## Rails 3.2.8 (Aug 9, 2012) ##
* Do not consider the numeric attribute as changed if the old value is zero
and the new value is not a string.
Fixes#7237.
*Rafael Mendonça França*
* Removes the deprecation of `update_attribute`. *fxn*
* Reverted the deprecation of `composed_of`. *Rafael Mendonça França*
* Reverted the deprecation of `*_sql` association options. They will be
deprecated in 4.0 instead. *Jon Leighton*
* Do not eager load AR session store. ActiveRecord::SessionStore depends on
the abstract store in Action Pack. Eager loading this class would break
client code that eager loads Active Record standalone.
Fixes#7160
*Xavier Noria*
* Do not set RAILS_ENV to "development" when using `db:test:prepare` and
related rake tasks.
This was causing the truncation of the development database data when using
RSpec.
Fixes#7175.
*Rafael Mendonça França*
Important Change: The YEAR(2) data type is now deprecated because it is problematic. Support for YEAR(2) will be removed in a future release of MySQL. For more information, see Section 11.3.4, “YEAR(2) Limitations and Migrating to YEAR(4)”.
=== 3.37.0 (2012-07-02)
* Allow specifying eager_graph alias base on a per-call basis using an AliasedExpression (jeremyevans)
* Allow bin/sequel to respect multiple -l options for logging to multiple files (jeremyevans)
* Correctly handle cases where SCOPE_IDENTITY is nil in the odbc/mssql adapter (stnoonan, jeremyevans)
* Add pg_interval extension, for returning interval types as ActiveSupport::Duration instances (jeremyevans)
* Save a new one_to_one associated object once instead of twice in the nested_attributes plugin (jeremyevans)
* Don't add unnecessary filter condition when passing a new object to a one_to_one setter method (jeremyevans)
* Differentiate between column references and method references in many_through_many associations (jeremyevans)
* Use :qualify=>:deep option when joining tables in model association datasets (jeremyevans)
* Support :qualify=>:deep option to Dataset#join_table to qualify subexpressions in the expression tree (jeremyevans)
* Support :qualify=>false option to Dataset#join_table to not automatically qualify keys/values (jeremyevans)
* Make filter by associations support use column references and method references correctly (jeremyevans)
* Call super in list plugin before_create (jeremyevans) (#504)
* Do not automatically cast String to text in pg_auto_parameterize extension (jeremyevans)
* Support alter_table validate_constraint on PostgreSQL for validating constraints previously declared with NOT VALID (jeremyevans)
* Support :not_valid option when adding foreign key constraints on PostgreSQL (jeremyevans)
* Support exclusion constraints on PostgreSQL (jeremyevans)
* Allow for overriding the create/alter table generators used per Database object (jeremyevans)
* Make casting to Date/(Time/DateTime) use date/datetime functions on SQLite (jeremyevans)
* Add pg_range_ops extension for DSL support for PostgreSQL range operators and functions (jeremyevans)
* The json library is now required when running the plugin/extension specs (jeremyevans)
* Use change migrations instead of up/down migrations in the schema_dumper (jeremyevans)
* Dump unsigned integer columns with a check >= 0 constraint in the schema_dumper (stu314)
* Switch the :key_hash entry to the association :eager_loader option to use the method symbol(s) instead of the column symbol(s) (jeremyevans)
* Add :id_map entry to the hash passed to the association :eager_loader option, for easier custom eager loading (jeremyevans)
* Fix dumping of non-integer foreign key columns in the schema_dumper (jeremyevans) (#502)
* Add nested_attributes :fields option to be a proc that is called with the associated object (chanks) (#498)
* Add split_array_nil extension, for compiling :col=>[1, nil] to col IN (1) OR col IS NULL (jeremyevans)
* Add Database#extension and Dataset#extension for loading extension modules into objects automatically (jeremyevans)
* Respect an existing dataset limit when updating on Microsoft SQL Server (jeremyevans)
