migration cost of 3, still need work be replaced by
    ON CONFLICT.
  - Remove some redundant regular expressions.
  - Fix escaped commas not working properly in MODIFY_TYPE. A MODIFY_TYPE
    value like `TABLE1:COL4:decimal(9\,6)` was leading to a column like
    `col4 decimal(8#nosep#3)` in the SQL dump file that was generated. This
    fixes the output to be `col4 decimal(8,3)`. Thanks to Nick Muerdter for
    the patch.
  - Strip default "empty_clob()" values from table dumps. This function does
    not exist in Postgres and is not necessary. Thanks to Nick Muerdter for
    the patch.
  - Fix undesired double quoting of column name in function based indexes.
  - Fix issue with Perl < 5.8 "Modification of a read-only value attempted"
  - Fix retrieving of table size on Oracle 8i.
  - Add auto double quoting of object name with unauthorized characters.
    Thanks to Magnus Hagander for the feature request.
  - Automatically double quote object name beginning with a number
  - Fix missing DESC part in descending indexes. Thanks to Magnus Hagander
    for the report.
  - Fix case where a column name in oracle is just a number (e.g. the column
    is called "7"), it will be created in postgres without quoted identifier,
    which fails. Thanks to Magnus Hagander for the report.
  - Fix "reqs/sec" display in debug mode. Thanks to Laurent Martelli for
    the patch
  - Fix export if Oracle procedure is created without a parameter. Thanks to
    dirkgently007 for the report.
  - Fix CSV report output.
  - Fix triggers from file parser.
  - Add a test on triggers return to handle case where it is triggered on
    DELETE + other(s) event(s). In this case a test is done on the TG_OP to
    return OLD if event is DELETE or NEW in other case. Thanks to Dominique
    Legendre for the suggestion.
  - Change NULL_EQUAL_EMPTY to be disabled by default to force change of the
    application instead of transforming the PL/SQL.
  - Change score of SYNONYM and DBLINK in the migration assessment.
  - Add conversion of Oracle type STRING into varchar(n) or text.
  - Add information about libaio1 requirement for instant client
  - Remove extra space when calling ora2pg_get_efile() used to export BFILE
    into EFILE. Thanks to Dominique Legendre for the export.


2015 06 01 - v15.3

This is a maintenance release only that fixes several minor bugs and typos.
The configuration file have been entirely rewritten to classify configuration
directives in section for better understanding.

Here is the full list of changes:

  - Ora2Pg will use DEFAULT_SRID when call to sdo_cs.map_oracle_srid_to_epsg()
    returns an ORA-01741 error. Mostly because there's no SRID defined for that
    column in ALL_SDO_GEOM_METADATA. The error message will still be displayed
    but a warning will explain the reason and ora2pg will continue with default
    value. Thanks to kazam for the report.
  - Add current setting for NLS_TIMESTAMP_FORMAT and NLS_DATE_FORMAT to the
    SHOW_ENCODING report.
  - Change default value for GEOMETRY_EXTRACT_TYPE to INTERNAL instead of WKT.
  - Change generic configuration file behavior with BINMODE parameter commented
    if it was previously uncommented. This will force to use the default value.
  - Fix potential issue with max open file limit with unclosed temporary file.
    Thanks to Marc Clement for the report.
  - Fix use of SECURITY DEFINER in SYNONYM export.
  - Fix parsing of editable function/procedure/package from input DML file.
  - Fix case where variable $2 and $3 was null after a too early call of a new
    substitution in read_view_from_file(). Thanks to Alex Ignatov for the patch.
  - Add support to "create or replace editionable|noneditionable" from input DML
    files. Thanks to Alex Ignatov for the report.
  - Fix unknown column HIGH_VALUE from *_TAB_PARTITIONS in Oracle 8i. Thanks to
    Sebastian Fischer for the patch.
  - Fix call to ALL_MVIEW_LOGS object which not exists with Oracle 8i. Thanks to
    Sebastian Fischer for the report.
  - Fix Error ORA-22905 while Get the dimension of the geometry by looking at
    number of element in the SDO_DIM_ARRAY. Thanks to jkutterer for the patch.
  - Remove reordering export of view for Oracle database before 11g. Thanks to
    kyiannis for the report.
  - Fix several some typos and a bunch of misspelled. Thanks to Euler Taveira
    for all the patches.
  - Fix missing Oracle database version before looking at function security
    definer. Thanks to kyiannis for the report.

2015 04 13 - v15.2

This new minor release fixes some issues and adds two new configuration
directives:

  * ORA_INITIAL_COMMAND to be able to execute a custom command just after
    the connection to Oracle, for example to unlock a security policy.
  * INTERNAL_DATE_MAX to change the behavior of Ora2Pg with internal date
    found in user defined types.

This version will also automatically re-order exported views taking into
account interdependencies.

Here is the full list of changes:

    - Add INTERNAL_DATE_MAX configuration directive with default to 49 to be
      used when reformatting internal date returned with a user defined type
      and a timestamp column. DBD::Oracle only return the internal date format
      01-JAN-77 12.00.00.000000 AM so it is difficult to know if the year value
      must be added to 2000 or 1900. We takes the default behavior where date
      are between 1950 and 2049.
    - Remove extra CHAR and BYTE information from column type. Thanks to Magnus
      Hagander for the report.
    - Re-order views taking into account interdependencies. Thanks to Kuppusamy
      Ravindran and Ulrike for the suggestion and the Oracle query.
    - Fix case sensitivity in function based indexes. Thanks to Kuppusamy
      Ravindran for the report.
    - Fix PERFORM wrong replacement and infinite loop processing DECODE in some
      condition. Thanks to Didier Brugat for the report.
    - Fix replacement of boolean value in DEFAULT value at table creation.
      Thanks to baul87 for the report.
    - Add ORA_INITIAL_COMMAND configuration directive to be able to execute a
      custom command just after the connection to Oracle, to unlock a policy for
      example. Thanks to Didier BRUGAT for the feature request.
    - Fix alias in from clause when an XML type is found. Thanks to Lance Jacob
      for the record.
    - Invert condition on excluding temporary file with Windows OS. Thanks to
      kazam for the report.
    - Remove start time and global number of rows from _dump_table() parameters
      they are not used anymore.
    - Remove use of temporary file on Windows operating system.
    - Disable parallel table export when operating system is Windows.
    - Fix export of objects with case sensitivity using ALLOW or EXCLUDE
      directives. Thanks to Alexey Ignatov for the report.
    - Fix export of triggers from recycle bin.
    - Fix count of synonym in assessment report.
    - Add list of tables created by OEM to the exclusion list.
    - Fix look at default configuration file and set mode of export_schema.sh
      to executable by default. Thanks to Kuppusamy Ravindran for the report.
    - Add AUTHORIZATION to the list of PostgreSQL reserved word. Thanks to
      Kuppusamy Ravindran for the report.
    - Display a warning when an index has the same name as the table itself so
      that you can renamed it before export. Thanks to Kuppusamy Ravindran for
      the feature request.
    - Fix export of function based indexes with multiple column. Thanks to
      Kuppusamy Ravindran for the report.
    - Modify ora2pg script to return 0 on success, 1 on any fatal error and 2
      when a child process die is detected.
    - Change the way the generic configuration file is handle during project
      initialization. You can use -c option to copy your own into the project
      directory. If the file has the .dist extension, ora2pg will apply the
      generic configuration on it. Thanks to Kuppusamy Ravindran for the report
      and features request.
    - Add debug information when cloning the Oracle connection.
    - Force return of OLD when the trigger is on DELETE event

2015 02 06 - v15.1

New minor release just to fix two annoying bugs in previous release.

    - Fix replacement of function name which include SELECT in their name by
      PERFORM. Thanks to Frederic Bamiere for the report.
    - Fix creation of sources subdirectories when initializing a new migration project.

2015 02 04 - v15.0

This major release improve PL/SQL code replacement, fixes several bugs and
adds some new useful features:

    - Add support to the PostgreSQL external_file extension to mimic BFILE
      type from Oracle. See https://github.com/darold/external_file for
      more information.
    - Allow export of Oracle's DIRECTORY as external_file extension objects
      This will also try to export read/write privilege on those directories.
    - Allow export of Oracle's DATABASE LINK as Oracle foreign data wrapper
      server using oracle_fdw.
    - Allow function with PRAGMA AUTONOMOUS_TRANSACTION to be exported through
      a dblink wrapper to achieve the autonomous transaction.
    - Allow export of Oracle's SYNONYMS as views. Views can use foreign table
      to create "synonym" on object of a remote database.
    - Add trimming of data when DATA_TYPE is used to convert CHAR(n) Oracle
      column into varchar(n) or text. Default is to trim both side any space
      character. This behavior can be controlled using two new configuration
      directives TRIM_TYPE and TRIM_CHAR.
    - Add auto detection of geometry constraint type and dimensions through
      spatial index parameters. This avoid the overhead of sequential scan
      of the geometric column.
    - Add support to export Oracle sub partition and create sub partition
      for PostgreSQL with the corresponding trigger.
    - ALLOW and EXCLUDE directives are now able to apply filter on the object
      type. Backward compatibility can not be fully preserved, older definition
      will apply to current export type only, this could change your export in
      some conditions. See documentation update for more explanation.
    - Add PACKAGE_AS_SCHEMA directive to change default behavior that use a
      schema to emulate Oracle package function call. When disable, all calls
      to package_name.function_name() will be turn into package_name_function_name()
      just like a function call in current schema.
    - Add FKEY_OPTIONS to force foreign keys options. List of supported options
      are: ON DELETE|UPDATE CASCADE|RESTRICT|NO ACTION.
    - Add rewriting of internal functions in package body, those functions will
      be prefixed by the package name. Thanks to Dominique Legendre for the
      feature request.

Some change can break backward compatibility and make configuration directives
obsolete:

    - The ALLOW_PARTITION configuration directive has been removed. With new
      extended filters in ALLOW/EXCLUDE directive, this one is obsolete.
      Backward compatibility is preserved but may be removed in the future.
    - ALLOW and EXCLUDE directives do not works as previously. Backward
      compatibility may be preserved with some export type but may be broken
      in most of them. See documentation.
    - It is recommended now to leave the NLS_LANG and CLIENT_ENCODING commented
      to let Ora2Pg handle automatically the encoding. Those directives may be
      removed in the future.

Here is the full changelog of the release:

    - Declares SYNONYM views as SECURITY DEFINER to be able to grant access to
      objects in other schema.
    - Fix wrong replacement of data type in function body. Thanks to Dominique
      Legendre for the report.
    - Fix missing column name replacement on trigger export when REPLACE_COLS
      is defined. Thanks to Dominique Legendre for the report.
    - Fix missing table replacement on trigger export when REPLACE_TABLES is
      defined. Thanks to Dominique Legendre for the report.
    - Fix case where IS NULL substitution was not working. Thanks to Dominique
      Legendre for the report.
    - Remove double exclusion clause when multiple export type is used with same
      column name and no values defined.
    - Allow parsing of DATABASE LINK and SYNONYM from a DDL file.
    - Add DIRECTORY export type to export all Oracle directories as entries for
      the external_file extension. This will also export read/write privilege
      on those directories. Thanks to Dominique Legendre for the feature request.
    - Review documentation about NULL_EQUAL_EMPTY.
    - Fix missing code to replace IS NULL as coalesce(...). Thanks to Dominique
      Legendre for the report.
    - Add external_file schema to search_path when BFILE is set to EFILE in
      directive DATA_TYPE. Thanks to Dominique Legendre for the request.
    - Remove IF EXIST clause to oracle function created by Ora2Pg for BFILE
      export. Thanks to Dominique Legendre for the report.
    - Add support to the PostgreSQL external_file extension to mimic BFILE type
      from Oracle. See https://github.com/darold/external_file for more information.
    - Add auto detection of geometry constraint type and dimensions through the
      spatial index parameters first. This avoid the overhead of sequential scan
      of the geometric column.
    - Remove lookup at package function when not required.
    - Fix issue with database < 10g that do not have the DROPPED column into the
      ALL_TABLES view. Thanks to Lance Jacob for the report.
    - Add trimming of data when DATA_TYPE is used to convert CHAR(n) Oracle
      column into varchar(n) or text column into PostgreSQL. Default is to
      trim both side any whitespace character. This behavior can be controlled
      using the new configuration directives TRIM_TYPE and TRIM_CHAR.
    - Update copyright year.
    - Add assessment cost for object TABLE SUBPARTITION and review cost for
      object DATABASE LINK.
    - Update documentation about SYNONYM export.
    - Allow export of SYNONYMS as views with a new export type: SYNONYM.
    - Fix object exclusion function with Oracle 8i and 9i. Thanks to Lance Jacob
      for the report.
    - Fix INTERVAL YEAR TO MONTH and DAY TO SECOND with precision.
    - Remove unused pragma from the cost assessment.
    - Suppress PRAGMA RESTRICT_REFERENCES, PRAGMA SERIALLY_REUSABLE and INLINE
      from the PLSQL code. There is no equivalent and no use in plpgsql.
    - Fix several issues in function/procedure/package extraction from file
      input and some other related bug.
    - Remove single slash and \\r from function code.
    - Remove schema from package name with input file to avoid creating
      function with SCHEMA.PKGNAME.FCTNAME
    - Fix ALLOW/EXCLUDE ignored with type COPY or INSERT. Thanks to thleblond
      for the patch.
    - Fix setting of NLS_NUMERIC_CHARACTERS and NLS_TIMESTAMP_FORMAT with
      multiprocess, the session parameters was lost with the cloning of the
      database handle. Thanks to thleblond for the patch.
    - Fix issue that could produce errors "invalid byte sequence" when dumping
      data to pg database by forcing the client_encoding when PG_DSN is set.
      Thanks to thleblond for the patch.
    - Fix issue to add parenthesis with function with no parameters and wrong
      use of PERFORM in cursor declaration. Thanks to hdeadman for the report.
    - Fix broken export of function or procedure without parameter in package
      body. Thanks to hdeadman for the report.
    - Fix ERROR: "stack depth limit exceeded" generated by an infinite loop in
      partition trigger when there is no default table when value is out of range.
    - Add support to Oracle sub partition export.
    - Fix issue with procedure in package without parameters.
    - Enable DISABLE_SEQUENCE in generic configuration file.
    - Fix unwanted alter sequence in data export when there is table allowed
      or excluded.
    - Fix initial default values of command line parameter that prevent value
      in configuration file to be taken.
    - Fix non working global definition of table in ALLOW and EXCLUDE directive
      with COPY and INSERT export.
    - Update ora2pg.spec, thanks to bbuechler for the patch.
    - Close temporary files before deleting them, on Windows if they are not
      explicitly closed there are not deleted. Thanks to spritchard for the
      patch.
    - Force schema name to be uppercase when PRESERVE_CASE is disable (default).
      Thanks to Jim Longwill for the report.
    - Add rewriting of internal functions in package body, those functions will
      be prefixed by the package name. Thanks to Dominique Legendre for the
      feature request.
    - Fix type replacement in user defined type. Thanks to Dominique Legendre
      for the report.
    - Add filter with INSTEAD OF triggers on views to TRIGGER export type. Thanks
      to Dominique Legendre for the feature request.
    - Fix replacement of function name when PACKAGE_AS_SCHEMA is disabled.
    - Fix PLSQL_PGSQL that was always set to 0 when -p was not used even if
      configuration directive PLSQL_PGSQL was activated. Thanks to Dominique
      Legendre for the report.
    - Remove ALTER SCHEMA ... OWNER TO ... when CREATE_SCHEMA is not enable.
      Thanks to Dominique Legendre for the report.
    - Add DBLINK export to be created as foreign data wrapper server. Thanks to
      the BRGM for the feature request.
    - Remove ALLOW_PARTITION configuration directive, with extended filter in
      ALLOW/EXCLUDE directive, this one is obsolete. Backward compatibility is
      preserved.
    - Add documentation about extended filters in ALLOW and EXCLUDE directive.
    - Update documentation about VIEW_AS_TABLE and remove statement change with
      export TYPE is VIEW.
    - Add filter to grant export on functions, sequences, views, etc.
    - Fix GRANT in ALLOW or EXCLUDE filters.
    - Add commented order: "REVOKE ALL ON FUNCTION ... FROM PUBLIC;" when the
      function is declared as SECURITY DEFINER.
    - Prevent collecting column information with SHOW_TABLE export type.
    - Fix default value SYSTIMESTAMP to CURRENT_TIMESTAMP, and remove DEFAULT
      empty_blob(). Thanks to hdeadman for the report.
    - ALLOW and EXCLUDE directives are now able to apply filter on the object
      type. Backward compatibility can not be fully preserved, older definition
      will apply to current export type only, this could change your export in
      some conditions. See documentation update for more explanation. Thanks to
      the BRGM for the feature request.
    - Force function to be created with SECURITY DEFINER when AUTHID in table
      ALL_PROCEDURES is set to DEFINER in Oracle. This only works with Oracle
      >= 10g. Thanks to Dominique Legendre for the feature request.
    - Add PACKAGE_AS_SCHEMA configuration directive to change default behavior
      to use a schema to emulate Oracle package function call. When disable all
      call to package_name.function_name() will be turn into package_name_function_name()
      just like a function call in current schema. Thanks to the BRGM for the
      feature request.
    - Add a note to documentation about the way to convert srid into Oracle
      database instead of in Ora2Pg. Thanks to Dominique Legendre for the hint.
    - Fix documentation about SHOW_ENCODING export type.
    - Remove use of REGEX_LIKE with Oracle version 9. Thanks to Lance Jacob for
      the report.
    - Replace new FKEY_OPTIONS by FKEY_ADD_UPDATE configuration directive with
      three possible values: always, never and delete. It will force or not
      Ora2Pg to add "ON UPDATE CASCADE" on foreign keys declaration.
    - Allow FORCE_OWNER to work with all exported objects. Thanks to BRGM for
      the feature request.
    - Add FKEY_OPTIONS to force foreign keys options. List of supported options
      are: ON DELETE|UPDATE CASCADE|RESTRICT|NO ACTION. Thanks to the BRGM for
      the feature request.
    - Fix ambiguous column in view extraction. Thanks to Dominique Legendre for
      the report.
    - Fix replacement of TYPE:LEN by boolean, ex: REPLACE_AS_BOOLEAN CHAR:1.
      Thanks to jwiechmann for the report.
    - Fix error ORA-00942 where Ora2Pg try to export data from a view defined
      in VIEW_AS_TABLE configuration directive.
    - Update list of excluded Oracle schema to the documentation.
    - Fix export of all views with comments when VIEW_AS_TABLE is set.
    - Fixed some typos in the generated sample configuration file. Thanks to
      Hal Deadman for the patch.
    - Limit column information export to the type of object extracted.
    - Remove call to MDSYS in SQL code. Thanks to Dominique Legendre for the
      report.
    - Add more Oracle schema to the exclusion list.
    - Fully remove join on DBA_SEGMENTS to retrieve the list of tables, views
      and comments. Replaced by ALL_OBJECTS. Thanks to Dominique Legendre for
      the help.
    - Exclude JAVA\$.* tables and fix tables list query to include newly created
      tables with no segments. Thanks to Dominique Legendre for the fix.
    - Fix regex that convert all x = NULL clauses to x IS NULL to not replace
      := NULL too.
    - Autodetect unusual characters in owner name when extracting data and used
      it embeded into double quote.
    - Replace single return with return new in trigger code. Thanks to Dominique
      Legendre for the report.

