Changes from previous:
Version 1.33, released February 05th, 2011
----------------------------------------------
[Bug fixes]
* Move test dependencies to (unreleased) Bundle::Test::SQL::Statement
to avoid circular dependencies
Version 1.32, released January 19th, 2011
----------------------------------------------
[Bug fixes]
* Fixed invalid check for escaped single quotes
* Fixed unpermitted modification of array source for table creation
(CREATE TABLE AS IMPORT(?),[[..],[..]])
* Fixing alias used in ORDER BY (RT#61384, thanks jvm)
* Fixing ORDER BY behavior for multiple sort columns
(slower, but guaranteed correct)
[Improvements]
* renamed fetch-method into fetch_row (keep fetch() as alias) and
add a fetch_rows() to fetch all rows at once
* Different accessors for direction of ORDER BY clause query part
and it's boolean equivalent "desc" (0 or 1, respectively)
* Add a lot of Pure-Perl DBD's as build dependency for testing
(skip DBD::AnyData for now, because it seems to be broken - check
for next release)
[Misc]
* Bump requirement of DBI to 1.616
* switch for fully external DBD tests from DBD::XBase to DBD::SQLite
* Document another limitation (lacking implicit creating temp table
during processing a query using the same table with different aliases
twice)
to trigger/signal a rebuild for the transition 5.10.1 -> 5.12.1.
The list of packages is computed by finding all packages which end
up having either of PERL5_USE_PACKLIST, BUILDLINK_API_DEPENDS.perl,
or PERL5_PACKLIST defined in their make setup (tested via
"make show-vars VARNAMES=..."), minus the packages updated after
the perl package update.
sno@ was right after all, obache@ kindly asked and he@ led the
way. Thanks!
Upstream changes:
Version 1.31, released August 16th, 2010
----------------------------------------------
[Bug fixes]
* Fix misbehaviour of DELETE/UPDATE for tables with only basic capabilities
and no matching where clause (reported by H.Merijn Brand and Drew ...)
* Fix some column identifier splits to respect quoted tables
[Improvements]
* Optimized some core routines between 25% and 50%.
Upstream changes:
Version 1.30, released August 1st, 2010
----------------------------------------------
[Bug fixes]
* remove blib directory from distfile
Version 1.29, released August 1st, 2010
----------------------------------------------
[Bug fixes]
* add some getters as documented in SQL::Statement::Structure
(fixes RT#59834, thanks John Wiersba)
* add missing import of function croak to SQL::Statement::Term::ColumnValue
* fix assignment of parser result (doesn't run with perl-5.13.3)
pkgsrc changes:
- require DBI-1.612 for testing
- add conflict for DBI<=1.612
Upstream changes:
Version 1.28, release July 15th, 2010
----------------------------------------------
[Improvements]
* Introduce new "capability" method for SQL::Statement and SQL::Eval::Table
+ Add capability for "insert_new_row" to allow DBD::DBM to fix PK
constrain on INSERT statements.
* Performance of IMPORT feature improved (thanks to Sven Probst, RT#57322)
[Bug fixes]
* expect every table object being derived from SQL::Eval::Table
* rewrite DELETE and UPDATE command based on table capabilities
* add abstract methods for all methods derived classes must override
(this means, open_table for SQL::Statement deriveds must be overridden
and all data access methods of tables - see SQL::Eval::Table for details)
* Tests are fixed to use TEMP TABLES explicitely when required
* check for invalid column names fixed
* Don't let depreciated parser structures stay alive in SQL::Statement when
reusing the Parser
[Documentation]
* Method documentation of SQL::Statement and SQL::Eval::Table are improved
* Add a Roadmap describing future plans for SQL::Statement (in addition to
DBD::File::Roadmap).