* Add pg_range extension, for dealing with PostgreSQL 9.2+ range types (jeremyevans)
* Make pg_array extension convert array members when typecasting Array to PGArray (jeremyevans)
* Make jdbc/postgres adapter convert array type elements (e.g. date[] arrays are returned as arrays of Date instances) (jeremyevans)
* Make the pg_inet extension handle inet[]/cidr[]/macaddr[] types when used with the pg_array extension (jeremyevans)
* Make the pg_json extension handle json[] type when used with the pg_array extension (jeremyevans)
* Fix schema parsing of h2 clob types (jeremyevans)
* Make the pg_array extension handle array types for scalar types handled by the native postgres adapter (jeremyevans)
* Generalize handling of array types in the pg_array extension, allowing easy support of custom array types (jeremyevans)
* Remove type conversion of int2vector and money types on PostgreSQL, since previous conversions were wrong (jeremyevans)
* Add eval_inspect extension, which makes Sequel::SQL::Expression#inspect attempt to return a string suitable for eval (jeremyevans)
* When emulating offset with ROW_NUMBER, default to ordering by all columns if no specific order is given (stnoonan, jeremyevans) (#490)
* Work around JRuby 1.6 ruby 1.9 mode bug in Time -> SQLTime conversion (jeremyevans)
=== 3.36.1 (2012-06-01)
* Fix jdbc adapter when DriverManager#getConnection fails (aportnov) (#488)
== v0.14.0 [2012-06-17] Michael Granger <ged@FaerieMUD.org>
Bugfixes:
#47, #104
New Methods for PostgreSQL 9 and async API support:
PG
- ::library_version
PG::Connection
- ::ping
- #escape_literal
- #escape_identifier
- #set_default_encoding
PG::Result
- #check
New Samples:
This release also comes with a collection of contributed sample scripts for
doing resource-utilization reports, graphing database statistics,
monitoring for replication lag, shipping WAL files for replication,
automated tablespace partitioning, etc. See the samples/ directory.
v0.12.2
- Respect the start_key in get_range. Resolves Issue #127.
- Fix issue with differences in gemspec and what is required. Resolves
Issue #125.
- Update to Cassandra 0.8.7 and 1.0.2.
## Rails 3.2.7 (unreleased) ##
* `:finder_sql` and `:counter_sql` options on collection associations
are deprecated. Please transition to using scopes.
*Jon Leighton*
* `:insert_sql` and `:delete_sql` options on `has_and_belongs_to_many`
associations are deprecated. Please transition to using `has_many
:through`
*Jon Leighton*
* `composed_of` has been deprecated. You'll have to write your own accessor
and mutator methods if you'd like to use value objects to represent some
portion of your models.
*Steve Klabnik*
* `update_attribute` has been deprecated. Use `update_column` if
you want to bypass mass-assignment protection, validations, callbacks,
and touching of updated_at. Otherwise please use `update_attributes`.
*Steve Klabnik*
## Rails 3.0.16 (Jul 26, 2012)
* No changes.
## Rails 3.0.14 (Jun 12, 2012)
* protect against the nesting of hashes changing the
table context in the next call to build_from_hash. This fix
covers this case as well.
CVE-2012-2695
Upstream ChangLog:
1.65 Mon Jul 23 20:16:08 PDT 2012
Finish the "Perl Data::Table Cookbook", should be a good learning material.
To download, visit https://sites.google.com/site/easydatabase/
Polish Data::Table::Excel for CPAN upload.
Minor patches to the code.
1.64 Sun Jul 8 22:01:17 PDT 2012
Add $keepRestCols to Data::Table::group();
We introduce new constants for fromCSV/fromTSV/fromFile/csv/tsv.
Data::Table::OS_UNIX = 0;
Data::Table::OS_PC = 1;
Data::Table::OS_MAC = 2;
Add method reorder(), redefine column orders
Add method melt() and cast(), concept borrowed from Reshape package in R
Add method each_group(), so one can apply a custom method to rows sharing the same key
Made a seemingly backward incompatible change to pivot()
pivot($colToSplit, $colToSplitIsNumeric, ...) is changed to
pivot($colToSplit, $colToSplitIsStringOrNumber, ...)
What is now pivot($colToSplit, $Data::Table::STRING, ...), where Data::Table::STRING has a value of 1,
was equivalent to pivot($colToSplit, 0, ...) in <= 1.63.