2014 11 12 - v14.1

This is a maintenance release only mainly to add patches that was not
been applied in previous major release.

    - Remove ALLOW_CODE_BREAK, it is no more useful.
    - Change output of SHOW_ENCODING to reflect change to default encoding.
    - Comment ALLOW_PARTITION in default configuration file
    - Add QUERY and KETTLE export type in configuration file comments.

2014 11 05 - v14.0

This major release adds full export of Oracle Locator or Spatial geometries into
PostGis, SDO_GEOM functions and SDO_OPERATOR are also translated. This export
adds the following features:

  1. Basic and complex geometry types support
  2. Geometry data conversion from Oracle to PostGIS
  3. Spatial Index conversion
  4. Geometry metadata / constraints support
  5. Spatial functions conversion

For spatial data export, you have three choice, WKT to export data using
SDO_UTIL.TO_WKTGEOMETRY(), WKB to export data using SDO_UTIL.TO_WKBGEOMETRY()
and INTERNAL to export geometry using a Pure Perl library. Unlike the first
two methods, INTERNAL is fast and do not raise Out Of Memory. The export is
done in WKT format so that you can verify your geometry before importing to
PostgreSQL.

Other additional major features are:

  - Parallel table processing.
  - Auto generation of migration template with a complete project tree.
  - Allow user defined queries to extract data from Oracle.

Parallel table processing is controlled by the -P or --parallel command line
options or the PARALLEL_TABLE configuration directive to set the number of
tables that will be processed in parallel for data extraction. The limit is
the number of cores on your machine. Ora2Pg will the open one connection to
Oracle database for each parallel table extraction. This directive, when upper
than 1, will invalidate ORACLE_COPIES but not JOBS, so the real number of
process that will be used is (PARALLEL_TABLES * JOBS).

The two options --project_base and --init_project when used indicate to Ora2Pg
to create a project template with a work tree, a generic configuration file
and a shell script to export all objects from the Oracle database. So that you
just have to define the Oracle database connection into the configuration file
and then execute the shell script called export_schema.sh to export your
Oracle database into files. Here a sample of the command and the project's tree.

    ora2pg --project_base /tmp --init_project test_project

    /tmp/test_project/
	config/
		ora2pg.conf
	data/
	export_schema.sh
	reports/
	schema/
		fdws/  functions/  grants/  kettles/
		mviews/  packages/  partitions/
		procedures/  sequences/  tables/
		tablespaces/  triggers/  types/  views/
	sources/
		functions/  mviews/  packages/
		partitions/  procedures/  triggers/
		types/  views/

It create a generic config file where you just have to define the Oracle
database connection and a shell script called export_schema.sh. The
sources/ directory will contains the Oracle code, the schema/ will
contains the code ported to PostgreSQL. The reports/ directory will
contains the html reports with the migration cost assessment.

Sometime you may want to extract data from an Oracle table but you need a
custom query for that. Not just a "SELECT * FROM table" like Ora2Pg do but
a more complex query. The new directive REPLACE_QUERY allow you to overwrite
the query used by Ora2Pg to extract data. The format is TABLENAME[SQL_QUERY].
If you have multiple table to extract by replacing the Ora2Pg query, you can
define multiple REPLACE_QUERY lines. For example:

    REPLACE_QUERY   EMPLOYEES[SELECT e.id,e.fisrtname,lastname FROM EMPLOYEES e
	      JOIN EMP_UPDT u ON (e.id=u.id AND u.cdate>'2014-08-01 00:00:00')]

Other new features are:

    - Export of declaration of language C function. Previous version was
      not exporting function with no code body like external C function.
    - Export of COMMENT from views.
    - Function to replace some call to SYS_CONTECT(USERENV, ...) by the
      PostgreSQL equivalent.
    - Add POSTGIS_SCHEMA configuration directive to add the dedicated
      PostGis schema into the search_path.
    - Add PG_SUPPORTS_IFEXISTS configuration directive to be able to suppress
      IF EXISTS call in DDL statement generated by Ora2Pg.
    - Triggers are now all excluded/allowed following the table names specified
      in the ALLOW and EXCLUDED directives
    - Allow automatic export of nested tables (TYPE+TABLE+COPY).

One change is not fully backward compatible: Ora2Pg now use UTF8 by default
on both side. On Oracle connection NLS_LANG is set to AMERICAN_AMERICA.AL32UTF8,
NLS_NCHAR to AL32UTF8. On PostgreSQL side CLIENT_ENCODING to UTF8. For export
that dump to files, Perl binmode is set to utf8. You can always change those
default setting in configuration file, but it is not recommanded.

Here is the full changelog of the release:

    - Fix inline comments into function declaration. Thanks to Marcel Huber
      for the report.
    - Fix case where SELECT ... INTO was wrongly replaced by PERFORM.
    - Fix DECODE() translation. Thanks to Dominique Legendre for the report.
    - Add replacement of SDO_OPERATOR into PostGis relationships.
    - Add replacement of SDO_GEOM spatial function to postgis ST_* functions.
    - Add GEOMETRY_EXTRACT_TYPE configuration directive to specify the geometry
      extracting mode: WKT (default), WKB and INTERNAL.
    - Add a pure Perl library to export SDO_GEOMETRY as a WKT representation.
      This is controlled by a new extraction type INTERNAL to use with the
      GEOMETRY_EXTRACT_TYPE configuration directive.
    - Remove USE_SC40_PACKAGE directive and any reference to this library,
      it is not useful now that we have the INTERNAL geometry extraction mode.
    - Fix replacement of varchar2 in PL/SQL function.
    - Fix bug in type replacement when default values used function.
    - Add export of declaration of language C function. Previous version was
      not exporting function with no code body like external function.
    - Fix create statement in export of view as table. Thanks to ntlis for the
      report.
    - Fix replacement of to_number without format.
    - Add export of COMMENT from VIEWS.
    - Add function to replace some call to SYS_CONTECT(USERENV, ...) by the
      PostgreSQL equivalent.
    - Fix parsing from file of tablespace.
    - Fix wrong alias name in FROM clause when extracting XML data. Thanks
      to Marc Sitges for the report.
    - Fix export of comments in FDW export, might be COMMENT ON FOREIGN TABLE.
      Thanks to David Fetter for the report.
    - Fix broken export of function based indexes. Thanks to Floyd Brown for
      the report.
    - Fix sequence with negative minvalue/maxvalue and negative increment.
      Thanks to jwiechmann for the report.
    - Fix forced owner to schema to the value of FORCE_OWNER when it is set
      to a user name.
    - Fix create schema when FORCE_OWNER is enabled. Thanks to Dominique
      Legendre for the report.
    - Add POSTGIS_SCHEMA configuration directive to add a schema to the
      search_path. Thanks to Dominique Legendre for the feature request.
    - Returns NULL when a geometry is NULL instead of calling ST_AsText with
      a null value. Thanks to Dominique Legendre for the report.
    - Add more explanation about values of CONVERT_SID.
    - Fix issue in DBMS_OUTPUT replacement.
    - Fix exclusion of default objects from type export.
    - When CONVERT_SRID is > 1 this value will be used to force the SRID value
      on all export.
    - Disable NULL_EQUAL_EMPTY in generic configuration when generating a project
      tree.
    - Add LOGMNR$ and RECAP$ in the exclusion objects list.
    - Fix performance issue in extracting data from geometry column and add
      AUDSYS,DVSYS and DVF to the list of schema to exclude.
    - Prefix table name with schema name on queries for retrieving data to
      avoid errors in multi schema export.
    - Add SDO_* cost to migration report.
    - Fix real number of Synonym that should be review.
    - Fix wrong report of CTXSYS synonym.
    - Enabled AUTODETECT_SPATIAL_TYPE by default.
    - Remove KETTLE and FDW export from the auto generated project.
    - Force the copy of /etc/ora2pg/ora2pg.conf.dist into the project directory
      with no more look at the current ora2pg.conf. Force autodetection of
      spatial type in the generic configuration.
    - Huge performance gain on querying information about Spatial column. Thanks
      to Dominique Legendre for the great help.
    - Fix wrong use of table alias with SEGMENT_NAME.
    - Add unified audit table (CLI_SWP$.*) from the exclusion list.
    - Fix operator in check condition of range partitions. Thanks to Kaissa
      Chellouche for the report.
    - Add to the internal exclusion list tables generated by spatial indexes
      MDRT_.*, sequences MDRS_.* and interMedia Text index DR$.*. Thanks to
      Dominique Legendre for the report.
    - Make REPLACE_TABLES and REPLACE_COLS work with VIEW. The view name and
      the columns aliases will be replaced. Take care that the table name or
      columns names in the statement will be kept untouched and need manual
      rewriting. Thanks to Sven Medin for the feature request.
    - Add PG_SUPPORTS_IFEXISTS configuration directive to be able to suppress
      IF EXISTS call in DDL statement generated by Ora2Pg. PostgreSQL below
      9.x do not support this keywords. Thanks to George Kowalski fot the
      feature request.
    - Fix wrong substitution in EXECUTE ... USING statement, where parameters
      number was not prefixed by a $ sign. Thanks to Dominique Legendre for
      the report.
    - Fix document about KEEP_PKEY_NAMES that also affect unique key and not
      only primary key as it was specified in the documentation. Thanks to
      Dominique Legendre for the report.
    - Add tables generated by statistics on spatial index (MDXT_.*) into the
      internal exclusion list. This join the already excluded table generated
      by partition logging (USLOG$_.*) and materialized view logs (MLOG$_.*,
      RUPD$_.*)
    - Add DEFAULT_SRID configuration direction to permit change of the internal
      default EPSG srid 4326.
    - Fix new line after search_path settings. Thanks to Dominique Legendre for
      the report.
    - Triggers are now all excluded/allowed following the table names specified
      in the ALLOW and EXCLUDED directive, no more on there own name which had
      little interest. Thanks to Dominique Legendre for the feature request.
    - Add support to COPY export with Spatial objects. Thanks to Legendre
      Dominique for the great help to solve this problem.
    - Fix default SRID value when a NULL value is returned by Oracle, SRID 8307
      and the corresponding EPSG SRID 4326.
    - Update documentation on relation between PARALLEL_TABLES and FILE_PER_TABLE
    - Add the -P or --parallel command line options and update documentation
      about parallel table processing.
    - Add PARALLEL_TABLES configuration directive to force ora2Pg to use on
      process and one connection per table up to the number of CPU specified.
      Thanks to menardorama for the feature request.
    - Add PARALLEL_TABLES configuration directive to force ora2Pg to use on
      process and one connection per table up to the number of CPU specified.
      Thanks to menardorama for the feature request.
    - Add --init_project and --project_base command line options to create a
      migration template with a complete project tree, a generic configuration
      file and script to automate export of all object in the project tree.
    - Fix unwanted space before AND returned by limit_to_tables(). Thanks to
      Alex Wang for the report.
    - Add note about regex inclusion/exclusion not working with 8i database in
      documentation
    - Fix regex inclusion/exlusion of table that was not more working since the
      inclusion of limit_to_tables() function. Thanks to alex wang for the patch
    - Exclude dropped tables (those who are in the recycle bin) from export.
    - When USER_GRANTS is disabled, aka login as dba user, force table list to
      be checked against DBA_SEGMENTS with SEGMENT_TYPE of type table or table
      partition. This could help solving some incomprehensible object found in
      Oracle view ALL_TABLES.
    - Fix query to retrieved list of tables, owner selection was set two time.
    - Add support to automatic nested table export (TYPE+TABLE+COPY).
    - Fix wrong export of materialized view log table. Thanks to Ronson Blossom
      for the report.
    - Update the SYSUSER array to exclude objects owned par those more users.
    - Fix unwanted export of overflow table of an index-organized table. Thanks
      to Ronson Blossom for the report.
    - Update the SYSUSER array to exclude objects owned par those users.
    - Display table owner in debug mode for SHOW_TABLE or SHOW_COLUMN.
    - Add a section to give hint about converting Oracle outer join syntax to
      ANSI. Thanks to Sven Medin for the links.
    - Fix issue #82 again. Thanks to Sven Medin fro the report.
    - Add first support to user defined queries to extract data from Oracle.
      This feature add a new configuration directive named REPLACE_QUERY.
    - Change program title when dump to file.
    - Fix MODIFY_TYPE directive that was broken when using type with space
      character. Thanks to Dmitry K. for the patch.
    - Show missing view name in debug mode when exporting some views as table.
    - Rewrite replace(a,b) with three arguments replace(a,b,'') for PostgreSQL.
      Thanks to Dominique Legendre for the report.
    - Convert all x <> NULL or x != NULL clauses to x IS NOT NULL. All x = NULL
      are converted into x IS NULL. Thanks to Dominique Legendre for the report.
    - Add warning at exit to signal when a OOM occurs. In that case, when a child
      Ora2Pg process was silently killed by the OOM killer there was no information
      that a failure occurs.

