Revision history for Perl extension List-SomeUtils

0.58     2019-10-26

[BACKWARDS INCOMPATIBILITY]

* Fixed a number of bugs in the handling of empty lists. The any_u(), all_u(),
  none_u(), notall_u(), and one_u() functions returned an empty list in list
  context instead of undef. This is in direct opposition to the documentation
  which has always said that it returned undef. Note that the
  List::SomeUtils::XS code had the _correct_ behavior, so most users probably
  have never experienced this bug.

[IMPROVEMENTS]

* Fixed some doc issues. Closes #7.


0.57     2017-12-12 (TRIAL RELEASE)

[BACKWARDS INCOMPATIBILITY]

* Fixed a number of bugs in the handling of empty lists. The any_u(), all_u(),
  none_u(), notall_u(), and one_u() functions returned an empty list in list
  context instead of undef. This is in direct opposition to the documentation
  which has always said that it returned undef. Note that the
  List::SomeUtils::XS code had the _correct_ behavior, so most users probably
  have never experienced this bug.


0.56     2017-07-22

* Make sure we depend on the latest LSU::XS if the system support XS.


0.55     2017-07-22

* Fixed incorrect comments in doc examples for uniq().

* Added a new function, mode(), requested by Jerrad Pierce. GH #2.


0.54     2017-06-01

* Moved issue tracking to GitHub.


0.53     2016-08-14

* Replaced Exporter::Tiny with Exporter.


0.52     2016-05-06

* Updated to latest ExtUtils::HasCompiler and simplified dynamic
  prereqs. Patch by Karen Etheridge. GitHub #3.


0.51     2016-02-27

* No code changes. The last release was missing a Changes entry.


0.50    2016-02-27

* This is List::MoreUtils under a different name with fewer configure-time
  prereqs.

* The "import as old versions" feature present in LMU does not exist in this
  distribution.


0.413	2015-06-10

- Fix compiling in c++ mode (depreciated, but some people seem to
  require it). Solves RT#104690


0.412	2015-05-19

- release 0.411_001 without further changes

0.411_001	2015-05-11

- move generation of test endpoints to author stage as requested per issue/#9
- add a rough guide for contributors
- fix rt#103251 to avoid removing bundled stuff by accident
- Fix compilation errors under cl (Thanks to jddurand)


0.410	2015-03-30

- release 0.409_003 after no further issues came up


0.409_003	2015-03-27

- update bundled bootstrap modules
  * Data::Tumbler to 0.010
  * Test::WriteVariants to 0.012
  * Config::AutoConf to 0.311
- fix spelling (and add stop-words for names etc. in author tests)


0.409_002	2015-03-23

- fix multiple mg_get can break weird tie's (thanks to leont)
- fix test run using PERL5OPT=d:Confess (thanks kentl & ribasushi)
- use base instead of parent, cause parent isn't bundled before 5.10.1
  (smoke report from SREZIC)
- update bundled modules (for bootstrapping) and ppport.h (from 3.25
  to 3.31)


0.409_001	2015-03-21

- fix RT#102885: uniq bug broke tied array (reported by louying@pwrd.com)
- fix issue/8: Macros introduced in dfd851147f cause problems with MSVC
  (reported by A. Sinan Unur)
- Update ppport.h from 3.25 to 3.31


0.408	2015-03-18

- fix RT#102840: uniq broken for call-by-function-return (reported by
  Jean-Damien Durand), with a new test case thanks to Thomas Sibley
- fix RT#102853: hent_val accesses (reported by Brad Forschinger with
  a reasonable patch)
- fix RT#102833: Compilation error with perl 5.21.7+ (reported by
  Slaven Rezic)
- fix regex for RT#44518 test


0.407	2015-03-17

- Added one(), onlyidx(), onlyval() (RT#73134, MHASCH) and onlyres()
- improve XS maintainability
- document how uniq/distinct deal with undef (RT#49800)
- add bsearchidx to satisfy RT#63470
- add singleton to satisfy RT#94382
- fix RT#82039 - uniq changes the type of its arguments
- fix RT#44518 again


0.406	2015-03-03

- add new functions firstres and lastres in addition to firstidx, lastidx,
  firstval and lastval
- regenerate MANIFEST to bundle README.md


0.405	2015-02-14

- fix RT#78527 - first_val/last_val in documentation
- fix RT#102055 - ExtUtils::MakeMaker required version absurdly high
- update README (deploy it as README.md now)
- fix compiler issue for older/ansi-c89 compilers
- remove local compat workarounds in favour for ppport.h


0.404	2015-01-28

- fix ancient toolchains (PREREQ_PM & Co. set appropriately),
  reported by ilmari
- bump version required of Test::More to 0.96 (#toolchain
  calls it a "sane subset")
- fix some meta-data #toolchain pointed out


0.403	2015-01-27

- remove most recent stable perl recommendation from meta to
  workaround misbehaving CPAN clients blocking update
- update copyright date
- ensure AUTHOR is a string on older toolchains