However, the $colToSplitIsStringOrNumber is now auto-guessed within the code, so the change is not very relevant.
Most existing code should run fine, without change.
Patch group(), piviot() to distinguish keys between empty string and undef.
Patch subTable() to take row mask array when {useRowMask=>1} is provided.
1.63 Tue Jun 12 17:05:43 PDT 2012
In this release, we patch addCol, delCol, addRow, rowMerge, colMerge to for an empty table
We introduce new methods isEmpty(), hasCol(), moveCol($colID, $newColIdx)
We introduce new constants for Data::Table::new()
Data::Table::ROW_BASED
Data::Table::COL_BASED
1.62 Fri May 25 11:40:09 PDT 2012
In this release, we address a few pain points
Data::Table::colMerge, update to support new options
{ renameCol => 1}
If specified, duplicate column names in the second table is automatically renamed (by appending _2) to avoid conflict
We introduce some constants, so we have fewer numbers to remember.
Data::Table::NUMBER
Data::Table::STRING
Data::Table::ASC
Data::Table::DESC
for sort(), you can use $t->sort('col2', Data::Table::NUMBER, Data::Table::DESC); it is equivalent to $t->sort('col2', 0, 1);
Data::Table::INNER_JOIN
Data::Table::LEFT_JOIN
Data::Table::RIGHT_JOIN
Data::Table::FULL_JOIN
for join(), you may use $t->sort($t2, Data::Table::FULL_JOIN, ['col1'], ['col1']);
it is equivalent to $t->sort($t2, 3, ['col1'], ['col1']).
match_string, match_pattern have been generating @Data::Table::OK, which is a class-level array.
$t->match_pattern() will now also store the results (array ref) in $t->{OK}, that should be used in the future.
However, @Data::Table::OK is still supported for compatibility reasons.
This is not a pain point, but conceptually nicer to be localized.
match_pattern_hash() is added. The difference is each row is fed to the pattern as a hash %_. In the case of
match_pattern, each row is fed as an array ref $_. The pattern for match_pattern_hash() becomes much cleaner.
If a table has two columns: Col_A as the 1st column and Col_B as the 2nd column, a filter "Col_A>2 AND Col_B<2"
is written before as
$t->match_pattern('$_->[0] > 2 && $_->[1] <2');
where we need to figure out $t->colIndex('Col_A') is 0 and $t->colIndex('Col_B') is 1, in order to build the pattern.
Now you can use column name directly in the pattern:
$t->match_pattern_hash('$_{Col_A} >2 && $_{Col_B} <2');
This method creates $t->{OK}, as well as @Data::Table::OK, same as match_pattern().
Data::Table::rowMerge, update to support new options
{ byName =>1, addNewCol => 1}
If byName is 1, rows in the second table are appended by matching their column names, so that the second table
can have columns in a different order.
If addNewCol is 1, columns not exist in the first table will be automatically added.
addNewCol is best used with byName. If used alone, addNewCol will just patch the two tables so that they have
the same number of columns.
Data::Table::subTable, update internal to remove side effect on column header array
Data::join add support for an option {renameCol => 1}.
If specified, duplicate column names in the second table is automatically renamed (by appending _2) to avoid conflict
1.61 Mon Feb 27 21:07:55 PST 2012
Data::Table::fromSQL now can take DBI::st instead of a SQL string. This is introduced, so that
variable binding (such as CLOB/BLOB) can be done outside the method.
1.60 Sat Feb 25 19:26:46 PST 2012
Data::Table::addRow now also can take a hash reference. Hash keys are column names,
undef will be the value, if a column name is not found in the hash.
Suggested by Federico
1.59 Sun Feb 5 00:20:00 PST 2012
I have never checked those CPAN ticket, happened to discover them and address them in this version.
Update document, explain Data::Table::fromCSV(\*STDIN, 1) can be used to read table from STDIN.
Add tbody and thead to Data::Table::html, if it's portrait.
Suggested by Ken Rosenberry.