2014 06 02 - v13.0

This major release adds first support to export Oracle Spatial Objects to PostGis
Spatial objects. There's also a new configuration directive to allow logging of
statement failures to prevent Ora2Pg to abort and continue to load valid data. 
The other main feature is the possibility to convert DDL files without needing an
Oracle database connection, until now this was reserved to files containing stored
procedures. There's also several bug fixes.

	- Allow error logging during data import. This feature controlled by the
	  LOG_ON_ERROR directive allow you to not abort the data import process
	  when an error is encountered and to log to a file the COPY or INSERT
	  statement that generate the error. After fixing the statement you will
	  be able to load the missing data. Thanks to menardoram for the feature
	  request.
	- Force export type to be INSERT when COPY is used and a table have a
	  GEOMETRY column. I can not find a solution to export as copy statement
	  for the moment. Thanks to Dominique Legendre and Vincent Picavet for
	  the help.
	- Fix export of user defined type as object. Thanks to Shanshan Wang for
	  the report.
	- Limit look up of objects to the ALLOW or EXCLUDE filter into the SQL
	  query instead of the Perl code to avoid retrieving huge list of objects
	  on such database. Thanks to menardorama for the feature request.
	- Add support to spatial data export in INSERT mode. Still need some work
	  in COPY export mode if possible.
	- Fix query to retrieve SRID that broken with patch on CONVERT_SRID.
	- Fix wrong filter with ALLOW directive when getting list of  partition.
	- Add GRANT export read from an input file.
	- Fix data type conversion when using input file and data type such
	  varchar2(10 BYTE).
	- Add export of comment with TABLE and VIEW exports using an input file.
	- Add extraction of TABLESPACE from an input file.
	- Add support to SEQUENCE extraction from input file.
	- Fix wrong filter with ALLOW directive when exporting partition. The
	  filter was done on partition name instead of table name, that mean
	  that setting ALLOW directive was resulting in no export at all. Thanks
	  to menardorama for the report.
	- Add CONVERT_SRID configuration directive to control the automatic
	  conversion of SRID to standard EPSG using the Oracle SDO function
	  sdo_cs.map_oracle_srid_to_epsg() Oracle function. Thanks to Dominique
	  Legendre for the help.
	- Fix a typo in the create index prefix on partitioned tables. Thanks
	  to menardorama for the patch.
	- Fix non replacement of destination during SHOW_COLUMN and COPY export.
	  Using MODIFY_TYPE was only working in TABLE export.
	- Force pl/sql conversion with TABLE export to replace advanced default
	  values. Fix code TRUNC(SYSDATE, MONTH) in default value and everywhere
	  that should be: date_trunc(month,LOCALTIMESTAMP). Thanks to menardorama
	  for the report.
	- Fix code regarding unique partition index naming. Thanks to menardorama
	  for the report.
	- Add PREFIX_PARTITION configuration directive. When enabled it will force
	  renaming all partition table name with the name of the parent table.
	  Thanks to menardoram for the feature request.
	- Add AUTODETECT_SPATIAL_TYPE in configuration file and documentation
	  about this new directive.
	- Add export of SDO_GEOMETRY column type. They are basically exported to
	  the non-constrained "geometry" type with SRID if defined. When the
	  configuration directive AUTODETECT_SPATIAL_TYPE is enable, Ora2Pg will
	  try to autodetect the geometry type, the dimension and the SRID used
	  to set a constrained geometry type. For example, in the first case
	  column shape with Oracle type SDO_GEOMETRY will be converted as:

		shape geometry(GEOMETRY) or shape geometry(GEOMETRY, 4326)

	  and in the second case, with constrained geometry type:

		shape geometry(POLIGONZ, 4326)

	  with a three dimensional polygon. Thanks to Vincent Picavet for the
	  feature request and specification.
	- Add support to spatial index read from file.
	- Add export of Oracle spatial index. For example, index:
		CREATE INDEX cola_spatial_idx ON cola_markets(shape) INDEXTYPE IS MDSYS.SPATIAL_INDEX;
	  will be exported as
		CREATE INDEX cola_spatial_idx ON cola_markets USING GIST(shape);
	  Thanks to Vincent Picavet / Oslandia for the feature request and explanations.
	- Allow TRIGGER export to parse an input file with Oracle DML orders.
	- Add PG_SUPPORTS_CHECKOPTION configuration directive to not remove
	  WITH CHECK OPTION in create view statement. It is supported in 
	  PostgreSQL 9.4.
	- Allow VIEW export to parse an input file with Oracle DML orders.
	- Allow TABLE export to parse an input file with Oracle DML orders.
	- Add SYNCHRONOUS_COMMIT configuration directive disabled by default.
	  This is the current behavior of Ora2Pg, it set synchronous_commit
	  to off before data import to PostgreSQL. This is only used when you
	  load data directly to PostgreSQL, the default is off to disable
	  synchronous commit to gain speed at writing data. Some modified or
	  old version of PostgreSQL, like Greenplum, do not have this setting.
	- Add some useful information for Windows user in documentation. Thanks
	  to Roger Park for the report.
	- Fix case when parentheses are omitted in index creation. Thanks to
	  Yuri Ushakov for the report.
	- Fix export type PACKAGE when ALLOW is defined to extract only some
	  packages. Thanks to Maciej Bak for the report.
	- Fix INSERT export where backslash should be escaped and single be
	  doubled in standard conforming string notation. Thanks to Yuri
	  Ushakov for the report.
	- Add important note about LONGREADLEN and DATA_LIMIT that could need
	  to be adjusted to avoid out of memory. Thanks to Mike Kienenberger
	  for the patch.
	- Fix case sensitivity issue with export of comment on column. Thanks
	  to Pierre Crumeyrolle for the report.
	- Fix export of RAW data in COPY mode, was missing a backslash. Thanks
	  to jwiechmann for the report.
	- Fix RAW data export in COPY and INSERT mode, RAW data type is returned
	  in hex by DBD::Oracle. Thanks to jwiechmann for the report.
	- Fix one release 8i condition.
	- Fix inexistent column USE_NO_INDEX with Oracle 8i and MVIEW export.
	- Enclose call to utf8::encode and utf8::valid into eval.
	- Fix export of constraint with Oracle 8i release.
	- Fix unrecognized fatal error with 8i database. Thanks to UnvorherSeba
	  for the patch.
	- Revert change level of error from fatal to error, when querying
	  materialized view.
	- Change level of error from fatal to error, when querying materialized
	  view.

2014 01 28 - v12.1

This is a maintenance release with some minor bug fixes and a new configuration
directive, INDEXES_SUFFIX, to allow appending a suffix to indexes names.

	- Fix example given for the WHERE configuration directive. Thanks to
	  Bob Treumann for the report.
	- Add INDEXES_SUFFIX configuration option to allow append a suffix to
	  indexes names.
	- Replace special charater ^M by \r as they are not supported by git.
	- Fix IF EXISTS in alter table of sub _drop_foreign_keys. Thanks to
	  Francis Corriveau for the patch.
	- Fix isolation level when exporting data. Thanks to Ludovic Penet for
	  the report.
	- Fix regression when ora2pg tries to create foreign keys on tables or
	  to tables that are not selected for export. Thanks to Ludovic Penet.
	- Add information about backslashed comma into directive MODIFY_TYPE
	  into Makefile.PL.
	- Add missing MODIFY_TYPE definition in documentation.
	- Allow backslashed comma into MODIFY_TYPE type redefinition. Example:
		TABLE1:COL3:decimal(9\,6),TABLE1:COL4:decimal(9\,6).
	  Thanks to Mike Kienenberger for the report
	- Fix missing single cote into create_materialized_view() call. Thanks
	  to Jacky Rigoreau for the patch.
	- Fix some typo in documentation, thanks to Mike Kienenberger for the
	  report.
	- Add a chapter about installing DBD::Oracle into documentation. Thanks
	  to Raghavendra for the patch.
	- Fix case sensitivity on external table name with FDW export type.
	  Thanks to Guillaume Lelarge for the report.
	- Fix export of materialized views when PG_SUPPORTS_MVIEW is disabled.
	  Thanks to Christian Bjornbak for the report.
	- Update copyright.

2013 10 22 - v12.0

This release fixes lot of issues and three new features. Using REORDERING_COLUMNS
directive you will be able to reorder columns to minimized the footprint on disc,
so that more rows fit on a data page. The PG_SUPPORTS_MVIEW will allow you to
export materialized with native PostgreSQL 9.3 syntaxe. The USE_TABLESPACE variable
will allow you to export object using their original tablespace.

	- Skip constraints on system internal columns (sys_nc...$) from export.
	- Fix missing output directory in generic psql file for data loading.
	- Add missing progress bar during TYPE and PARTITION export type.
	- Remove duplicated message in debug mode during Oracle reconnection.
	- Allow file input with create type declaration to use ora2pg converter.
	  Unsupported syntax is signaled into the output file.
	- Exclude MLOG$.* and RUPD$.* table from export.
	- Prevent export of indexes and constraints during FDW export type.
	- Fix wrong total number of sequences shown in progress bar.
	- Remove warning when PG_DSN is define during a export type that do not
	  support direct import into PostgreSQL.
	- Auto switch prefix from DBA to ALL when error 942 is returned when
	  looking at tables informations. A hint is also displayed to ask for
	  activating USER_GRANTS or connect using a user with DBA privilege.
	- Add REORDERING_COLUMNS configuration directive to allow reordering
	  columns during the TABLE export. This could help to minimized the
	  footprint on disc, so that more rows fit on a data page. Thanks to
	  Christian Bjornbak for the feature request.
	- Fix call to unblessed reference at disconnect when direct import to
	  pg is not used. Thanks to Christian Bjornbak for the report.
	- Fix regression in drop/create foreign keys and index during data
	  export. Thanks to Christian Bjornbak for the report.
	- Fix truncate table error with parallel and direct data copy. Thanks
	  to keymaper for the report.
	- Fix several other issues with parallel and direct data import.
	- Fix trigger export on multi files when FILE_PER_FUNCTION is enabled.
	- Fix issue on converting boolean values with non default values.
	  Thanks to Christian Bjornbak for the report.
	- Fix boolean value for disabled key in default %BOOLEAN_MAP key/value.
	- Fix case where INTO was wrongly replaced by INTO STRICT. Thanks to
	  Jacky Rigoreau for the report.
	- Fix case where label after a END was not removed. Thanks to Jacky
	  Rigoreau for the report.
	- Fix discard of input file parsing. Fix PERFORM replacement in PL/SQL
	  code wirh cursor. Thanks to Jacky Rigoreau for the report.
	- Enable PG_SUPPORTS_MVIEW by default and update documentation.
	- Replace DBA_DATA_FILES by USER_SEGMENTS to get database size to avoid
	  error ORA-00942. Thanks to Pierre Boizot for the report.
	- Fix trigger conversion error. Thanks to Pierre Boizot for the report.
	- Add support to PostgreSQL 9.3 materialized view syntaxe, this need a
	  new configuration directive PG_SUPPORTS_MVIEW to be enabled.
	- Update default configuration file and documentation about USE_TABLESPACE.
	- Add USE_TABLESPACE configuration directive to force ora2pg to use Oracle
	  tablespace name with table, constraints indexes and indexes if tablespace
	  in not in the default (TEMP, USERS, SYSTEM). Thanks to Rob Moolhuijsen
	  for the feature request.
	- Allow DEFER_FKEY, when enabled during TABLE export, to create all foreign
	  keys as DEFERRABLE and INITIALLY DEFERRED. Thanks to David Greco for the patch.
	- Fix non working ON_ERROR_STOP set to 0 during data export.
	- Lot of code changes to fix dump to file in multiprocess mode. Ora2Pg will
	  also only drop/create constraints and indexes related to the allow/exclude
	  tables, thanks to Maciej Bak for the report.
	- Force decimal character from Oracle output to be a dot. Thanks to Maciej Bak
	  for the report.
	- Add default exclusion of Oracle recycle bin objects with name begining by BIN$.
	- Fix escaping quote in table and column comments. Thanks to realyota for the report.
	- Reduce DECODE migration cost from 2 to 1 unit.
	- Reduce OUTER JOIN (+) migration cost from 3 to 1 unit.
	- Add Time::HiRes to the requirement chapter for Perl <= 5.8. Thanks to
	  Mike Kienenberger for the report.
	- Replace wrong use of --config instead of --conf into the documentation. Thanks
	  to Mike Kienenberger for the report.
	- Fix regex used to rewrite CREATE VIEW code. Thanks to David Greco for
	  the patch.
	- Fix an issue with oracle copies when primary key was negative. Thanks
	  to David Greco for the patch.
	- Fix case sensitivity with SEQUENCE when preserve_case is enabled.
	  Thanks to Jean-Max Reymond for the report.
	- Fix table COMMENT export when preserve_case is enabled. Thanks to
	  Jean-Max Reymond for the report.

2013 05 28 - v11.4

This release fixes others several major issues on migration cost assessment that
was not addressed in previous release, please upgrade.

	- Fix other major issues in migration cost assessment.
	- Redefine some migration cost values to be more precise.

2013 05 27 - v11.3

This release fixes several major issues on migration cost assessment, especialy
with stored procedures with lot of lines or if you have lot of comments in that
code. You may want to run your database evaluation again as the estimated times
can be up to tree time lower on huge PL/SQL code. 

	- Add full details about PL/SQL evaluation by ora2pg when --estimate_cost
	  or ESTIMATE_COST is enable. This will display cost units per keywords
	  detected in the function/package code.
	- Fix wrong cost unit assessment on PL/SQL code size, this bug generated
	  very high migration cost assessment for functions/packages with lot of
	  lines. Please run your tests again, estimated times can be up to tree
	  time lower on huge code.
	- Remove comments before code evalution.
	- Fix file input parser for PL/SQL packages export when IS or AS was in
	  the next line than the CREATE PACKAGE BODY ...
	- Exclude NOT NULL constraint from the count of CHECK constraints into
	  the TABLE report.
	- Fix decimal precision in table migration assessment cost.
	- Fix typo in changelog.

2013 05 01 - v11.2

This release fixes several major issues especially with direct import of data
into PostgreSQL and Windows port that was both broken.

	- Update doc about Windows multiprocess issues and acknowledgements.
	- Fix Windows OS issues using multiprocessing options by disabling
	  multiprocess support on this plateform. When -J or -j will be used a
	  warning will be displayed and Ora2Pg will simply run single process
	  like in previous 10.x versions. Thanks to Jean Marc Yao Adingra for
	  the report.
	- Fix RAW and LONG RAW export to ByteA. Thanks to Prabhat Tripathi for
	  the report and testing.
	- Fix patch regression on multiple TRUNCATE call for a single table.
	  Thanks to David Greco for the report.
	- Placed calls to DB handle InactiveDestroy outside the forked process
	  to prevent fatal errors on Windows. Thanks to Jean Marc Adingra for
	  the report.
	- Forked running processes are renamed into more readable name like
	  "ora2pg logger" for the progress bar, "ora2pg - querying Oracle" when
	  used with -J option and "ora2pg - sending to PostgreSQL" to better
	  know  what is the current job of the process.
	- Removed the use of /Y flag in Windows install script, this was causing
	  error "dmake:  Error code 130, while making install_all". Thanks to
	  Jean-Marc Adingra for the report.
	- Fix direct import to PostgreSQL that was just producing nothing. Thank
	  to David Greco for the patch.
	- Fix ora2pg usage documentation.
	- Add an underscore to CLIENT ENCODING in SHOW_ENCODING output to be the
	  same as the configuration directive.