* POD spelling fixes provided by H.Merijn Brand and Pod::Spell::CommonMistakes
(thanks Tux)
* POD grammar fixes and reasonable sentences created by Martin Evans
[Things that may break your code]
* SQL::Statement 1.28 is expected not to work proper in combination with
DBI 1.611 and below
* SQL::Statement::ColumnValue expects now every table being derived from
SQL::Eval::Table
Upstream changes:
Version 1.27, release May 06th, 2010
----------------------------------------------
[Bug fixes]
* use originally given table name for open_table() on SELECT, too
(all other command still use the originally given name)
* warn/die -> carp/croak
* fix delete_one_row & update_one_row ability using
[Documentation]
* Apply spelling fix patch from Ansgar Burchardt (RT#56475)
pkgsrc changes:
- Adjust dependencies (mark fullfilled by perl CORE)
Upstream changes:
Version 1.25, release March 15th, 2010
----------------------------------------------
[Bug fixes]
* Keep org_table_names with schema information to allow derived
table classes to handle as it seems reasonable there
* Separate columns with "\0" in multi-column aggregation to be able to
difference between ('1','1foo') and ('11','foo')
[Misc]
* More resources added to META.yml
Version 1.24, release March 15th, 2010
----------------------------------------------
[Misc]
* Ignore *.rej in MANIFEST.SKIP (fixes RT #52081 reported by
Lars Thegler)
[Bug fixes]
* Add missing import of _INSTANCE in SQL::Statement::Function::NumericEval
fixes RT #52356 - reported by Detlef Pilzecker)
* Fix wrongly discarded DISTINCT clause (RT#53186)
[Improvements]
* Fix parsing errors of plain numbers (RT#16931)
* Fix parsing errors of nested calculation / functions (RT#16931, RT#52356)
* Rewrite result calculation of aggregation functions (simplify code,
speed up)
* Upgrade Makefile.PL (patch from Alexandr Ciornii) to handle different
EU::MM versions and abilities properly
* Update documentation to show how 'column_defs' and SQL::Statement::Term
instances shall be used
* Introduce SQL::Dialect::Role providing ini-style data access to
SQL::Dialects (patch from Michael Schwern)
[Things that may break your code]
* SQL::Parser now didn't deliver a struct containing 'column_names',
'computed_columns' and 'set_functions' - it's combined into one member
'column_defs'.
* Minimum required perl version is now 5.8 - upcoming next version of DBI
requires perl 5.8, too - and I could simplify some code that's why
Version 1.23, release November 20th, 2009
----------------------------------------------
[Misc]
* Applied patch from Marc Espie which fixes several orthographic
errors in SQL::Statement::Syntax documentation.
* Added a fixed version of test reported via RT #34121
[Bug fixes]
* Fix an issue in UPDATE command which 'shift's the values from
the list of parameters which causes there're no more parameters
left after first row get's updated (Fixes RT #50788)
* Fix aggregate function handling of new code since 1.21_01
* Correct handling of DISTINCT in aggregate functions
[Improvements]
* Add support for tables/columns starting with '_' for CSV and AnyData,
which is usually forbidden by ANSI SQL
* Add support for inserting multiple lines with one statement
(fixes RT #31730)
* Handle ANSI 'IS NULL' and CVS/AnyData 'IS NULL' different
[Things that may break your code]
* row_value now expects up to two arguments
pkgsrc changes:
- Adding license definition
- Adjusting dependencies
- Ensure build
- remove *.orig removal (may break some SunOS builds)
Upstream changes:
Version 1.22, release October 10th, 2009
----------------------------------------------
[Misc]
* Add missing changelog - no code changes
Version 1.21, release October 10th, 2009
----------------------------------------------
[Misc]
* remove version dependency to check previously installed version
* add DBD::File as "Test" requirement
Version 1.21_8, release October 5th, 2009
----------------------------------------------
[Bug fixes]
* Add additional test for bug-fix in 1.21_7 to t/18bigjoin.t
[Misc]
* Correct some typo's in POD
Version 1.21_7, not public released
----------------------------------------------
[Bug fixes]
* Don't fail for non-existent columns introduced by functions
in joins
Version 1.21_6, release September 24th, 2009
----------------------------------------------
[Bug fixes]
* Don't abort Makefile.PL when in automated smoke tests
Version 1.21_5, release September 23th, 2009
----------------------------------------------
[Bug fixes]
* table order isn't wrongly used in order of appearance when SQL::Parser
couldn't determine an order
* Some internal fixes
[Misc]
* Updated dependency to Params::Util to non-leaking 1.00
* Note dependency to Carp and Data::Dumper
* rely on version to compare versions
* Update POD for terms
* Update
Version 1.21_4, release September 21th, 2009
----------------------------------------------
[Bug fixes]
* modify regex to match types to fix problems with Perl 5.6.2
* add DESTROY methods to ensure clean up
* fix lower casing internal table names when joining tables
* replace parameter shifting by assigning @_ to the list of parameters
Version 1.21_3, release September 17th, 2009
----------------------------------------------
[Things that may break your code]
* When someone accesses the where_clause attribute of the SQL::Statement
instance - be aware that now IN and BETWEEN can be native entries
[Bug fixes]
* convert operation to upper case when surely initialized