0.402	2014-12-17

- bump Config::AutoConf and Test::WriteVariants requirement for
  improved 5.6 compatibility (fixes rt#101121)
- use base instead of parent in configure stage (improves building
  on 5.6)
- fix rt#101067 by applying patch from Father Chrysostomos (thanks
  to Lukas Mai (MAUKE) for reporting and explaining)


0.401	2014-12-08

- update bundled Config::AutoConf to 0.307
- release after long testing period


0.400_010	2014-12-08

- bundle configure_requires using inc::latest
- fix RT#96596 by checking types before starting logic ...
  * RT#86260 reported the same issue
- lower minimum perl required to 5.6
- switch to check_produce_loadable_xs_build of Config::AutoConf 0.306


0.400_009	2014-05-05

- improve documentation (David Golden, Jens Rehsack)
- bundle non-core modules (compared to 5.14)


0.400_008	2014-04-24

- fix none for 0.24 and clarify API tag documentation (David Golden)
- refactor import tags for clarity (David Golden)


0.400_007	2014-04-22

- cut out exporter-related cruft; it was only necessary when needing
  to choose between multiple implementations (Toby Inkster)
- Reorganize and clarify documentation (David Golden)
- revise SYNOPSIS and DESCRIPTION for revised export model (David
  Golden)
- introduce ":like_*" import tags (Toby Inkster, Jens Rehsack)
- remove Data::Tumbler and Module::Pluggable from configure
  dependencies, they're coming with Test::WriteVariants


0.400_006	2014-04-01

- fix typos in POD (RT#87490 - thanks to David Steinbrunner)
- refactor LMU as discussed with David Golden, Tim Bunce and
  Toby Inkster


0.400_005	2014-03-24

- rename implementations from alias => relax and tassilo => strict
- remove 'sno' implementation
- add precedence 'default' in addition to 'all' for those who prefer
  strict over relax
- move dependency Module::Runtime from configure to runtime


0.400_004	2014-03-21

- Switch from Sub::Exporter to Exporter::Tiny (Toby Inkster)
- fix issues on older perls back to 5.8.1 (Config::AutoConf
  will not do out of the box, but this can be fixed)
- fix backward compatibility issues (RT#94013 in conjunction
  with RT#93995)
  Details needs to be discussed with Moose community (unless
  they stop caring) to get out of distinguishing hell as soon
  as possible
- fix some spelling issues reported by David Steinbrunner
  (RT#86347)
- clarify depedencies, especially recommended ones
- add some additional tests to prove reported bugs (informed
  reporters when not reproducable)


0.400_003	2014-03-18

- fix compile error on threaded perls (RT#93934 - thanks Andreas
  Koenig for reporting)
- fix exporter configuration (RT#93929 - thanks Andreas Koenig for
  reporting)
- fix RT#40905 by allowing choose an appropriate implementation
- add test for RT#76749 - seems not reproducable (but hopefully
  reporter David J. Oswald can fix the test to help fixing the
  issue beyond)


0.400_002	2014-03-16

- reduce minimum perl version to 5.8.1
- split implementations between existing authors
- switch to DynaLoader and Sub::Exporter
- rely for testing on Test::WriteVariants and Data::Tumbler
  (DBI::Test technology to improve tests)
- fix 64-bit integer precision (RT#93207 reported by Dana Jacobsen)


0.400_001	2013-10-11

- Reformat Changes as per CPAN::Changes::Spec
- taking FIRSTCOME power and move repository to GitHub
- merge Tassilo's 0.25_nn dev releases back
    * mark "any" and "all" as "to be discussed"
	  ==> API changes made by Alias/ADAMK 
- bump version to clarify new age (contributors welcome!)
- bump minimum perl version to 5.8.3


0.33	2011-08-04

- Updated can_xs to fix a bug in it


0.32	2011-05-20

- Production release, no other changes


0.31_02	2011-03-21

- More accurate detection of XS support (ADAMK)


0.31_01	2011-03-21

- Updating copyright year (ADAMK)
- Teak documentation of all() and none() (WYANT)
- Memory leak fixed for apply() and XS version restored (ARC)
- Memory leak fixed for indexes() and XS version restored (ARC)
- Memory leak fixed for part() and XS version restored (ARC)


0.30	2010-12-16

- Change the way we localise PERL_DL_NONLAZY to false to remove
  a warning that some people were seeing. The new approach is taken
  from the way that List::Util does it.


0.29	2010-12-08

- Removed an erroneous Test::NoWarnings dependency


0.28	2010-12-07

- Switching to a production release
- Restored the regression test for RT #38630 from 0.23.
  As apply() was disabled in 0.27_04 this test will only act
  to validate the future XS restoration of apply().
- Adding uniq warning tests, disabled initially


0.27_04	2010-12-06

- The behaviour of any/all/none/notall has changed when
  passed a null list to treat a null list as a legitimate list.
  Instead of returning C<undef> the functions now return the
  following: any {} == false, all {} == true, none {} == true,
  notall {} == false.
  Resolves #40905: Returning undef when none is passed an empty
- Disabled the leaking XS versions of part(), apply() and indexes()


0.27_03	2010-12-06

- General house cleaning


0.27_02	2010-12-01

- Reduced test suite peak memory consumption by 5-10 meg
- Added the 'distinct' alias for the uniq function, for people that
  like their chained map/grep/sort pipelines with a SQL'ish flavour.
- Expanded test suite for the any() group of functions.
- The any() group of functions now strictly always return scalar
  boolean true, false and undef to match the XS version.