UPGRADE: please reinstall all as most of the files have changed.

2013 04 07 - v11.1

This release adds partition data speed improvement by exporting data directly
from and into the destination partitioned table. There's also some bug fix on
RAW or LONG RAW data export and PL/SQL to PL/PGSQL code rewrite.

	- Adjust cost assessment for indexes, tables and tables partition.
	- Add comment to report of index partition about local index only.
	- Fix position of TRUNCATE TABLE in output file.
	- Fix export of data from RAW or LONG RAW columns, they was exported
	  as hex string. Now data are converted using utl_raw.cast_to_varchar2()
	  function before being escaped for insert into a bytea. Thanks to Alex
	  Delianis for the report.
	- Fix issue with Oracle TIMESTAMP(0) data export that add a single
	  ending point, ex: "2008-08-09 00:00:00.", this ending character is
	  now removed by format_data_type(). Thanks to Pierre-Marie Petit for
	  the report.
	- Fix typo on MODIFY_STRUCT description.
	- Force DEBUG to off in default configuration file.
	- Change range PARTITION operators in the check conditions, >= and <
	  replaced by > and <=, corresponding to Oracle VALUES LESS THAN.
	- Add ALLOW_PARTITION to limit data export to a list of partition name.
	- PLSQL: Fix wrong replacement of SELECT by PERFORM during VIEW export.
	- Partitioned tables data is now imported directly into the destination
	  tables instead of inserted into the main table and dispatched by the
	  trigger. Ora2Pg will automatically detect the in/out table partition,
	  there's nothing to configure.
	- PL/SQL: Do not allow decode() rewrite by case/when/else when there
	  is a function call in it.
	- Fix Error when Compress::Zlib is not installed, this module is not
	  mandatory.

UPGRADE: please reinstall all as all files have changed.

2013 03 24 - v11.0

This is a new major release because it adds support to multiprocessing to export
data in parallel mode, this allow to improve speed during data import by more
than ten times. This multiprocessiing capabilities allow Ora2Pg to be closer than
the speed of any ETL. To compare speed or allow using Kettle for data import,
there's now a new export type to obtain Kettle XML transformation files. This
release adds also lot of work on speed improvement to scan Oracle database with
huge number of object.

	- Add documentation about JOBS, ORACLE_COPIES, DEFINED_PK configuration
	  directive and informations about KETTLE export type.
	- Add KETTLE export type to generate XML transformation file definition
	  for Penthatlo Data Integrator (Kettle). Thanks to Marc Cousin for the
	  work. Example of use:
		ora2pg -c ora2pg.conf -t KETTLE -j 12 -J 4 -o loaddata.sh
	- Fix major bug in export of auto generated named constraint. Thanks to
	  mrojasaquino fot the report.
	- Show number of rows in the top largest tables.
	- Add TOP_MAX description to the documentation.
	- Add the TOP_MAX directive to default configuration file and update
	  documentation. Directive used to control the top N tables to show.
	- Add top N of largest tables in SHOW_TABLE, SHOW_COLUMN and SHOW_REPORT
	  export type.
	- Fix progressbar output when ora2pg is interrupted by ctrl+c.
	- Add JOBS, ORACLE_COPIES and DEFINED_PK directives to configuration file.
	  JOBS replacing THREAD_COUNT but backward compatibility is preserve.
	- Add 3 new command line options, -j | --jobs and -J | --copies, used to
	  set the number of connection to PostgreSQL and Oracle for parallel
	  processing. The third, -L | --limit is used to change DATA_LIMIT at
	  command line.
	- Add multiprocess support on data export. With the help of Thomas Ogrisegg.
	- Add more schema in SYSUSERS that should not be exported.
	- Add full detailed information about SYNONYM in SHOW_REPORT.
	- Add MODIFY_TYPE configuration directive to allow some table/column
	  type to be changed on PostgreSQL side during the export.
	- Fix objects type count in progressbar of SHOW_REPORT.
	- Restrict table and index in SHOW_REPORT to the tables defined in ALLOW
	  and EXCLUDE directives.
	- Show total number of rows in SHOW_TABLE and SHOW_REPORT output.
	- Add top 10 of tables sorted by number of rows in SHOW_TABLE and
	  SHOW_REPORT output.
	- Fix typo in SYNONYM objects.
	- Add report of top ten tables ordered y number of rows.
	- Rewrite most of the Oracle schema storage information extraction for
	  speed improvement.
	- Use Hash to store column informations.
	- Fix %unique_keys declaration in _table() method.
	- Remove call to _table_info() from SHOW_REPORT code as those informations
	  are already loaded with the _table() method.
	- Fix missing column definition on TABLE export.
	- Add progress bar during output generation following export type.
	- Add STOP_ON_ERROR configuration directive to enable/disable the call to
	  ON_ERROR_STOP into generated SQL scripts. Thanks to Ludovic Penet for
	  the feature request.
	- Huge speed improvement on columns informations retrieving.
	- Fix progress bar to keep the total number of tables related to the ALLOW
	  or EXCLUDE configuration directives. Thanks to Ludovic Penet for the report.
	- Change return type of function _table_info(), it now returns data instead
	  of the database handle.
	- Improve speed on indexes and constraints extraction for database with huge
	  number of tables.
	- Improve performance to retrieve columns information and comments.
	- Remove report of column details during export in debug mode, use SHOW_COLUMN
	  instead.
	- Remove call to upper() in objects owner condition to improve performance
	  with database with huge number of objects.
	- Add a fix to not export foreign key for exclude tables. Thanks to Ludovic
	  Penet for the report.
	- Fix Windows install issue with copying ora2pg.conf.dist. Thanks to
	  Dominique Fourdrinoy for the report.
	- Increase the cost of Oracle function not converted to PG automatically.

UPGRADE: reinstall all is required to override the old installation, you may use the
new ora2pg.conf.dist file which included the new configuration directives. 

2013 01 15 - v10.1

This release adds HTML report for migration cost assessment and some bug fix.

	- Fix global where should not be overwritten. Thanks to Dan Harbin for
	  the patch.
	- Fix bug/typo in boolean replacement, where a colon was used instead
	  of a single quote. Thanks to Alex Delianis for the patch.
	- Update copyright.
	- Add detection of additional Oracle functions for better migration
	  cost assessment.
	- Update documentation.
	- Force report detail in lowercase.
	- Added information about the migration cost value to the reports.
	- Add --dump_as_html command line option and DUMP_AS_HTML configuration
	  directive.
	- Allow migration report to be generated as HTML.
	- Separate report generation code from data collection code.

2012 12 12 - v10.0

This is the first version of Ora2Pg 10.x series, that is a major release.
Overall numerous improvements and bugs fixes there's now a new export type:
SHOW_REPORT that will output a report of all objects contained in your Oracle
database and some comments on how they will be exported. With this report you
can use a new directive ESTIMATE_COST to ask to Ora2Pg to evaluate the database
migration cost in terms of man days. There's also an other new configuration
directive EXTERNAL_TO_FDW, disable by default, to permit the export of all
Oracle external tables as file_fdw foreign tables.

The database content report and the migration cost estimation is a work in
progress so all feedback on these new features are welcome. Here is the complete
changelog:

	- Update documentation about ora2pg usage and new feature.
	- Fix quote escaping on table comments. Thanks to Sebastian Fischer.
	- Fix some other issues with 8i databases, added database version auto-
	  detection to avoid printinf warning. Thanks to Sebastian Fischer for
	  the help.
	- Allow null value in BFILE to the oar2pg_get_bfilename().
	- Update documentation about BFILE export.
	- Add drop function ora2pg_get_bfilename() when necessary.
	- Add support to BFILE external path export by creating a function
	  ora2pg_get_bfilename( p_bfile IN BFILE ) to retrieve path from BFILE.
	  BFILE will be exported as text field with the full path to the file as
	  value. Note that this is the first time that Ora2Pg need write access
	  to the Oracle database, if you do not have BFILE or you have set the
	  corresponding PostgreSQL type asd bytea (the default) the function
	  will not be created.
	- Fix a performance issue when extracting BLOB with a LongReadLen upper
	  than 1MB.
	- Fix priviledge on schema created from Oracle package body. Thanks to
	  Dominique Legendre for the report.
	- Add object type in comment before priviledge extraction.
	- Order output of grant to groups grants by object types. This is useful
	  to quickly disable some SQL orders corresponding of not already loaded
	  objects. Thanks to Dominique Legendre for the feature request.
	- Fix progress bar output.
	- Fix priviledge on sequence, tablespace and schema.
	- Fix backward compatibility with Oracle 8i, remove query with JOIN.
	  Thanks to Sebastian Fischer for the report.
	- Fix backward compatibility with Oracle 8i on priviledge extraction.
	  Thanks to Sebastian Fischer for the report.
	- Fix backward compatibility with Oracle 8i on index extraction. Thanks
	  to Sebastian Fischer for the report.
	- Add more precision in cost estimation.
	- Add somme other PL/SQL uncovered code detection.
	- Add more debug information during data extraction.
	- Removed progress bar when debug is enabled.
	- Add report and estimate cost about CHECK constraint and function
	  based indexes.
	- Update documentation about new export directives SHOW_REPORT and
	  ESTIMATE_COST.
	- Add --estimate_cost and --cost_unit_value command line options.
	- Add ESTIMATE_COST and COST_UNIT_VALUE to default configuration file.
	- Rewritte and extend support to ROWNUM replacement.
	- Remove incompatible grants between Oracle and the PortgreSQL export,
	  especially on views.
	- Limit GRANT export to VALID object. Activate EXPORT_INVALID to enable
	  grants export on all object.
	- Add export of VALID only views. To export all with INVALID ones you
	  must activate the EXPORT_INVALID directive. Thanks to Dominique
	  Legendre for the feature request.
	- Fix issue in substr() pl/sql replacement, thanks to Dominique
	  Legendre for the report, plus add other code replacements in pl/sql.
	- Fix issue with function name not on the same line as the create
	  statement - was affecting file input only.
	- Add report of number of JOB object in the database (SHOW_REPORT).
	- Add PL/SQL replacement of various form of EXEC function call.
	- Remove creation of password with users that are not requiring
	  password. Thanks to Dominique Legendre for the feature request.
	- A sql type and a precision can now be used in REPLACE_AS_BOOLEAN to
	  replace all filed with that type as a boolean, example:
	  NUMBER:1 will replace all field of type NUMBER(1) as a boolean.
	- Fix grants on partition export, will now used all_ and user_ tables.
	- Fix removing of newline in the DECLARE clause. Thanks to Dominique
	  Legendre for the report.
	- PostgreSQL client_encoding is now forced to UTF8 when BINMODE is set
	  to utf8. Thanks to Dominique Legendre for the report.
	- Replace DISABLE TRIGGER ALL by DISABLE TRIGGER USER following the value
	  if USER_GRANTS to avoid permission denied on constraint trigger when
	  data are load under a non PG superuser. Thanks to Dominique Legendre
	  for the report.
	- Rename DISABLE_TABLE_TRIGGERS to DISABLE_TRIGGERS and set default value
	  to 0. Other values are USER or ALL following the connected user.
	- Fix missing newline after comment in PL/SQL code. Thanks to Dominique
	  Legendre for the report.
	- Fix report message on external table export.
	- The export TYPE have been entirely rewritten to only export supported
	  user defined types. Exported are: Nested Tables, Object type, Type in
	  herited and Subtype, Varrays. Associative Arrays, Type Body and type
	  with member method are not supported.
	- When FILE_PER_INDEX is enable, SQL order to move indexes in their
	  respective tablespace will be written into a dedicated file prefixed
	  by TBSP_INDEXES_.
	- Fix location on external table export. Thanks to Thomas Reiss for
	  the help.
	- PG_SUPPORTS_INSTEADOF is now activated by default, that mean that
	  migration should be done on PG >= 9.1.
	- Remove obsolete --xtable commande line option, should be replaced by
	  --allow, backward compatibility is preserved.
	- Add EXTERNAL_TO_FDW configuration directive, disable by default, to
	  export all Oracle external tables as file_fdw foreign tables.
	- Fix an other case where user defined type were not exported with an
	  ending semi-colon. Thank to Dominique Legrendre for the report.
	- Fix export of user defined type with extra ");" at end of the type
	  definition and remove system types from export. Thanks to Dominique
	  Legendre for the report.
	- Add PLSQL replacemement of currval. Thanks to Thomas Reiss for the
	  patch.
	- Add PLSQL replacement of PIPELINED and PIPE ROW by SETOF and RETURN
	  NEXT.
	- Add rewrite of Oracle DETERMINISTIC function into PostgreSQL
	  IMMUTABLE function.
	- Fix copy during install on MacOSx and add /Y option to windows
	  install copy to force overwrite existing files. Thanks to Dennis
	  Spaag for the report.
	- Fix issue exporting rows with perl ARRAYS ref. Thanks to Sorin
	  Gheorghiu for the report.
	- Add report of number of database link in SHOW_REPORT output.
	- Fix major bug on export of NUMBER with precision, they was all
	  exported as bigint. Thanks to Dominique Legendre for the report.
	- Add progress bar during SHOW_REPORT export.
	- Add detailed report about index in SHOW_REPORT output.
	- Fix data export when schema was given in lower case. Thanks to
	  Dominique Legendre for the report.
	- Add SHOW_REPORT export type to display a full summary of the Oracle
	  database content.
	- PLPGSQL: add the name keyword to XMLELEMENT call. Thanks to Thomas
	  Reiss for the hint.
	- Add SHOW_VERSION export type to display the version of Oracle.
	- Add COLLATION to the keyword list. Thanks to Dominique  Legendre for
	  the report
	- Change documentation to add more detail on exporting Oracle views as
	  PostgreSQL tables based on the new VIEW_AS_TABLE directive.
	- Add -a | --allow option and --view_as_table to ora2pg script.
	- Add VIEW_AS_TABLE configuration option to allow export of view as
	  table and permit the additional use of the ALLOW or/and EXCLUDE
	  directive. Thanks to Dominique Legendre for the feature request.
	- Removed conflict with transaction when DEFER_FKEY was enabled and
	  allow DEFER_FKEY and DROP_FKEY to be enabled both. Before, only
	  DEFER_FKEY was used in this case, now both are used and of course
	  DEFER_FKEY is wasted. Thanks to Dominique Legendre for the report.
	- Directives ALLOW and EXCLUDE are now usable with all kind of object
	  following the export type.
	- Rename TABLES directive as ALLOW to be less confusing, backward
	  compatibility is preserved.
	- Thanks to Dominique Legendre for the feature request.
	- Remove auto ordering of table export following the foreign keys to
	  fix an infinite loop. Thanks to Siva Janamanchi for the report.
	- Rewrite the view as table export to reuse the same code as table
	  export, old code was resulting in issues with disable triggers and
	  deferring constraints.
	- Remove alter session to set NLS_NCHAR that was returning error on some
	  installation. Thanks to Dominique Legendre for the report.
	- Fix replacement of IS SELECT wrongly replaced by IS PERFORM in some
	  case. Thanks to Dominique Legendre fot the report.

UPGRADE: Almost all files have changed so a new installation is required.