Modify Data::Table::html and Data::Table::html2, so that it can accept coloring via CSS
The color now can be either specified as an array as before, or as three CSS class names
Suggested by Xavier Robin
1.58 Thu Feb 2 20:33:03 PST 2012
Patch join(), prior version of join considers two NULL keys to be equal
update document, clarify that rowMerge assumes table columns in the same order
Thanks to Ulrik Stervbo.
1.57 Thu Apr 23 15:22:36 PDT 2009
Patch pivot(), it throws warning before, when colToFill is undef.
1.56 Fri Aug 22 15:53:29 PDT 2008
When the first line in a TSV is not a header, but contains strings such as \t.
The program will not transform \t to a tab.
Modify fromTSV, so that \t, \N (etc) transformation is optional.
Add transform_element flag to fromTSV method to turn on/off the transformation.
Thanks to Bin Zhou.
1.55 Mon May 5 10:29:44 PDT 2008
Patch parseCSV. fromFile guesses the wrong delimiter if some ending columns are empty.
Version 5.0.4
-------------
Released: 22th March 2012
Bug fixes.
* Fix#3468882 "UPDATE single row with varchar key: empty where clause", reported by "Dirk Kraemer"
* Fix XSS in function.php, reported by Mateusz Goik
* Fix bug where renaming a tablespace was leading to an error
* Fix a bug where clicking to "Show all schemas" on schema privilege page leads back to the latest tab used in database level
* Fix a bug when executing a query from the history
* Fix bad confirmation message when droping an autovacuum table setup
* Fix bug #3429633 '"Back" link from "Browse" leads to error'
* Fix a PHP warning when Slony conf parameters are not set
* Show cancel/kill actions in process page only if role is a superuser
* Some more small code cleanup and fixes.
Version 5.0.3
-------------
Released: 3rd October 2011
Some bug fixes.
* Fix bug with enable/disable triggers on multiple triggers (Dawid 'DeyV' Polak)
* Fix bug #3353670: Error when executing a report with paging
* Fix bug with OpenBSD about where setting application_name using PGOPTIONS was raising an ERROR
* Two security fix about code injection, reported by Mateusz Goik, fixed by ioguix
* Fix character-encoding problem with autocomplete
- bug #3521416 [interface] JS error when editing index
- bug #3521313 [core] Call to undefined function __()
- bug #3521016 [edit] NOW() function incorrectly selected
- bug [GUI] Invalid HTML code on transformation_overview.php
- bug #3522930 [browse] Missing validation in Ajax mode
- bug Fix popup message on build SQL of import
- bug #3523499 [core] Make X-WebKit-CSP work better
- replace Highcharts with jqplot for query profiling, zoom search
- bug #3531584 [interface] No form validation in change password dialog
- bug #3531585 [interface] Broken password validation in copy user form
- bug #3531586 [unterface] Add user form prints JSON when user presses enter
- bug #3534121 [config] duplicate line in config.sample.inc.php
- bug #3534311 [interface] Grid editing incorrectly parses ENUM/SET values
- bug #3510196 [core] More clever URL rewriting with ForceSSL
Due to updated PKG_DEVELOPER checks, postgresql83-server started failing
with the following message:
ERROR: lib/postgresql/plpgsql.so: missing libintl.so.8
The plpgsql.so library had no rpath set other than what the base compiler
adds by default. On DragonFly, the libintl.so library would likely have
been found anyway as /usr/pkg/lib is part of the LD_CONFIG hint file
search path, but that isn't a guarantee.
The new patch adds ${libdir} to the plpgsql.so rpath so it passes
PKG_DEVELOPER checks now.
Due to updated PKG_DEVELOPER checks, postgresql84-server started failing
with the following message:
ERROR: lib/postgresql/plpgsql.so: missing libintl.so.8
The plpgsql.so library had no rpath set other than what the base compiler
adds by default. On DragonFly, the libintl.so library would likely have
been found anyway as /usr/pkg/lib is part of the LD_CONFIG hint file
search path, but that isn't a guarantee.
The new patch adds ${libdir} to the plpgsql.so rpath so it passes
PKG_DEVELOPER checks now.
Changes:
* autoconf 2.68 -> 2.69, automake 1.11.1 -> 1.12.2,
libtool 2.2.10 -> 2.4.2.