[Improvements]
* IN and BETWEEN are now native operations - they are not expanded
to OR'ed equalize operations anymore
Version 1.21_2, release September 15th, 2009
----------------------------------------------
[Things that may break your code]
* modify behavior for unquoted identifiers - they're converted and
returned lower cased now (instead upper cased as in 1.21_1)
Fixes bug RT #48502
Version 1.21_1, release July 30th, 2009
------------------------------------------
[Things that may break your code]
* removed SQL::Statement::Column
* don't instantiate SQL::Statement::Functions objects
* Reworked internal column and function handling to reduce code complexity
* rows and columns aren't setable from outside a table or eval object
anymore
[Bug fixes and other changes]
* Fixed bugs:
+ RT #47292: Test failures with recent DBI
+ RT #44512: Patch for CREATE TABLE parsing
+ RT #42676: tests 16 failed
- Updating module to 1.20
Upstream changes:
Version 1.20, released March 5th, 2009
------------------------------------------
* Fixed Makefile in MANIFEST (reported by Havard Eidnes in RT #43586)
* Fixed invalid label FETCHROW used (reported by Michael in RT #42982)
* separated update_one_row and update_specific_row method names for tables
to avoid confusion
Auto-Oked by rhaen@, because I maintain the CPAN module
Pkgsrc changes:
o Added new needed dependencies
o Removed no-longer-needed patch-aa
Upstream changes:
Version 1.19, released February 6th, 2009
------------------------------------------
* Fixed OUTER JOIN behaviour
* Added version info to all *.pm files to allow CPAN::Reporter find updates
* Correct META-Files
* Fixed reported bugs:
Version 1.18_02, no public release
-----------------------------------
Additional (profiled) optimizations for complicated where clauses
Version 1.18_01, released January 12th, 2009
---------------------------------------------
No code changes within SQL::Statement - but deliver and execute additional
tests.
Pkgsrc changes:
o Update patch to Makefile.PL to avoid reading stdin.
Upstream changes: (Nothing marked 1.17 from upstream...)
Version 1.16_04, released 4 January, 2009
------------------------------------------
* added tests to prove valid quoting (most of them fail)
* Reformat the source
* add some (profiled) tweaks as removing useless regex to speed up SELECT
* Fixed Bugs:
* 14217 Does not correctly handle SQL statements with comments
* 15686 Join syntax is case-sensitive, and common columns in natural
joins are "ambiguous" [patch]
* 13080 Cannot update a field based on its previous value
* 26058 functions on computed columns aliased to the underlying column
name are not called
Version 1.16_03, released 1 January, 2009
------------------------------------------
* removed *.orig and *.rej relicts
Version 1.16_02, released 1 January, 2009
------------------------------------------
* Changing join_2_tables to reduce memory usage when joining a lot of tables
* Fixed Bugs:
* 15688 Columns aliased with double quotes are a fatal error
* 16579 Speed optimizations
* 30590 Bug in SQL::Statement::is_number()
* 41875 Bug in synopsis example
Version 1.16_01, released 1 January, 2009
------------------------------------------
* With this release, I'd like to welcome Jens Rehsack as co-maintainer of
the SQL::Statement and SQL::Parser modules. Jens has added in some
great improvements.
Thanks Jens! -- Jeff
* Adding a lot of join tests (once from Jeff, 48 from
PostgreSQL official handbook) - no error of them will be corrected in the
first run
Thanks to Alexander Breibach <alexander.breibach@gmx.de> -- Jens
to trigger/signal a rebuild for the transition 5.8.8 -> 5.10.0.
The list of packages is computed by finding all packages which end
up having either of PERL5_USE_PACKLIST, BUILDLINK_API_DEPENDS.perl,
or PERL5_PACKLIST defined in their make setup (tested via
"make show-vars VARNAMES=...").
ChangeLog:
Changes log for Perl extension SQL::Statement
Version 1.15, released 2 February, 2006
----------------------------------------
* fixed placeholder bug in SQL::Statement::UPDATE
thanks for bug report Tanktalus
Version 1.14, released 21 April, 2005
----------------------------------------
* fixed circular dependency in tests (one mistakenly required AnyData)
Version 1.13, released 18 April, 2005
----------------------------------------
* pod fixes
Version 1.12, released 18 April, 2005
----------------------------------------
* added support for GROUP BY
(several people sent suggestions for this in the past, please email me
so I can credit you, sorry I lost the names)
* added support for true LIMIT - if a LIMIT clause is specified and
no ORDER BY clause is specified, the SELECT will stop searching
when the limit is reached; with an ORDER BY clause it will still
search the entire table because we can only ORDER a set; using
LIMIT without an ORDER BY will greatly increase speed
* added support for CREATE/DROP keyword|operator|type|function
* optimized process_predicate to only look up scalars once
* completely re-wrote the POD
* fixed bug in primary key search optimization
thanks for bug report and test scripts: Jim Lambert, <jimlambrtATmac.com>
* fixed problem with all_cols slowing inserts
thanks for patch and test Cosimo Streppone <cosimoATcpan.org>
* cleaned up case of temp table column names
thanks for bug report: Dan Wright
* added a META.YML and extra tests
developer is officially maintaining the package.