2012 10 07 - v9.3

	- Add auto detection of Oracle character set and the corresponding
	  PostgreSQL client encoding to use. NLS_LANG and CLIENT_ENCODING
	  configuration directives can be leaved commented, Ora2Pg will set
	  their values automatically.
	- Add PL/SQL replacement of CURSOR IS SELECT by CURSOR FOR SELECT and
	  IS REF CURSOR by REFCURSOR. Thanks to Dominique Legendre for the
	  report.
	- Fix missing set client_encoding orders into fonction or procedure
	  export file. Thanks to Dominique Legendre for the report.
	- Fix not working SKIP configuration directive. Thanks to Siva
	  Janamanchi for the report.
	- Add configuration directive NULL_EQUAL_EMPTY to disable the IS NULL
	  and IS NOT NULL replacement into PL/SQL code. Enabled by default.
	  Thanks to Dominique Legendre for the feature request.
	- Remove exclusion of object names with the dollar sign. Thanks to
	  Konrad Beiske for the suggestion.
	- Fix timestamp with time zone when microsecond is enabled. Thanks
	  to Siva Janamanchi for the report.
	- Fix extra semi-column when PKEY_IN_CREATE is enabled. Thanks to
	  Siva Janamanchi for the report.
	- Update configuration about boolean replacement.
	- Allow any column type replacement as a boolean in PostgreSQL, values
	  will be converted as well. Thanks to Konrad Beiske for the feature
	  request.
	- Add REPLACE_AS_BOOLEAN and BOOLEAN_VALUES configuration directives
	  to allow type replacement with a boolean. Thanks to Konrad Beiske
	  for the feature request.
	- Add new configuration directive PKEY_IN_CREATE to add primary keys
	  definition in CREATE TABLE statement instead of creating them after
	  with an ALTER TABLE statement. For Greenplum database, primary key
	  must be created with the CREATE TABLE statement so you may want to
	  enable this configuration directive. Thanks to Siva Janamanchi for
	  the feature request.
	- Add new configuration directive USE_RESERVED_WORDS to force Ora2Pg to
	  auto-detect PostgreSQL reserved words in Oracle object's names and
	  automatically double quote them. Thanks to Siva Janamanchi for the
	  feature request.
	- SHOW_TABLE and SHOW_COLUMN will now display a warning when Oracle
	  object's name is a PG reserved words. Those names will need to be
	  renamed or double-quoted (see USE_RESERVED_WORDS).
	- Add TIMESTAMP WITH LOCAL TIME ZONE Oracle type conversion to timestamp
	  and force timestamp with time zone format to use TZH:TZM. Thanks to
	  Siva Janamanchi for the report.
	- Fix table and column replacement issues introduced with path that
	  removed double-quote when PRESERVE_CASE is disabled. Thanks to Steve
	  DeLong for the report.
	- PLPGSQL convertion: Fix SELECT replacement with PERFORM in VIEW
	  declaration. Thanks to Thierry Capitaine for the report.
	- Add display Ora2Pg type conversion map between Oracle originals types
	  and PostgreSQL's types when using export type SHOW_COLUMN. Thanks to
	  Thierry Capitaine for the feature request.
	- Reorder command line options in ora2pg script usage and documentation.
	- Add call to quote_ident() and quote_literal() into materialized
	  functions to secure parameters.
	- Fix major issue in pl/sql to pl/pgsql conversion with multiple package
	  declaration in the same code record. Thanks to Marc Cousin for the
	  report.
	- Add data type TIMESTAMP WITH TIME ZONE. Thanks to Siva Janamanchi for
	  the report.
	- Add new export type: MVIEW to allow export of all materialized views
	  as snapshot materialized view (fully reload of the view).
	- Add -e | --exclude option to Perl script ora2pg to exclude some given
	  objects from the export. It will override any value of the EXCLUDE
	  directive. The value is a coma separated list of object name or regex.
	- Update domumentation about the EXCLUDE directive change.
	- Allow exclusion from export of functions, procedures and functions in
	  package by specifying a list of name or regex in EXCLUDE directive.
	  Thanks to Keith Fiske from Omniti for the feature request.

UPGRADE: Almost all files have changed so a new installation is required.

2012 09 05 - v9.2

	- In plpgsql conversion, SELECT without INTO becomes PERFORM.
	- In plpgsql conversion, EXECUTE IMMEDIATE replaced by EXECUTE.
	- Fix DATA_TYPE value in configuration file.
	- Fix case sensitivity on data export using COPY mode.
	- Directive XML_PRETTY is now disabled by default as it is better to
	  use getClobVal() to get the XML data out of an xmltype column.
	- Add documentation about regex usage int EXCLUDE and TABLES directives.
	- Remove all double-quote around object name when CASE_SENSITIVY is
	  disabled. Thanks to Dominique Legendre for the suggestion.
	- Rename CASE_SENSITIVE by PRESERVE_CASE to be less confusing, backward
	  compatibility preserved. Thanks to Dominique Legendre for the request.
	- Add support to user defined type data export. Before it will simply
	  export an array reference ARRAY(0xa555fb8), now the array is explored
	  and inserted as ROW(col1,col2,...). Thanks to Mathieu Wingel for the
	  feature request.
	- Fix bug in direct data import in postgresql with COPY: pg_putcopydata
	  can only be called directly after issuing a COPY FROM command. Thanks
	  to Steve Delong for the report.
	- Add warning at any debug level before abort when a data file already
	  exist during data export.
	- Fix issue with oracle sensitivity when exporting data.
	- Fix search_path on package export, indexed and constraints files on
	  TABLE export.
	- Remove obsolete ORA_SENSITIVE configuration directive, thanks to
	  Dominique Legendre it is no more used.
	- Force automatic conversion of PL/SQL when entry is an input file.
	- Fix errors in main file for package loader with FILE_PER_FUNCTION
	  enabled.
	- Fix case where package body where not exported.
	- Add missing EXPORT_INVALID directive into default configuration file.
	- Fix replacement of END followed by the name of the function, the semi-
	  colon was removed.
	- Fix case sensitivity issue in INDEX creation.
	- Fix case sensitivity issue in CHECK constraint and a typo in a
	  progress bar variable.
	- Replace old DATA export type by INSERT in configuration file.
	- Fix case sensitivity issue in ALTER TABLE ... ADD CONSTRAINT. Thanks
	  to David Greco for the report.
	- Add set client_encoding before table output to handle character
	  encoding into comments and possibly objects names.
	- Fix some case sensitivity issue with schema name. Thanks to Dominique
	  Legendre for the report.
	- Do not display warning message about direct import if no connection
	  to a PostgreSQL database is defined.
	- Allow multiple export type to be specified into the ora2pg -t command
	  line option.
	- Dump progress bar to stderr instead of stdout to separate logs.
	- Add new -q | --quiet option to perl script ora2pg to disable progress
	  bar.

2012 08 19 - 9.1

	- Add progress bar to show data export progression.
	- Add -q | --quiet option to ora2pg perl script to disable progress bar.
	- Change documention about tnsnames.ora to mark it is not necessary.
	- Add progress bar during data export, per table and globaly.
	- Replace export type DATA by INSERT to mark the difference with COPY
	  and avoid confusion. Documentation is updated and full backward
	  compatibility preserved.
	- Improve Oracle case sensitivity detection on column and update
	  documentation about ORA_SENSITIVE directive
	- Direct import for COPY statement now used DBD::Pg and pg_putcopydata()
	  instead of a pipe to psql command.
	- Fix case sentitivity issue on disabling/enabling all triggers.
	- Add autodetection of case sensitivity with column name.
	- Move trunc() to data_truc() convertion into the ALLOW_CODE_BREAK part.
	- Update comment about FILE_PER_FUNCTION in configuration file.
	- Fix NOT NULL constraint add twice, the first time in the column
	  definition and the second time in an ALTER TABLE ... ADD CONSTRAINT
	  ... CHECK ( ... NOT NULL). Reported by Dominique Legendre.
	- Add support to direct CALL of stored procedures in trigger definition.
	  Reported by Dominique Legendre.
	- Remove index creation on primary and unique key autogenerated by
	  PostgreSQL.
	- Fix PL/SQL to PLPGSQL automatic convertion on index when exporting
	  data with DROP_INDEX activated.
	- Fix DROP_INDEX to only delete indexes that will be created at end.
	- Fix search path when exporting data with EXPORT_SCHEMA disabled
	- Add missing documentation about the LOGFILE directive
	- Fix case sensitivity on sequence export. They will now always be
	  insensitive as in PostgreSQL its called is converted between quotes:
	  nextval('seq_name'). Reported by Dominique Legendre.
	- Limit export of primary and unique key if KEEP_PKEY_NAMES is enabled
	  to those who are not autogenerated by Oracle. Reported by Dominique
	  Legendre.
	- Trigger export is now limited to those belonging to table that are not
	  excluded from the export (see TABLES and EXCLUDE directives). Reported
	  by Dominique Legendre
	- Fix case sensitivity on trigger export.
	- Fix data export failure in table with column name with accent.
	  Reported by Dominique Legendre.
	- Fix set client_encoding syntax. Reported by Dominique Legendre
	- Add automatic try with oracle sensitivity when an error occurs during
	  retreving table information. This additionaly also fixes an error when
	  table has accent on his name.
	- Fix replacement of user defined data type with directive DATA_TYPE.
	  Reported by Dominique Legendre.
	- Fix function or procedure detection with external input file. Reported
	  by Arul Shaji.
	- Update documentation about Windows installation and ActiveState Perl
	  distribution. Thanks to Stephan Hilb for the report.
	- Fix date format issue by forcing NLS_DATE_FORMAT to format:
	  YYYY-MM-DD HH24:MI:SS. Thanks to Stephan Hilb for the report.
	- Remove obsolete pod documentation in Ora2Pg.pm.
	- Add new configuration directive CREATE_SCHEMA to disable the sql
	  command of schema creation at top of the output file during TABLE
	  export type. Patch by David Greco.
	- Added converting INSERTING/UPDATING/DELETING to PG_OP=INSERT, etc.
	  Patch by David Greco.
	- Fix parsing leading ':' on triggers, as they generally have :NEW and
	  :OLD to denote the new and old recordsets. Patch by David Greco
	- Add new PG_INTEGER_TYPE configuration directive activated by default,
	  to limit conversion into postgresql integer or bigint of Oracle number
	  without scale - NUMBER(p), PG_NUMERIC_TYPE is now reserved to convert
	  NUMBER(p,s). Patch by David Greco.
	- Limit numeric with precision <= 9 to be converted as integer, numeric
	  with precision >= 10 will be converted to bigint to handle integer
	  above 2147483647. Patch by David Greco.
	- Add plsql to plpgsql automatic conversion on check constraints. Patch
	  by David Greco.
	- Add plpgsql replacement, patch by David Greco:
	  REGEX_LIKE( string, pattern ) => string ~ pattern
	- Update documentation about NOESCAPE and STANDARD_CONFORMING_STRING
    	- Change place of the ENABLE_MICROSECOND into the documentation.
	- Fix forgot to add documentation about encryption with Oracle server.
	- Add missing DISABLE_COMMENT configuraton directive in default
	  configuration file and update documentation

2012 07 15 - 9.0

	- Remove call to obsolete BINDIR and MANDIR variables into packaging
	  scripts to reflect the changes in Makefile.PL.
	- Update documentation about installation of Ora2Pg under Windows.
	- Automatically set LONGREADLEN to ORA_PIECE_SIZE length if the last one
	  is larger, for CLOB export.
	- Change Makefile.PL and source tree to fully support installation under
	  Windows OSes.
	- Change double quote by single in Makefile.PL perl replacement call.
	- Replace double quote by single one in $CONFIG_FILE default setting to
	  simplify automatic replacement at install.
	- Fix CLOB export that was limited to 64Kb even with LONGREADLEN defined
	  to an upper value. Patch use the ora_piece_size DBD::Oracle prepare
	  attribute. Patch by Mohamed Gargouri. See here for more detail:
		http://search.cpan.org/~pythian/DBD-Oracle-1.46/lib/DBD/Oracle.pm#Piecewise_Fetch_with_Polling
	- Add a note into documentation about encrypted communication between
	  Ora2Pg and Oracle. Note by Jenny Palomino.
	- Change documentation to reflect change to the format of the Oracle
	  timestamp with millisecond. This format is now enabled by default in
	  configuration file.
	- Fix bug with LONGREADLEN and LONGTRUNCOK when exporting LOB that was
	  not applied even after change into the configuration file. Reported
	  by Mohamed Gargouri
	- Fix microsecond format FF3 not compatible with Oracle 8i. Set to FF.
	- Add a warning to stderr when a table export need that ORA_SENSITIVE
	  be enabled.
	- Fix case where Oracle indexes with same name as a constraint was not
	  exported - Rodrigo

	The following are old patches that was not applied to v8.10 and the git
	repository:

	- Fix creation of bad constraint for each indexes.
	- Add DISABLE_COMMENT configuration directive to remove comments from
	  TABLE export type. Comments are exported by default.
	- Fix a bug in removing function name repetion at end
	- Add PL/SQL to PLGPSQL replacement of the to_number function
	- Fix PL/SQL to PLGPSQL replacement of substr into substring
	- Add replacement of specials IEEE 754 values BINARY_(FLOAT|DOUBLE)_NAN
	  and BINARY_(FLOAT|DOUBLE)_INFINITY by NaN and Infinity on PLPGSQL
	  conversion and on data export - Thanks to Daniel Lyons.
	- Fix return type of function with just OUT or INOUT params. Thanks to
	  Krasi Zlatev for the patch.
	- Add schema name on functions or procedures export when EXPORT_SCHEMA
	  is activated. Thanks to Krasi Zlatev for the patch.
	- Fix case sensitivity issue with schema on partition export.
	- Fix case sensitivity issue with --xtable option.
	- Fix issues with case sensitivity on the schema owner set into the
	  SCHEMA configuration directive.
	- Add default search_path on schema for contraints, index and data
	  export when EXPORT_SCHEMA is activated.
	- Fix case sensitivity issue in search_path.
	- Force Oracle datetime format to be YYYY-MM-DD HH24:MI:SS.FF in client
	  session to prevent other defined output format. Thanks to Aaron Culich
	  for the patch.
	- Add export/import of table and column comment. Thanks to Krasi Zlatev
	  for the patch.

2012 06 26 - 8.13

	- Fix broken export with missing single quote in Oracle timestamp export
	  formating with to_char(timestampcolumn, YYYY-MM-DD HH24:mi:ss). Thanks
	  to Steve Delong for the report.

2012 06 22 - 8.12

	- Add nex configuration directive ENABLE_MICROSECOND to allow timestamp
	  to be exported with millisecond precision. Thanks to Patrick King for
	  the feature request.
	- Fix multiple quote on foreign keys column names. Thanks to Vitaliy for
	  the report.
	- Add new export type FDW to allow table export as foreign table for
	  oracle_fdw. Thanks to David Fetter for the feature request.
	- Fix typo in LongTruncOk variable name. Thanks to Magnus Hagander for
	  the patch.
	- Add XML_PRETTY configuration directive to replace getStringVal() by
	  getClobVal() when extracting XML data. Thanks to Magnus Hagander for
	  the patch.
	- Fix case sensitivity issue in ALTER TABLE and TRUNCATE statement.
	  Thanks to Magnus Hagander for the patch.

UPGRADE: Ora2Pg.pm and ora2pg perl scripts have changed as well as configuration
file. Documentation has been updated too so you'd better install all again.

2012 04 30 - 8.11

	- Fix an error when running ora2pg directly against PG, index and
	  constraints are created against PG instead of being written to
	  the output file. Thanks to David Greco for the report.
	- Ora2Pg will now output a warning message when direct import to PG
	  is set with other import type than COPY and DATA.
	- Fix NUL character removing on LOB to bytea export. Thanks to info31
	  on PostgresqlFr for the report.