* Corrected dereference of NULL pointer in db_deinit_limit()
and db_deinit_threshold() introduced in the 1.0.2 version.
* Converted all flags to bool data type.
Provided by Andrey Simonenko in PR 46692.
from 1.35nb1 to 1.37.
Upstream changes (since 1.35):
1.37 to be released
- Updated to SQLite 3.7.12.1 (ISHIGAKI)
1.36_04 Sat 19 May 2012
- Final developer release
- Updated to SQLite 3.7.12 (ISHIGAKI)
- Tweaked Makefile.PL to behave better during the Bsymbolic
check (HMBRAND)
- Added SQLITE_WITHOUT_ZONEMALLOC for older MacOS X (ISHIGAKI)
1.36_03 Mon 7 May 2012
- Updated to SQLite 3.7.11 (ISHIGAKI)
- Fix >32bit integer truncation and other sqlite_set_result
condition issue (Yuriy Kaminskiy)
- Fix integer overflow in passing argument to perl function
(Yuriy Kaminskiy)
- Convert unsigned -> int64 when possible (Yuriy Kaminskiy)
- Turned datatype mismatch error (introduced in 1.34_02) into
a warning (you can disable this warning by setting PrintWarn
attribute to false). (ISHIGAKI)
- Refactored sqlite_is_number to fix various corner cases
(ISHIGAKI)
1.36_02 Thu 23 Feb 2012
- Downgraded SQLite to 3.7.9, as 3.7.10 turned out to be
broken on the latest MacOS X (due to a missing symbol),
and broke other modules that typically use temporary tables
under a few environments too. As of this writing, would-be
3.7.11 seems fine, but it would take another month to be
released. (ISHIGAKI)
1.36_01 Thu 19 Jan 2012
*** CHANGES THAT MAY POSSIBLY BREAK YOUR OLD APPLICATIONS ***
- Updated to SQLite 3.7.10 (ISHIGAKI)
Note that this release changed the default schema format
number, that means newly created database files will be
unreadable by SQLite version prior to 3.3.0 (2006-01-10)
(or DBD::SQLite prior to 1.12) unless you explicitly issue
"PRAGMA legacy_file_format=ON".
- Enabled SQLITE_ENABLE_FTS4
- Enabled SQLITE_ENABLE_STAT3
- Resolved#73159: FTS tokenizer segfault (ISHIGAKI)
- Resolved#73787: sqlite_see_if_its_a_number causes a buffer
overflow (ISHIGAKI)
- Resolved#73314 for DBD-SQLite: binding of 64bit integers fail
on 1.34_02 (ISHIGAKI)
- Implemented sqlite_trace and sqlite_profile methods for simpler
tracing/profiling; use DBI_TRACE/DBI_PROFILE for more
complicated cases (ISHIGAKI)
1.622.
pkgsrc changes:
- fixing download directory
Upstream changes:
Changes in DBI 1.622 (svn r15327) 6th June 2012
Fixed lack of =encoding in non-ASCII pod docs. RT#77588
Corrected typo in DBI::ProfileDumper thanks to Finn Hakansson.
Changes in DBI 1.621 (svn r15315) 21st May 2012
Fixed segmentation fault when a thread is created from
within another thread RT#77137, thanks to Dave Mitchell.
Updated previous Changes to credit Booking.com for sponsoring
Dave Mitchell's recent DBI optimization work.