The rationale for changing this from "tech-pkg" to "pkgsrc-users" is
that it implies that any user can try to maintain the package (by
submitting patches to the mailing list). Since the folks most likely
to care about the package are the folks that want to use it or are
already using it, this would leverage the energy of users who aren't
developers.
* added support for CREATE TABLE AS SELECT ... and CREATE TABLE AS IMPORT()
* added support for in-memory tables and heterogeneous operations
* added many new built-in functions see SQL::Statement::Functions.pm
* added support for user-defined functions, see SQL::Statement::Functions.pm
* added support for column name aliases
* added support for comparison to empty string (e.g. WHERE col1='')
currently returns the same as WHERE col1 IS NULL
* clean up a number of bugs in parentheses parsing and in the
predicates IN and BETWEEN as well as speed things up considerably
* fixed function detection regex in SQL::Parser::ROW_VALUE
to accomodate arbitrary spacing
* support concat operator '||' expressions
* added following functions to SQL::Statement::Functions:
- COALESCE/NVL
- DECODE (same as Oracle DECODE)
- CONCAT
- REPLACE/SUBSTITUTE
* fixed/adapted SQL::Statement::get_row_value(),
SQL::Statement::SELECT(), for join'ed
resultsets
* fixed infinite recursion bug with empty IN() predicate
* fixed case issues with table aliases in joins
* column and table name hashes now default to case sensitive
* where() method now supported as per the docs
module directory has changed (eg. "darwin-2level" vs.
"darwin-thread-multi-2level").
binary packages of perl modules need to be distinguishable between
being built against threaded perl and unthreaded perl, so bump the
PKGREVISION of all perl module packages and introduce
BUILDLINK_RECOMMENDED for perl as perl>=5.8.5nb5 so the correct
dependencies are registered and the binary packages are distinct.
addresses PR pkg/28619 from H. Todd Fujinaka.
* added support for MySQL-like "DROP TABLE IF EXISTS"
* fixed bug in dotted column names e.g. tableA.colB
* fixed bug in MAX and MIN
* fixed bug in ORDER BY (when col names not in SELECT list)
* added support for delimited identifiers (inside double quotes);
these are case sensitive and can contain spaces and other
special chars
* added support for two forms of escaping single quotes inside
quoted values: 'O\'Brien' or 'O''Brien'
* added support for both C-Style and SQL-Style double-hypen
comments, e.g. /* comment */ or -- comment
* added GetInfo.pm for use with $dbh->get_info()
* fixed bug in update that refers to its own columns
(e.g. SET num = num + 2)
* fixed bug in MIN and MAX when used with strings
* identifiers (names of columns, tables, and table name
aliases) are now all case insensitive as required by the SQL
standard. all older versions including the XS versions used
case sensitive column names
* added numerous examples to test.pl
* improved and/or fixed bugs in:
* placeholder support
* ORDER BY clause
* LIKE/CLIKE/RLIKE/IN predicates
* table name aliases in explicit joins
Added backwards compatiblity: both SQL::Statement and SQL::Parser now
work in perl version 5.004 and above.
Changed defaults for DBD::CSV so it now accepts new SQL without adding
extra flags to scripts
Added support for SQL comments
Fixed bugs in qualified column names (e.g. tableA.*), and in joins
using ON or WHERE
Fixed bug in UPDATE that caused the new value to be a hash rather
than a scalar.
New features:
Explicit and implicit joins
Table name aliases
Set functions
String functions
String concatenation
Numeric expressions
IN predicate
BETWEEN predicate
Alphabetic comparison in WHERE clauses
Ordering of text that looks like a number
Verbose error messages for both Parsing and Execution errors
Pure perl, new maintainer.
The automatic truncation in gensolpkg doesn't work for packages which
have the same package name for the first 5-6 chars.
e.g. amanda-server and amanda-client would be named amanda and amanda.
Now, we add a SVR4_PKGNAME and use amacl for amanda-client and amase for
amanda-server.
All svr4 packages also have a vendor tag, so we have to reserve some chars
for this tag, which is normaly 3 or 4 chars. Thats why we can only use 6
or 5 chars for SVR4_PKGNAME. I used 5 for all the packages, to give the
vendor tag enough room.
All p5-* packages and a few other packages have now a SVR4_PKGNAME.