2012 03 11 - 8.10

	- Add two configuration directives to control the BLOB export.
	  LONGREADLEN to set the database handle's 'LongReadLen' attribute
	  to a value that will be the larger than the expected size of the
	  LOB. LONGTRUNKOK to bypass the 'ORA-24345: A Truncation' error.
	  Thanks to Dirk Treger for the report.
	- Fix install problem on non-threaded Perl and the threads package.
	  Replace use() by require() call. Thanks to Ian Sillitoe for the patch.
	- Fix strange Oracle behaviour where binary_double infinity is exported
	  from Oracle as '~'. Replaced by 'inf'. Thanks to Daniel Lyons for the
	  report.

UPGRADE: only Ora2Pg.pm have changed so you can just override it. See also
documentation for new configuration directives: LONGREADLEN and LONGTRUNKOK.
	
2011 11 07 - 8.9

	- Fix double quote into file name of package function export when case
	  sensitivity is preserved.
	- Add support to XMLType data extraction. Thanks to Aaron Culich for
	  the report. Before this release, xml data was exported as a Perl
	  array reference.
	- Fix bug during foreign key export when foreign keys have different
	  owners. Thanks to Krasi Zlatev for the patch.
	- Add support to plpgsql conversion during index extraction as many
	  index use some Oracle function on their declaration. Thanks to
	  Sriram Chandrasekaran fot the feature request.
	- PLSQL: Add replacement of Oracle subtr() by PostgreSQL substring().
	  Thanks to Sriram Chandrasekaran fot the feature request.
	- PLSQL: Add replacement of Oracle decode() by PostgreSQL CASE/THEN/ELSE.
	  Thanks to Sriram Chandrasekaran fot the feature request.
	  Note that this two replacement was not implemented because they could
	  break the code if there's complex subqueries inside their declaration.
	  This is why you can enable it by setting ALLOW_CODE_BREAK to 1 (new).
	  In later release this directive will be enable by default.
	- Add output ordering on some object name so that results between two
	  runs can be compared. Thanks to Henk Enting for the patch.
	- Fix misshandling of all cases of RAISE_APPLICATION_ERROR rewrite into
	  RAISE EXCEPTION concatenations. Thanks to Krasi Zlatev for the report.

UPGRADE: only Ora2Pg.pm and Ora2Pg/PSQL.pm have changed so you can just override them
if you dont want to reinstall all.

2011 10 13 - 8.8

	- Before that release when you upgraded Ora2Pg using Makefile, the old
	  ora2pg.conf was renamed as ora2pg.old. This can lead to lost previous
	  configuration, the old ora2pg.conf is now untouched and the new one is
	  installed as ora2pg.conf.new.
	- Renamed ora2pg.pl into ora2pg_pl in the package before installation
	  to avoid the copy of the perl script into the site Perl install dir.
	  It is still installed as ora2pg in /usr/local/bin by default.
	- Fix errors that appeared to be due to no quoting on the field names
	  when ORA_SENSITIVE is enabled. Thank to Sam Nelson for the patch.
	- Limit case sensitivity on check constraints to column names only,
	  before that if there was a value between double quote into the check
	  constraint, it was wrongly changed to lower case.
	- Fix broken case sensitivity at data export when disabling/enabling
	  triggers and truncating tables with copy or insert statement.
	- Change Ora2Pg version in packaging files that was still in 8.5.

UPGRADE: only Ora2Pg.pm have changed so you can just override it.

2011 09 07 - 8.7

	- The escape_bytea() function has been rewritten using a prebuild array
	  to gain twice of performances. Thanks to Marc Cousin from Dalibo for
	  the patch.
	- Improve speed of bulkload data by disabling autocommit by issuing a
	  BEGIN at the start and COMMIT at the end.
	- Add multi-threading support. It is only used to do the escaping to
	  convert LOBs to byteas, as it is very cpu hungry. There's a lot of
	  CPU-waste here. The threads number is controlled by a new configuration
	  directive: THREAD_COUNT. Putting 6 threads will only triple your
	  throughput, if your machine has enough cores. If zero (default value),
	  do not use threads, do not waste CPU, but be slower with bytea.
	  Performance seems to peak at 5 threads, if you have enough cores, and
	  triples throughput on tables having LOB. Another important thing:
	  because of the way threading works in perl, threads consume a lot of
	  memory. Put a low (5000 for instance) DATA_LIMIT if you activate
	  threading. Many thanks to Marc Cousin for this great patch.
	- Fix standard_conforming_string usage on export as INSERT statement.
	- Fix an issue with importing Oracle NULL character (\0 or char(0)) with
	  bytea and character data with UTF8 encoding. Now whatever is the data
	  type or the encoding, this character is simply removed to prevent the
	  well known 'ERROR: invalid byte sequence for encoding "UTF8": 0x00.'
	  Thanks to Jean-Paul Argudo from Dalibo for the report.
	- Fix an incorrect syntax for "for each statement" triggers.
	  Thanks to Henk Enting for the report.
	- Add comment at end of line to signal on which cursor the replacement
	  on " EXIT WHEN (...)%NOTFOUND " is done. This will return something 
	  like "IF NOT FOUND THEN EXIT; END IF; -- apply on $1". Thanks to
	  jehan Guillaume De Rorthais from Dalibo for the report this help a lot
	  during Pl/Pgsql code review.
	- Fix table/column name replacement on constraint creation and dropping
	  when REPLACE_TABLES/REPLACE_COLS is set during DATA/COPY export.
	- Fix table/column name replacement on indexes creation and dropping
	  when REPLACE_TABLES/REPLACE_COLS is set during DATA/COPY export.
	- Remove unused table name parameter in _drop_indexes() function.
	- Add support to REPLACE_TABLES/REPLACE_COLS during schema export.
	  Before this release those replacements were only applied to DATA or
	  COPY export. You can now use it in schema export, it will replace
	  table and/or column names in the TABLE/INDEX/CONSTRAINT schema export.
	  MODIFY_STRUCT is still limited to DATA or COPY export as it have no
	  sense outside this export. Unfortunately those replacements can not be
	  done easilly in other export type like TRIGGER, FUNCTION, etc. so you
	  must still edit this code by hand.
	- Use the bundled Perl Config module to detect if Perl is compiled with
	  useithread. This mean that the old local defined %Config hash has been
	  replaced by %AConfig.
	- SKIP indices is now obsolete and must be replaced with SKIP indexes.
	  backward compatibility is preserved.
	- The file generated when FILE_PER_INDEX is activated has been renamed
	  into INDEXES_... instead of INDICES_...
	- Add a warning on tablespace export when Oracle user is not a dba.
	- Fix fatal error when dumping to one file per function with double
	  directory output.
	- Fix double print of FATAL messages and dirty database disconnect on
	  fatal errors.
	- Add setting of client_encoding into each export type as defined in
	  the configuration file.
	- Update web site documentation.

UPGRADE: Ora2Pg.pm, Ora2Pg/PGSQL.pm and ora2pg have changed so they must be
overwritten.  There's also changes in the configuration file and documentation
has changed as well. Backward compatibility is fully preserved.


2011 07 07 - 8.6

	- Remove "use strict" from head of Ora2Pg.pm that breaks view export.
	  This is usually removed before public release, but not this time.
	  Thanks to Jehan Guillaume de Rorthais from Dalibo for the report.
	- Add a new configuration directive called COMPILE_SCHEMA that force
	  Oracle to compile the PL/SQL before code extraction to validate
	  code that was invalidate for any reason before. If you set it to 1,
	  you will force compiling of the user session schema, but you can
	  specify the name of the schema to compile as the value too. Thanks
	  to Jean-Paul Argudo from Dalibo for the solution.
	- Add new configuration directive EXPORT_INVALID to allow export of all
	  PL/SQL code even if it is marked as invalid status. The 'VALID' or 
	  'INVALID' status applies to functions, procedures, packages and user
	  defined types.
	- Excluded from export all tables, types, views, functions and packages
	  that contains a $ character. Most of the time they don't need to be
	  exported.
	- PLSQL: add automatic conversion of Oracle SYS_REFURSOR as PostgreSQL
	  REFCURSOR.
	- Rewrite entirely the parser of DBMS_OUTPUT du to concatenation errors.
	- PLSQL: add automatic replacement of some Oracle exception errors:
          INVALID_CURSOR=>INVALID_CURSOR_STATE, ZERO_DIVIDE=>DIVISION_BY_ZERO,
          STORAGE_ERROR=>OUT_OF_MEMORY.

UPGRADE: Ora2Pg.pm and Ora2Pg/PGSQL.pm have changed so they must be overwritten.
There's also changes in the configuration file and documentation has changed as
well. Backward compatibility is fully preserved.


2011 07 01 - 8.5

	- When FILE_PER_FUNCTION is activated and export type is PACKAGE, Ora2Pg
	  will now save all functions/procedures of a package body into a
	  directory named as the package name and into different files. This
	  will allow to load each function separatly or load them all with the
	  OUTPUT SQL script generated by Ora2Pg.
	- Fix Oracle package body parsing failure when a procedure is declared
	  inside an other.
	- Add new configuration options FILE_PER_CONSTRAINT and FILE_PER_INDEX
	  to generate three files during the schema extraction. One for the
	  'CREATE TABLE' statements, one for the constraints (primary keys,
	  foreign keys, etc.) and the last one for indices. Thanks to Daniel
	  Scott for the feature request.
	- Allow to process PL/SQL Oracle code from file instead of a database
	  to apply Ora2Pg code conversion. Thank to Mindy Markowitz for the
	  feature request. See -i or --input_file command line option to ora2pg
	  perl script or INPUT_FILE new configuration option.
	- Add new configuration directive STANDARD_CONFORMING_STRINGS that is
	  used only during DATA export type to build INSERT statements. This
	  should avoid 'WARNING:  nonstandard use of \\ in a string literal'.
	  Please check that this behavior is backward compatible with your
	  PostgreSQL usage as this is enabled by default now.

UPGRADE: The new features has changed Ora2Pg.pm and ora2pg.pl so that they must
be overwritten. There's also changes in the configuration file and documentation
has changed as well. Take care of backward compatibility with escaped strings in
DATA export type and the new behavior on PACKAGE export. 

2011 06 07 - 8.4

	- Moves Ora2Pg to SourceForge.net.
	- Fix an issue on setting owner in "ALTER SEQUENCE ... SET OWNER TO".
	  Thanks to Herve Girres for the report.
	- Bugfix on lower case convertion for check constraints extraction.
	  Thanks to Alexander Korotkov for the patch.

UPGRADE: There's no new functionality, this is a bug fix release.

2011 05 11 - 8.3

	- Fix issue on inherited user defined types converted to inherited tables.
	  Add comment on unsupported inherited type in PostgreSQL too. Thanks to
	  Mathieu Wingel for the report.
	- Fix issue on column default values. Oracle all this kind of strange
	  syntax: counter NUMBER(4) default '' not null, that was translated to
	  counter smallint DEFAULT '' by Ora2Pg. Thanks to Mathieu Wingel this is
	  now rewritten as DEFAULT NOT NULL.
	- Fix case sensitivity on create view when there was double quote on the
	  column name statement part. Thanks to Mathieu Wingel or the report.
	- Fix bad patch applied on column name case sensitivity issue during check
	  constraint export. Thanks to Philippe Rimbault for the report.
	- Fix bug on package export introduced into version v8.2. The issue was
	  related to end of package procedure detection. hanks to Mathieu Wingel
	  or the report.

UPGRADE: There's no new functionality, this is a bug fix release and every one
should upgrade to it.


2011 05 01 - 8.2

	- PLSQL: automatic replacement of EXIT WHEN cursor%NOTFOUND; by Pg
	  synthax: IF NOT FOUND THEN EXIT; END IF;. Works with additional
	  condition too.
	- PLSQL: Automatic replacement of SQL%NOTFOUND by NOT FOUND.
	- PLSQL: Add detection of TOO_MANY_ROW to NO_DATA_FOUND to add STRICT.
	- Completely rewrite the parsing of Oracle package body to handle all
	  cases and especially prodedure declared into an other procedure.
	  Those procedure are renamed INTERNAL_FUNCTION and must be rewritten.
	- Fix type usage of ora2pg Perl script.
	- Add a new directive FORCE_OWNER. By default the owner of the database
	  objects is the one you're using to connect to PostgreSQL. If you use
	  an other user (postgres for exemple) you can force Ora2Pg to set the
	  object owner to be the one used in the Oracle database by setting the
	  directive to 1, or to a completely different username by setting the
	  directive value to that username. Thanks to Herve Girres from Meteo
	  France for the suggestion and patch.
	- Add --forceowner or -f command line option to ora2pg program.
	- Add SHOW_ENCODING extract type to return the Oracle session encoding.
	  For example it can return: NLS_LANG AMERICAN_AMERICA.AL32UTF8 
	- Remove SYS_EXTRACT_UTC from index creation as Pg always stores them
	  in UTC. Thanks to Daniel Scott for the patch.
	- In PLSQL code SYS_EXTRACT_UTC is replaced by the Pg syntaxe:
	  field AT TIME ZONE 'UTC'.
	- Fix a pending problem with "Wide character in print at" on COPY mode.
	  Thanks to Bernd Helmle from Credativ GmbH for the patch.
	- PLSQL: Add automatic rewrite of FOR ... IN REVERSE ... into Pg synthax
	- Fix column name case sensitivity issue during check constraint export.
	  Thanks to Daniel Berger for the report.
	- Remove the possibility to add comment after a configuration directive
	  it may not be used and it was generating an issue with the passwords
	  configuration directives for examples. Thanks to Daniel Berger for the
	  report.
	- Complete rewrite of user defined type extraction. Add support of inherited
	  type using Oracle UNDER keyword as well as better support to custom type
	  with BODY. Thanks to Mathieu Wingel for the report.
	- Fix case sensitivity on user defined types. Thanks to Mathieu Wingel for
	  the report.

UPGRADE: All files have changed so you need a fresh install/upgrade.
Previous release used to remove any string starting from a # in the config file,
this was to allow comments after a configuration directive. This possibility
have been removed in this release so you can no more add comments after a
configuration directive.


2011 03 28 - 8.1

	- Prevent Ora2PG to export twice datas when using FILE_PER_TABLE and
	  the data output file exist. This is useful in case of export failure
	  and you don't want to export all data again. This also mean that if
	  you want to get new data you have to remove the old files before.
	- Fix parsing of procedure/function into pl/sql Oracle package.
	- Fix bug in IS NULL/IS NOT NULL replacement. Thanks to Jean-Paul Argudo
	  from Dalibo for the report.
	- Add CREATE OR REPLACE on RULE creation.
	- Add DROP TRIGGER IF EXISTS before trigger creation.
	- Replace Oracle date "0000-00-00" by NULL.
	- Fix infinite loop in package/fonction type replacement.
	- Add one file per package creation if FILE_PER_FUNCTION is enabled.
	- Fix double quote in name of custom type extraction.
	- Add extraction of custom type IS VARRAY as an custom type of table
	  array. Thank to Jean-Paul Argudo from Dalibo for the patch.
	- Fix multiple double quote in name of create index definition.
	- Apply excluded and limited table to tablespace extraction.
	- Fix function and procedure detection/parsing on package content.
	- Fix schema prefix in function name declaration in package export. 
	- PLSQL: Replace some way of extracting date part of a date :
	  TO_NUMBER(TO_CHAR(...)) rewritten into TO_CHAR(...)::integer when
	  TO_NUMBER just have one argument.
	- Fix Makefile.pl error when trying to modify file ora2pg now renamed
	  into ora2pg.pl
	- Add 3 new export types SHOW_SCHEMA, SHOW_TABLE and SHOW_COLUMN. Those
	  new extraction keyword are use to only display the requested information
	  and exit. This allow you to quickly know on what you are going to work.
	  The SHOW_COLUMN allow a new ora2pg command line option: '--xtable relname'
	  or '-x relname' to limit the displayed information to the given table.
	- Add type replacement for BINARY_INTEGER and PLS_INTEGER as integer.