Changes:
* Fix incorrect password transformation in contrib/pgcrypto's DES crypt()
function (Solar Designer)
* Ignore SECURITY DEFINER and SET attributes for a procedural language's call
handler (Tom Lane)
* Make contrib/citext's upgrade script fix collations of citext arrays and
domains over citext (Tom Lane)
* Allow numeric timezone offsets in timestamp input to be up to 16 hours away
from UTC (Tom Lane)
* Fix timestamp conversion to cope when the given time is exactly the last DST
transition time for the current timezone (Tom Lane)
* Fix text to name and char to name casts to perform string truncation
correctly in multibyte encodings (Karl Schnaitter)
* Fix memory copying bug in to_tsquery() (Heikki Linnakangas)
* Ensure txid_current() reports the correct epoch when executed in hot standby
(Simon Riggs)
* Fix planner's handling of outer PlaceHolderVars within subqueries (Tom Lane)
* Fix planning of UNION ALL subqueries with output columns that are not simple
variables (Tom Lane)
* Fix slow session startup when pg_attribute is very large (Tom Lane)
* Ensure sequential scans check for query cancel reasonably often (Merlin
Moncure)
* Ensure the Windows implementation of PGSemaphoreLock() clears
ImmediateInterruptOK before returning (Tom Lane)
* Show whole-row variables safely when printing views or rules (Abbas Butt, Tom
Lane)
* Fix COPY FROM to properly handle null marker strings that correspond to
invalid encoding (Tom Lane)
* Fix EXPLAIN VERBOSE for writable CTEs containing RETURNING clauses (Tom Lane)
* Fix PREPARE TRANSACTION to work correctly in the presence of advisory locks
(Tom Lane)
* Fix truncation of unlogged tables (Robert Haas)
* Ignore missing schemas during non-interactive assignments of search_path (Tom
Lane)
* Fix bugs with temporary or transient tables used in extension scripts (Tom
Lane)
* Ensure autovacuum worker processes perform stack depth checking properly
(Heikki Linnakangas)
* Fix logging collector to not lose log coherency under high load (Andrew
Dunstan)
* Fix logging collector to ensure it will restart file rotation after receiving
SIGHUP (Tom Lane)
* Fix "too many LWLocks taken" failure in GiST indexes (Heikki Linnakangas)
* Fix WAL replay logic for GIN indexes to not fail if the index was
subsequently dropped (Tom Lane)
* Correctly detect SSI conflicts of prepared transactions after a crash (Dan
Ports)
* Avoid synchronous replication delay when committing a transaction that only
modified temporary tables (Heikki Linnakangas)
* Fix error handling in pg_basebackup (Thomas Ogrisegg, Fujii Masao)
* Fix walsender to not go into a busy loop if connection is terminated (Fujii
Masao)
* Fix memory leak in PL/pgSQL's RETURN NEXT command (Joe Conway)
* Fix PL/pgSQL's GET DIAGNOSTICS command when the target is the function's
first variable (Tom Lane)
* Ensure that PL/Perl package-qualifies the _TD variable (Alex Hunsaker)
* Fix PL/Python functions returning composite types to accept a string for
their result value (Jan Urbanski)
* Fix potential access off the end of memory in psql's expanded display (\x)
mode (Peter Eisentraut)
* Fix several performance problems in pg_dump when the database contains many
objects (Jeff Janes, Tom Lane)
* Fix memory and file descriptor leaks in pg_restore when reading a
directory-format archive (Peter Eisentraut)
* Fix pg_upgrade for the case that a database stored in a non-default
tablespace contains a table in the cluster's default tablespace (Bruce Momjian)
* In ecpg, fix rare memory leaks and possible overwrite of one byte after the
sqlca_t structure (Peter Eisentraut)
* Fix contrib/dblink's dblink_exec() to not leak temporary database connections
upon error (Tom Lane)
* Fix contrib/dblink to report the correct connection name in error messages
(Kyotaro Horiguchi)
* Fix contrib/vacuumlo to use multiple transactions when dropping many large
objects (Tim Lewis, Robert Haas, Tom Lane)
* Update time zone data files to tzdata release 2012c for DST law changes in
Antarctica, Armenia, Chile, Cuba, Falkland Islands, Gaza, Haiti, Hebron,
Morocco, Syria, and Tokelau Islands;
Changes:
* Fix incorrect password transformation in contrib/pgcrypto's DES crypt()
function (Solar Designer)
* Ignore SECURITY DEFINER and SET attributes for a procedural language's call
handler (Tom Lane)
* Allow numeric timezone offsets in timestamp input to be up to 16 hours away
from UTC (Tom Lane)
* Fix timestamp conversion to cope when the given time is exactly the last DST
transition time for the current timezone (Tom Lane)
* Fix text to name and char to name casts to perform string truncation