UPGRADE: Please make a full upgrade asap to this release.

2011 03 15 - 8.0

This major release simplify and improve Oracle to PostgreSQL export. Ora2Pg v8.x
now assume that you have a modern PostgreSQL release to take full advantage of
the Oracle compatibility effort of the PostgreSQL development team. Ora2Pg since
v8.x release will only be compatible with Pg >= 8.4.

	- Remove addition of AS for alias as with modern PG version this can
	  be optional (Pg >= 8.4).
	- Fix CREATE with missing USER/ROLE for grant extraction. Thanks to
	  Herve Girres for the report.
	- Apply missing psql_pgsql converter to view definition.
	- PLSQL : Normalize HAVING ... GROUP BY into GROUP BY ... HAVING clause
	- PLSQL : Convert call to Oracle function add_months() in Pg syntax
	- PLSQL : Convert call to Oracle function add_years() in Pg syntax
	- Apply missing psql_pgsql converter to triggers WHEN clause.
	- Fix DECLARE CURSOR rewrite.
	- Allow one file per function / procedure / package exported with a new
	  configuration option FILE_PER_FUNCTION. Useful to editing and testing.
	  Thank to Jean-Paul Argudo from DALIBO for the feature request.
	- The FILE_PER_TABLE configuration option is now also applied to views.
	- Remove obsolete PG_SUPPORTS_INOUT as it is supported by with modern
	  PG version (Pg >= 8.4).
	- Remove obsolete PG_SUPPORTS_DEFAULT as it is supported by with modern
	  PG version (Pg >= 8.4).
	- Allow to adjust PostgreSQL client encoding with a new configuration
	  directive: CLIENT_ENCODING.
	- Add TRUNCATE_TABLE configuration directive to add TRUNCATE TABLE
	  instruction before loading data.
	- Add type conversion of Oracle XMLTYPE into PostgreSQL xml type.
	- PLSQL: SYSDATE is now replaced by LOCALTIMESTAMP to not use timezone.
	  Thanks to Jean-Paul Argudo from DALIBO for the report.
	- Use 'CREATE OR REPLACE' on create trigger function instruction.
	- Fix prefixing by OUTPUT_DIR when file per table/function is enabled.
	- Use 'CREATE OR REPLACE' on create view.
	- PLSQL_PGSQL is now enabled by default. If you want to export Oracle
	  original function/procedure/package code, disable it.
	- PLSQL: WHERE|AND ROWNUM = N; is automatically replaced by LIMIT N;
	- PLSQL: Rewrite comment in CASE between WHEN and THEN that makes Pg
	  parser unhappy.
	- PLSQL: Replace SQLCODE by SQLSTATE

UPGRADE: You must reinstall all and review your configuration file

2011 02 14 - 7.3

	- Remove PG_SUPPORTS_INOUT, now Ora2Pg assumes the PostgreSQL database
	  destination support it (Pg > v8.1).
	- Remove PG_SUPPORT_ROLES, now Ora2Pg assumes the PostgreSQL database
	  destination support it (Pg > v8.1).
	- Complete rewrite of the GRANT (user/role/grant) export type. It now
	  should be only related to the current Oracle database. Note that do
	  not try to import rights asis as you may import have errors or worse
	  miss handling of the rights! Just remember that for example in Oracle
	  a schema is nothing else than a user so it must not be imported like
	  this.
	- Fix multiple errors in partitionning definition. Thank to Reto Buchli
	  for the report.
	- PLSQL: reordering cursor Oracle declaration "DECLARE CURSOR name" into
	  "DECLARE name CURSOR". Thank to Reto Buchli (WSL IT) for the report.
	- Fix miss handling of DEFAULT parameters value in convert_function().
	  Thanks to Leonardo Cezar for the patch.
	- Fix Oracle tablespace export where Pg tablespace location was based on
	  Oracle filename. This fix extract the path and replace the filename
	  with tablespace name. Thank to Reto Buchli (WSL IT) for the report.
	- Fix parsing of ending function code. Thanks to Leonardo Cezar for the
	  patch. 
	- Fix call to _convert_procedure() that is in fact the same function as
	  _convert_function(). Thanks to Leonardo Cezar for the report.
	- Fix multiple call on tablespace alter index on the same object. Thank
	  to Reto Buchli (WSL IT) for the report.
	- PSQL: Rewrite RAISE EXCEPTION concatenations. Double pipe (||) are
	  replaced by % and value is set as parameter a la sprintf. Thank to
	  Reto Buchli (WSL IT) for the report.
	- Add missing comment of PARTITION export type into configutation file.
	- Complete rewrite of the table partition export part has it was not
	  handling all case and was really buggy. 
	- PLSQL: add normalisation of the to_date() function.
	- Ora2Pg now warns during grant export when it is not connected as an
	  Oracle DBA user. GRANT export need rights of Oracle DBA or it fail.
	- Fix install of changelog into Makefile.PL, name was wrong. Thanks to
	  Julian Moreno Patino for the patch.


2011 01 12 - 7.2

	- Fix escaping of BLOB/RAW to bytea data import causing import to crash.
	  Thanks to Ozmen Emre Demirkol for the report.
	- Add support to default value into functions parameter (PG >= 8.4).
	  Can be activated with a new configuration directive: PG_SUPPORTS_DEFAULT. 
	  Default is 1, activated.
	- Fix bad ending of exported function: remove trailing chars after END.
	- Add support to WHEN clause on triggers (PG >= 9.0), can be activated
	  with a new configuration directive: PG_SUPPORTS_WHEN.
	- Add support to INSTEAD OF usage on triggers (incoming PG >= 9.1). Can
	  be activated with a new configuration directive: PG_SUPPORTS_INSTEADOF.
	- Fix error using SKIP directive. Thanks to Laurent Renard from Cap Gemini
	  for the report.
	- Fix missing perl object instance in format_data() function.
	- Fix duplicate procedure or function when export type use both FUNCTION
	  and PROCEDURE.

2010 12 04 - 7.1

	- Improve direct DBD::Pg data export/import speed by 10.
	- Add --section=3 in pod2man call into Makefile.PL. Thanks to Julian
	  Moreno Patino for the report.
	- Renamed ChangeLog into changelog to avoid upstream warning with Debian
	  package. Thanks to Julian Moreno Patino for the suggestion.
	- Fix some spelling mistakes in doc/Ora2Pg.pod. Thanks to Julian Moreno
	  Patino for the fix.
	- Fix release version into Ora2Pg.pm and PLSQL.pm, was still in 6.5.
	- Fix direct data export/import using DBD::Pg. Thanks to Laurent Renard
	  from Cap Gemini for the report.
	- Fix drop/create contraints and index during direct data export/import
	  using DBD::Pg. Thanks to Thierry Grasland from Cap Gemini for the report.

2010 11 23 - 7.0

	- Rename ora2pg perl script into ora2pg.pl in sources because Windows
	  users can't extract the tarball. During install it is renamed into
	  ora2pg. Thanks to Andrew Marlow for the report.
	- Fix doinst.sh for SlackWare Slackbuid packaging.
	- The DEFER_FKEY configuration directive has been fixed as it only
	  works in a transaction. Note that foreign keys must have been created
	  as DEFERRABLE or it also will not works. Thanks to Igor Gelman for the
	  report.
	- Add DROP_FKEY configuration directive to force deletion of foreign keys
	  before the import and recreate them and the end of the import. This may
	  help if DEFER_FKEY not works for you.
	- Add DROP_INDEX configuration directive to force deletion of all indexes
	  except the automatic index (primary keys) before data import and to
	  recreate them at end. This can be used to gain speed during import.
	- Add TSMSYS, FLOWS_020100 and FLOWS_FILES to the owners exclude list.
	  This concern the SRS$ table and all tables begining with 'WWV_FLOW_'
	- Change the way DATA_LIMIT is working. It must be used now to set the
	  bulk size of tuples return at once. Default is 10000.
	- Improve data export speed by 6! The data export code has been entierly
	  rewritten and the speed gain is really fun.
	- Add OUTPUT_DIR configuration directive to set a base directory where all
	  dumped files must be written. Default: current directory.
	- Change value of default numeric(x) type from float to bigint and change
	  default numeric(x,y) type to double precision.
	- Change conversion type for BFILE from text to bytea.

2010 09 10 - 6.4

	- Configuration directives SHOWTABLEID, MIN and MAX are now obsolete
	  and has been definitively removed. They were never used and add too
	  much confusion.
	- Fix bug in column name replacement where table name was also replaced.
	  Thank to Jean-Paul Argudo from DALIBO for the report.
	- Fix case sensitive errata in PG schema search path. Thank to Jean-Paul
	  Argudo from DALIBO for the report.
	- Remove double \n at end of debug message.
	- Fix debug mode not activated if the DEBUG directive is enable and
	  the -d command line is not present.
	- Add unbuffered output for debug message.

UPGRADE: simply override the Ora2Pg.pm Perl module where it is installed.


2010 07 22 - 6.3

	- Fix Oracle 8i compatibility error during schema extraction complaining
	  that column CHAR_LENGTH doesn't exist. Thanks to Philippe Rimbault for
	  the report. Note that the error message is still displayed but tagged
	  as WARNING only.
	- Fix error using the IMPORT option on a read_conf method call. Thanks
	  to Diogo Biazus for the report.
	- Fix export of sequences that does not handle maxvalue well and can be
	  lower than minvalue. Thanks to Nathalie Doremieux for the report.

UPGRADE: Just override Ora2Pg.pm

2010 06 15 - 6.2

	- Change default transaction isolation level from READ ONLY to
	  SERIALIZABLE to ensure consistency during data export. Thanks to
	  Hans-Jurgen Schonig from postgresql-support.de
	- Add the TRANSACTION configuration directive to allow overriding of
	  the isolation level. Value can be readonly, readwrite, committed and
	  serializable. The last is the default.

2010 05 07 - 6.1

	- Fix error on partition export following schema definition.
	- Add first support to export Oracle user defined types.
	- Add CTXSYS,XDB,WMSYS,SYSMAN,SQLTXPLAIN,MDSYS,EXFSYS,ORDSYS,DMSYS,
	  OLAPSYS to the sysuser default exclusion list.
	- PLSQL.pm: Add automatic translation of Oracle raise_application_error
	  and dup_val_on_index to PG RAISE EXCEPTION and UNIQUE_VIOLATION.
	- Change/fallback to a lower case package name (ora2pg-6.x.tar.gz).
	- Change default convert type for 'LONG RAW' to bytea.
	- Add PG_SCHEMA configuration directive to defined a coma delimited
	  list of schema to use in SET search_path PostgreSQL command.


2010 02 28 - 6.0

	- Publish a dedicated site to Ora2Pg at http://ora2pg.darold.net/
	- Add export of Oracle table partitoning. See export type PARTITION.
	- Add command line arguments to perl script ora2pg. See --help for a
	  full listing of these option. The most interesting is --type to change
	  the export type directly at command execution without needing to edit
	  the configuration file, --plsql to directly enable PLSQL to PLPSQL
	  code conversion and --source, --user --password to set Oracle data
	  source. There's also --namespace to set the Oracle schema.
	- Create all file for standard Perl module install. Install is now done
	  with: perl Makefile.PL && make && make install
	- Move Ora2Pg license from Perl Artistics to GPLv3.
	- Move PLSQL package as Ora2Pg::PLSQL for standard Perl module install.
	- Remove use of Perl module String::Random.
	- Rename program ora2pg.pl into ora2pg for standard usage.
	- Fix extra double quote on column name of index export. Thanks to
	  Guillaume Lelarge for the patch.
	- Add packaging facilities to build RPM, SlackBuild and Debian packages.
	- Fix miss handling of Ora2Pg.pm options at object instance init.
	- Configuration file ora2pg.conf is now generated by Makefile.PL

2009 12 18 - 5.5

	- Fix CONSTANT declaration in Procedure/Function/Package export.
	- Fix length of char and varchar on multibyte like UTF8 encoding. Thanks
	to Ali Pouya for the patch.
	- Fix view export where alias to column in Oracle not use 'AS' and
	PostgreSQL required it. Thanks to Ali Pouya for the report.
	- Add type replacement of sql variable in PLSQL code (PLSQL.pm). Thanks
	to vijay for the patch.

2009 07 15 - 5.4

	- Fix bug introduced in multiple output file feature. This bug force
	Ora2pg to crach after the first table export when output is wanted in
	a single file. Thanks to Leo Mannhart for the report.
	- Fix debug filename output on multiple export file. Thanks to Leo
	Mannhart for the report.

2009 07 07 - version 5.3

	- Fix wrong escaping of data named as column during view export. Thank
	to Andrea Agosti for the patch.
	- Allow export of datas into one file per table. See FILE_PER_TABLE
	configuration directive. Thanks to Alexandre - Aldeia Digital for the
	idea.

2009 06 19 - version 5.2

	- Fix order of the column name of the view which was not preserved. Now
	ordered by COLUMN_ID. Thank to Andrea Agosti for the report.
	- Fix case sensitivity in VIEW extraction. Thank to Andrea Agosti for
	the patch.

2009 03 06 - version 5.1

	- Fix missing -U username at Pg connection. Thanks to Brendan Richards.
	- Fix $ENV{ORACLE_HOME} and $ENV{NLS_LANG} to not being overwritten
	by configuration settings if they are already defined in environment.
	- Fix typo in ora2pg.pl where keep_pkey_names was replaced by
	keep_pkeys_name and so prevent use of KEEP_PKEY_NAMES in configuration.
	Thanks to Olivier Mazain for the report.
	- Configuration file directives are now case insensitive.
	- Force $type parameter given to _sql_type() to be uppercase in that
	methode instead of during function call. Thanks to Ali Pouya for the
	report.
	- Modify ora2pg.pl to remove the obsolete call to export_data(). Use
	only export_schema() now.
	- Modify ora2pg.pl to simplify it. Reading configuration is now done
	internally by Ora2Pg.pm as well as all other initialization process.
	You can always overwrite all configuration options into call to new
	Ora2Pg(). Now ora2pg.pl can be as simple as:

		use Ora2Pg;
		my $schema = new Ora2Pg('config' => "/etc/ora2pg.conf");
		$schema->export_schema();
		exit(0);

	This will be less confusing. You can upgrade Ora2Pg.pm without carring
	about that, backward compatibility with previous version is preserved.
	- Review entire documentation with the great help of Ali Pouya.
	- Add type BOOLEAN converted to boolean.
	- PG_SUPPORTS_INOUT is now enabled by default in the configuration file
	- SQL and PL/SQL to PLPGSQL converter:
	    .Replace MINUS call to EXCEPT
	    .Replace DBMS_OUTPUT.put, DBMS_OUTPUT.put_line, DBMS_OUTPUT.new_line
	     by the PLPGSQL equivalent: RAISE NOTICE
	    .Rewrite function/procedure/package convertion functions.
	This Oracle SQL converter for function/procedure/package is now only
	applied if the configuration directive PLSQL_PGSQL is enable, else
	these Oracle code are exported as is. Thanks to Ali Pouya for the help.

	- Reserved call to sql transaction only for DATA export type. Others
	export type now use \set ON_ERROR_STOP ON. Thanks to Ali Pouya.
	- Fix tablespace creation into schema (missing search_path). Thanks to
	Olivier Mazain.
	- Fix the type returned by the _sql_type() method in the case of a
	numeric with null len and pg_numeric_type is set. Thanks to Ali Pouya.
	- Change function body delimiter to $body$ to allow use of $$ into the
	body as quote replacement. Thanks to Ali Pouya.
	- Fix returns type from function. If multiple OUT parameters: RECORD,
	if only one OUT parameter, return his type. If no out parameter: return
	VOID. Thanks to Ali Pouya.
	- Fix export DATA when the name of a column in the table match COMMENT,
	AUDIT or any other defined reserved words. These reserved words are
	defined in a new configuration variable ORA_RESERVED_WORDS. It accept
	a list of comma separated reserved words. Thanks to Andrea Agosti for
	the report.
	- Fix configuration parser that omit custom SYSUSERS definition.

2009 02 13 - version 5.0

	- Fix places where $self->{prefix} where not used. This prefix is
	used to replace DBA_... objects into ALL_... objects. Thanks to Daniel
	Scott report and patch.
	- Fix some problem on trigger export (missing ending semicolon, return
	opaque replaced by return trigger, add missing return new value, single
	quote for delimitating the function body hits against quotes inside the
	function). Thanks to Luca DallOlio for reports and patches.
	- Add first attempt to rewrite plsql code to plpgsql code (see function
	plsql_to_plpgsql in new perl module PLSQL.pm). There's a configuration
	option named PLSQL_PGSQL to activate the convertion.

2008 12 16 - version 4.11

	- Fix Ora2Pg failure on Oracle database with case sensitive tables.
	Thanks to Marc Cousin for report and patch.
	- Fix missing schema name in query when extract views as tables.

2008 12 04 - version 4.10

	- Fix missing replacement of table name on disable triggers when
	  required.
	- Fix some malformed debug output messages.
	- Add the capability to extract data from view as if it was a table.
	  This is usefull if you want to export/import data from an Oracle
	  view into a Pg table. There's nothing special to do, just to give
	  the view name into the TABLES configuration directive and set TYPE
	  to DATA or COPY. If views are not specified in the TABLES directive
	  there's not view export but only table data.
	- Add capability to extract views structure as table schema. There's
	  nothing special to do, just to give the view name into the TABLES
	  configuration directive and set TYPE to TABLE. This will not extract
	  constraints or other table tuning from table used in the view. Thanks
	  to Groupe SAMSE for the feature request.

2008 10 27 - version 4.9

	- Modify the DISABLE_TABLE_TRIGGERS configuration option. Should be now
	  replaced by DISABLE_TRIGGERS, but compatibility is preserved.
	- Add DISABLE_SEQUENCE configuration option to not export alter
	  sequence after COPY or DATA export.
	- Fix extraction of function based index that appears as SYS_NC....
	  Thanks to Bozkurt Erkut from SONY for the report

2008 09 04 - version 4.8

	- Add SYSUSERS configuration option that allow you to specify a coma
	  separated list of Oracle System user/schema to exclude from extracted
	  object. By default it only exclude user SYS,SYSTEM,DBSNMP,OUTLN and 
	  PERFSTAT
	- Add support to other binary mode output than ':raw' to avoid the Perl
	  error message:"Wide character in print". See the BINMODE configuration
	  directive. This will help a lot if you have UTF-8 records.
	  Thank to Guillaume Demillecamps for the report.
	- Fix double escaping of special character.
	  Thank to Guillaume Demillecamps for the report.

2008 01 25 - version 4.7

	- Add support to regular expressions in the exclusion list. Thanks to
	  Peter Eisentraut
	- Fix misformatted SQL string in function _extract_sequence_info.
	  Thanks to Bernd Helmle.
	- Add escaping of backslash on COPY output. Thanks to Peter Eisentraut

2008 01 03 - version 4.6

	- Applied a patch to add ALTER SEQUENCE statements to the dump to
	  adjust the sequence to the correct values after DATA and COPY dumps.
	  Thanks to Bernd Helmle.
	- Applied a patch which fixes problems with broken COPY output when
	  extracting data from Orace databases with embedded tabs, carriage
	  returns and line feeds. Thanks to Bernd Helmle.
	- Move the project to PgFoundry

2007 06 20 - version 4.5

	- Fix columns order in index extraction. Thanks to Ugo Brunel from BULL.

2007 04 30 - version 4.4

	- Fix missing single quote in role extraction.
	- Add configuration directive NOESCAPE to enable/disable
	  escaping characters during data extraction. Default is enabled.
	- Add TIMESTAMP, BINARY_FLOAT and BINARY_DOUBLE data type translation.
	- Add DATA_TYPE configuration directive to allow user defined data type
	  translation.
	- Add NLS_LANG configuration directive to set Oracle database encoding
	  and enforce a default language-setting in ora2pg.pl. Thanks to Lars
	  Weber

2007 04 03 - version 4.3

	- Fix duplicate view export. Add schema selector to views. Thank to
	  Ugo BRUNEL from BULL for the fix.
	- Remove 'use strict' to prevent failure on certain condition.
	  Thank to Andrea Schnabl for the report.

2006 06 08 - version 4.2

	- Fix a miss taping on constraint type search that convert unique key
	  to primary key. Thank to Ugo BRUNEL (BULL) for the patch.
	- Fix case sensitivity on CHECK constraint that could cause problem when
	  check value is uppercase. Thank to Ugo BRUNEL (BULL) for the patch.

2006 03 28 - version 4.1

	- Fix a problem when using data_limit and where clause. Thank to
	  Rene Bentzen for the patch.
	- Add enable/disable trigger on data import. Thank to Bernd Helmle.
	- Fix escaping of chr(13) MS crashing data import into PG. Thank
	  to Ugo Brunel (BULL).

2006 03 22 - version 4.0

	- Add validation of the requested schema in the database before all.
	  Thanks to Max Walton for the idea.
	- Add multiple export type at the same time. Thanks to Max Walton
	  for the idea.
	- Add support for in/out/inout function parameter. See PG_SUPPORTS_INOUT
	  configuration option. Thanks to Bernd Helmle for this great
	  contribution/patch.
	- Add support for ROLES with Pg v8.1+. See PG_SUPPORTS_ROLE configure
	  option.

2006 02 10 - version 3.4

	This release add better support to Oracle grant, function and grant
	extraction. Great thanks to the Pg team!

	- Add preservation of oracle primary key names. See KEEP_PKEY_NAMES
	  configuration option. Thanks to Antonios Christofides for this patch.
	- Fix bug in case insensitive check constrainte. Thanks to Wojciech
	  Szenajch for the patch.
	- Fix saving data to files correctly (binmod) when the oracle database
	  contains utf8 chars. Thanks to Richard Chen for the report.
	- Fix bug on view extraction when a column contains the word WITH.
	  Thanks to Richard Chen for the patch.
	- Fix wrong mapping between tge data type in Oracle "number(10)" and
	  Postgresql, which should be "integer" and not "bigint". Thanks to
	  Sergio Freire for the patch.
	- Fix bug in EXCLUDE configuration directive parsing. Thanks to Matt
	  Miller for the patch.

2005 02 22 - version 3.3

	- Fix bug "Modification of a read-only value attempted"

2005 02 11 - version 3.2

	- Fix patch error on column position sort
	- Replace 'now' by CURRENT_TIMESTAMP on SYSDATE replacement
	- Fix bytea type that was not quoted.

2005 02 10 - version 3.1

	- Fix bug on deferrable constraint. Thanks to Antonios Christofide for
	  the patch.
	- Fix problem on defer_fkey that should be in a transaction. Thanks to
	  Antonios Christofide for the patch.
	- Add sort by column position during schema extraction.
	- Add support to SYSDATE. Thanks to David Cotter-Alatto Technologies Ltd

2004 12 24 - version 3.0

	- Add 'TABLESPACE' extraction type to create PostgreSQL v8 tablespace.

2004 12 24 - version 2.9

	- Debuging output rewrite. Thanks to Antonios Christofide for help.
	- Add 'PG_NUMERIC_TYPE' configuration option to replace portable
	  numeric type into PostgreSQL internal type (smallint, integer,
	  bigint, real and float).

2004 12 24 - version 2.8

	- Fix/add support to data export of type BLOB, RAW and LONG RAW.
	  Thanks to Antonios Christofide for help.

2004 12 23 - version 2.7

	- Add 'FKEY_DEFERRABLE' configuration option to force foreign key
	  constraints to be exported as deferrable. Thanks to Antonios
	  Christofide for help.
	- Add 'DEFER_FKEY' configuration option to defer all foreign key
	  constraints during data export. Thanks to Antonios Christofide
	  for help.

2004 12 23 - version 2.6

	- Fix duplicate output during export. Thanks to Adriano Bonat for the
	  report.
	- Fix data limit infinite loop during data extraction. Thanks to Thomas
	  REISS for the report.
	- Add 'GEN_USER_PWD' configuration option allowing to generate a random
	  password.  Thanks to Antonios Christofide for help.
	  (Require String::Random from CPAN).
	- Fix USER/ROLES/GRANT extraction problem. Now all users are dumped.
	  All roles are translated to PostgreSQL groups. All grants are
	  exported. YOU MUST EDIT the output file to rewrite real privilege
	  and match your needs. Thanks to Antonios Christofide for help.
	- Fix split COPY export into multiple transaction for large data export.
	  The number of row per transaction is set to 'DATA_LIMIT' value. A
	  value of O mean all in a single transaction.

2004 10 13 - version 2.5

	- Fix extraction problem when the connection to Oracle DB is not as DBA.

2004 08 22 - version 2.4

	- Fix bug in DBI errstr call.
	- Add CASE_SENSITIVE configuration option to allow case sensitivity on
	  Add a new configuration directive 'USER_GRANTS' to do that. Thanks to
	  Octavi Fors for the report.
	  object name. Thanks to Thomas Wegner.
	- Fix major bug in unique keys extraction. Thanks to Andreas Haumer and
	  Marco Lombardo for their great help.
	- Add CHECK constraint extration. Thanks again to Andreas Haumer.
	- Add IMPORT configuration option to include common configuration file
	  throught multiple configuration files.Thanks to Adam Sah and Zedo Inc.
	- Add SKIP configuration option to turning off extraction of certain
	- schema features. Thanks to Adam Sah and Zedo Inc.
	- Fix bug in excluded tables
	- Fix backslash escaping. Thanks to Adam Sah and Zedo Inc.
	- Add REPLACE_TABLES configuration option to change table name during
	  data extraction.
	- Add REPLACE_COLS configuration option to change columns name during
	  data extraction.
	- Add WHERE configuration option to add where clause to each table or
	  specific tables during extraction. Usefull for replication. Thanks
	  to Adam Sah and Zedo Inc.
	- Add progress indicators (per 1000 rows) and performance results
	  during data extraction in debug mod. Thanks to Adam Sah and Zedo Inc.
	- Add Gzip and Bzip2 compress to output file if extension .gz or .bz2.
	  Gzip compress require perl module Compress::Zlib from CPAN. Thanks
	  to Adam Sah for the idea.

2004 04 13 - Version 2.3

	- Fix bug in date/time conversion when using data export limit. Thanks
	  to Andreas Haumer.
	- Add sort order when extracting tables and data to respect the TABLES
	  limited extraction array write order. Usefull if you have foreign key
	  constraints. Thanks to Andreas Haumer for the idea.

2004 04 13 - Version 2.2

	- Add EXCLUDE configuration option to allow table exclusion
	  from all extraction.
	- Fix a bug in escaping single quote on data export.

2004 03 09 - Version 2.1

	- Fix COPY output by replacing special character.
	- Add configuration file usefull for people who don't have Perl in mind
	  Thank's to Tanya Krasnokutsky to force me to do that :-)
	- Fix other minor problem.

2002 12 26 - Version 2.0

	- Clean code.
	- Fix COPY output on column value with EOL and add column naming.
	- Add support to the PostgreSQL 7.3 schema. So Oracle schema can now be
	  exported. (see export_schema init option)
	- Remove data extraction limit (old default: 10) so each tuple will be
	  dump by default.

2002 12 03 - Version 1.12

	I have fixed 2 bugs when using it against Oracle 817R3 on linux.

	- Fix problem regarding RI constraints, the owner name was not
	  getting into the sql statement. Thank to Ian Boston.
	- Moved all the RI constraints out of the create table statement.
	  Thank to Ian Boston for this contribution. This was a major request
	  from Ora2pg users.

2002 09 27 - Version 1.11

	- Fix a problem when retrieving package+package body. Thanks to Mike
	  WILHELM-HILTZ.
	- Set LongReadLen to 100000 when exporting table information. Many
	  users reports this kind of error: A-01406 LongReadLen too small and/or
	  LongTruncOk not set. This should fix the problem else you must
	  increase the value.
	- Filtering by owner for better performance when retreiving database
	  schema. Thanks to Jefferson MEDEIROS.

2002 07 29 - Version 1.10

	- Fix a problem with local settings regarding decimal separator (all ,
	  are changed to .) Thank to Jan Kester.

2002 06 04 - Version 1.9

	- Fix a problem on exporting data which fill NULL instead of 0 or
	  empty string. Thanks to Jan Kester.
	- Add time + date when export data [ tochar('YYYY-MM-DD HH24:MI:SS') ].
	  Thanks to Paolo Mattioli.

2002 03 05 - Version 1.8

	- Add Oracle type FLOAT conversion to float8.
	- Add column alias extraction on view.
	  Thanks to Jean-Francois RIPOUTEAU
	- Add PACKAGE extraction (type => DATA).

2002 02 14 - Version 1.7

	- Remove export of OUTLINE object type. Thanks to Jean-Paul ARGUDO.

2002 01 07 - Version 1.6

	- Fix problem exporting NULL value. Thanks to Stephane Schildknecht.

2001 12 28 - Version 1.5

	- Fix LongReadLen problem when exporting Oracle data on LONG and LOB
	  types. Thanks to Stephane Schildknecht for report and test.
	- Add more precision on NUMBER type conversion
	- Add conversion of type LONG, LOB, FILE
	- Fix a problem when extracting data, sometime table could need to be
	  prefixed by the schema name.
	- Fix output of Oracle data extraction. It now require a call to
	  function export_data().

2001 06 27 - Version 1.4

	- Add online Oracle data extraction and insertion into PG database.
	- Data export as insert statement (type => DATA)
	- Data export as copy from stdin statement (type => COPY)

2001 06 20 - Version 1.3

	- Grant/privilege extraction are now done separatly with option
	  type=>'GRANT'
	- Sequence extraction with the option type=>'SEQUENCE'
	- Trigger extraction with the option type=>'TRIGGER'
	- Function extraction with the option type=>'FUNCTION' and
	  type=>'PROCEDURE'
	- Complete rewrite of the foreign key extraction
	- Fix incorrect type translation and many other bug fix
	- Add schema only extraction by option schema => 'MYSCHEM'

2001 05 11 - Version 1.2

	- Views extraction is now really done with the option type=>'VIEW'
	- Add indexes extraction on tables.
	- Changes name of constraints, default is now used.
	- Add debug printing to see that the process is running :-)
	- Add extraction of only required tablename.
	- Add extraction of only n to n table indice. Indices of extraction
	  can be obtained with the option showtableid set to 1.
	- Fix print of NOT NULL field.
	- Complete rewrite of the grant extraction
	- Complete rewrite of most things

2001 05 09 - Version 1.1

	- Add table grant extraction based on group.
	  Oracle ROLES are exported as groups in PG

2001 05 09 - Initial version 1.0

------------------------------------------------------------------------------

Special thanks to Ali Pouya for documentation review. All my recognition
to Ali Pouya and Olivier Mazain for their great work in the package and
function export. Thanks to Jean-Paul Argudo for the time spent to heavily
testing Ora2Pg.

Special thanks to Josian Larcheveque and Stephane Silly as Oracle DBA
and their "patience".

Special Thanks to Dominique Legendre for his help on Spatial support and
all the tests performed.

Many thanks for all congratulation message, idea and bug report+fix I received.

Very special thanks to Jean-Paul Argudo that represent Ora2Pg at Linux Solution Paris 2005.

Gilles DAROLD <gilles (at) darold (dot) net>