correctly in multibyte encodings (Karl Schnaitter)
* Fix memory copying bug in to_tsquery() (Heikki Linnakangas)
* Ensure txid_current() reports the correct epoch when executed in hot standby
(Simon Riggs)
* Fix planner's handling of outer PlaceHolderVars within subqueries (Tom Lane)
* Fix slow session startup when pg_attribute is very large (Tom Lane)
* Ensure sequential scans check for query cancel reasonably often (Merlin
Moncure)
* Ensure the Windows implementation of PGSemaphoreLock() clears
ImmediateInterruptOK before returning (Tom Lane)
* Show whole-row variables safely when printing views or rules (Abbas Butt, Tom
Lane)
* Fix COPY FROM to properly handle null marker strings that correspond to
invalid encoding (Tom Lane)
* Ensure autovacuum worker processes perform stack depth checking properly
(Heikki Linnakangas)
* Fix logging collector to not lose log coherency under high load (Andrew
Dunstan)
* Fix logging collector to ensure it will restart file rotation after receiving
SIGHUP (Tom Lane)
* Fix WAL replay logic for GIN indexes to not fail if the index was
subsequently dropped (Tom Lane)
* Fix memory leak in PL/pgSQL's RETURN NEXT command (Joe Conway)
* Fix PL/pgSQL's GET DIAGNOSTICS command when the target is the function's
first variable (Tom Lane)
* Fix potential access off the end of memory in psql's expanded display (\x)
mode (Peter Eisentraut)
* Fix several performance problems in pg_dump when the database contains many
*bjects (Jeff Janes, Tom Lane)
* Fix pg_upgrade for the case that a database stored in a non-default
tablespace contains a table in the cluster's default tablespace (Bruce Momjian)
* In ecpg, fix rare memory leaks and possible overwrite of one byte after the
sqlca_t structure (Peter Eisentraut)
* Fix contrib/dblink's dblink_exec() to not leak temporary database connections
upon error (Tom Lane)
* Fix contrib/dblink to report the correct connection name in error messages
(Kyotaro Horiguchi)
* Fix contrib/vacuumlo to use multiple transactions when dropping many large
*bjects (Tim Lewis, Robert Haas, Tom Lane)
* Update time zone data files to tzdata release 2012c for DST law changes in
Antarctica, Armenia, Chile, Cuba, Falkland Islands, Gaza, Haiti, Hebron,
Morocco, Syria, and Tokelau Islands;
Changes:
* Fix incorrect password transformation in contrib/pgcrypto's DES crypt()
function (Solar Designer)
* Ignore SECURITY DEFINER and SET attributes for a procedural language's call
handler (Tom Lane)
* Allow numeric timezone offsets in timestamp input to be up to 16 hours away
from UTC (Tom Lane)
* Fix timestamp conversion to cope when the given time is exactly the last DST
transition time for the the current timezone (Tom Lane)
* Fix text to name and char to name casts to perform string truncation
correctly in multibyte encodings (Karl Schnaitter)
* Fix memory copying bug in to_tsquery() (Heikki Linnakangas)
* Fix planner's handling of outer PlaceHolderVars within subqueries (Tom Lane)
* Fix slow session startup when pg_attribute is very large (Tom Lane)
* Ensure sequential scans check for query cancel reasonably often (Merlin
Moncure)
* Ensure the Windows implementation of PGSemaphoreLock() clears
ImmediateInterruptOK before returning (Tom Lane)
* Show whole-row variables safely when printing views or rules (Abbas Butt, Tom
Lane)
* Fix COPY FROM to properly handle null marker strings that correspond to
invalid encoding (Tom Lane)
* Ensure autovacuum worker processes perform stack depth checking properly
(Heikki Linnakangas)
* Fix logging collector to not lose log coherency under high load (Andrew
Dunstan)
* Fix logging collector to ensure it will restart file rotation after receiving
SIGHUP (Tom Lane)
* Fix WAL replay logic for GIN indexes to not fail if the index was
subsequently dropped (Tom Lane)
* Fix memory leak in PL/pgSQL's RETURN NEXT command (Joe Conway)
* Fix PL/pgSQL's GET DIAGNOSTICS command when the target is the function's
first variable (Tom Lane)
* Fix potential access off the end of memory in psql's expanded display (\x)
mode (Peter Eisentraut)
* Fix several performance problems in pg_dump when the database contains many
objects (Jeff Janes, Tom Lane)
* Fix contrib/dblink's dblink_exec() to not leak temporary database connections
upon error (Tom Lane)
* Fix contrib/dblink to report the correct connection name in error messages
(Kyotaro Horiguchi)
* Update time zone data files to tzdata release 2012c for DST law changes in
Antarctica, Armenia, Chile, Cuba, Falkland Islands, Gaza, Haiti, Hebron,
Morocco, Syria, and Tokelau Islands; also historical corrections for Canada.
Changes:
* Fix incorrect password transformation in contrib/pgcrypto's DES crypt() function (Solar Designer)
* If a password string contained the byte value 0x80, the remainder of the password was ignored, causing the password to be much weaker than it appeared. With this fix, the rest of the string is properly included in the DES hash. Any stored password values that are affected by this bug will thus no longer match, so the stored values may need to be updated. (CVE-2012-2143)
* Ignore SECURITY DEFINER and SET attributes for a procedural language's call handler (Tom Lane)
* Applying such attributes to a call handler could crash the server. (CVE-2012-2655)
* Allow numeric timezone offsets in timestamp input to be up to 16 hours away from UTC (Tom Lane)
* Some historical time zones have offsets larger than 15 hours, the previous limit. This could result in dumped data values being rejected during reload.
* Fix timestamp conversion to cope when the given time is exactly the last DST transition time for the current timezone (Tom Lane)
* This oversight has been there a long time, but was not noticed previously because most DST-using zones are presumed to have an indefinite sequence of future DST transitions.
* Fix text to name and char to name casts to perform string truncation correctly in multibyte encodings (Karl Schnaitter)
* Fix memory copying bug in to_tsquery() (Heikki Linnakangas)
* Fix slow session startup when pg_attribute is very large (Tom Lane)
* If pg_attribute exceeds one-fourth of shared_buffers, cache rebuilding code that is sometimes needed during session start would trigger the synchronized-scan logic, causing it to take many times longer than normal. The problem was particularly acute if many new sessions were starting at once.
* Ensure sequential scans check for query cancel reasonably often (Merlin Moncure)
* A scan encountering many consecutive pages that contain no live tuples would not respond to interrupts meanwhile.
* Ensure the Windows implementation of PGSemaphoreLock() clears ImmediateInterruptOK before returning (Tom Lane)
* This oversight meant that a query-cancel interrupt received later in the same query could be accepted at an unsafe time, with unpredictable but not good consequences.
* Show whole-row variables safely when printing views or rules (Abbas Butt, Tom Lane)
* Corner cases involving ambiguous names (that is, the name could be either a table or column name of the query) were printed in an ambiguous way, risking that the view or rule would be interpreted differently after dump and reload. Avoid the ambiguous case by attaching a no-op cast.
* Ensure autovacuum worker processes perform stack depth checking properly (Heikki Linnakangas)
* Previously, infinite recursion in a function invoked by auto-ANALYZE could crash worker processes.
* Fix logging collector to not lose log coherency under high load (Andrew Dunstan)
* The collector previously could fail to reassemble large messages if it got too busy.
* Fix logging collector to ensure it will restart file rotation after receiving SIGHUP (Tom Lane)
* Fix PL/pgSQL's GET DIAGNOSTICS command when the target is the function's first variable (Tom Lane)
* Fix several performance problems in pg_dump when the database contains many objects (Jeff Janes, Tom Lane)
* pg_dump could get very slow if the database contained many schemas, or if many objects are in dependency loops, or if there are many owned sequences.
* Fix contrib/dblink's dblink_exec() to not leak temporary database connections upon error (Tom Lane)
* Update time zone data files to tzdata release 2012c for DST law changes in Antarctica, Armenia, Chile, Cuba, Falkland Islands, Gaza, Haiti, Hebron, Morocco, Syria, and Tokelau Islands; also historical corrections for Canada.