0.27_01 2012-12-01

- Refactored the split test scripts into a common test module
  to be shared between both the Perl and XS versions.
- Reapply fix for http://rt.cpan.org/Ticket/Display.html?id=39847
  "minmax error: unpredictable results with lists of 1 element"


0.26	2010-11-23

- No changes
- Some parts of the CPAN cloud were confusing my 0.24
  release with the older deleted 0.24. Bumping version
  past Tassilo's to clarify things.


0.24	2010-11-22

- No changes, switching to a production version


0.23_01	2010-09-25

- First attempt at repackaging the List::SomeUtils code in
  Makefile.PL and release toolchain similar to Params::Util


0.25_02	2009-08-01

- MS VC++ 7 doesn't like inline nor 'long long'
  (patch provided by Taro Nishino (taro DOT nishino AT gmail.com)
- Newx isbn't around in older perls so use New(0,...) instead


0.25_01	2009-07-30

- it seems the only way of handling the stack that works on all 
  flavors of the multicall API is by making a shallow copy of it
  and use that between the PUSH/POP_MULTICALL bracket
- fix awkward ok() override in List-SomeUtils.t so that it reports
  line numbers in test failures properly


0.24	2009-07-19

- List::SomeUtils was not handling the stack properly when the stack was grown
  from inside code-references
- a couple of tests for each_arrayref were calling each_array 


0.23	2009-04-19

- BACKWARDS INCOMPATIBLE CHANGE:
  fixed: Returning undef when none is passed an empty array is counterintuitive
  (http://rt.cpan.org/Ticket/Display.html?id=40905)
- fixed: minmax error: unpredictable results with lists of 1 element 
  (http://rt.cpan.org/Ticket/Display.html?id=39847)
- fixed: bug: uniq doesn't like undef values.
	     uniq warns on undef values
  (http://rt.cpan.org/Ticket/Display.html?id=37533)
  (http://rt.cpan.org/Ticket/Display.html?id=43214)
- fixed: bug in pairwise when $a and $b are lexically defined using my
  (http://rt.cpan.org/Ticket/Display.html?id=44518)
- fixed: Big memory leak with XS part()
  (http://rt.cpan.org/Ticket/Display.html?id=41097)
- fixed: memory leak in indexes() [XS]
  (http://rt.cpan.org/Public/Bug/Display.html?id=41494)
- reduced memory-requirements for the part() tests as that was responsible
  for a lot of unnecessary test-failures
- new function bsearch() which performs a binary search


0.22	2006-07-02

- SvPV_nolen doesn't exist on pre 5.6 perls


0.21	2006-06-18

- propagate dies from inside the code-reference of pairwise to 
  caller


0.20	2006-04-25

- part() would destroy the list elements when changing
  an array in place (@list = part { ... } @list)


0.19	2006-03-13

- working down myself the queue of suggestions:
  part() added
  (Ricardo SIGNES <rjbs AT cpan DOT org>)


0.18	2006-02-25

- each_arrayref (XS) couldn't deal with refs to list literals
  (brought up by David Filmer <usenet AT DavidFilmer DOT com>
   in comp.lang.perl.misc)


0.17	2005-12-07

- each_arrayref had no XS implementation and wasn't
  mentioned in the PODs
  (patch by Anno Siegel <siegel AT zrz DOT tu-berlin DOT de>)


0.16	2005-11-14

- a dangling semicolon in some macros prevented
  the XS portion to compile on some compilers
  (Lars Thegler <lars AT thegler DOT dk>)


0.15	2005-11-11

- 0.13 and 0.14 broke the module on 5.6.x
  (spotted by Thomas A. Lowery <tlowery AT cc3 DOT com>)
- internals changed to make use of the new MULTICALL API 
  which had to be backported to 5.005_x


0.14	2005-11-10

- 0.13 fixed the leaks but rendered the XS part uncompilable
  for perls < 5.6.0: Fixed
  (spotted by Lars Thegler <lars AT thegler DOT dk>)


0.13	2005-11-09

- nearly all functions receiving a CODE-block as first
  argument had a hefty memory-leak: Fixed
  (spotted by Thomas A. Lowery <tlowery AT cc3 DOT com>)


0.12	2005-09-28

- first_index and each_arrayref weren't exportable
  (spotted by Darren Duncan)


0.11	2005-09-27

- make sure that Test::Pod and Test::Pod::Coverage are
  installed in the required minimum versions
  (thanks to Ricardo Signes <rjbs AT cpan DOT org>)


0.10	2005-04-01

- new function minmax() with comparisons in O(3n/2 - 2)
- some POD corrections (Adam Kennedy)
- POD- and POD-coverage tests


0.09	2004-12-04

- 0.08 only fixed uniq() for scalar context


0.08	2004-12-03

- uniq() was not mentioned in the perldocs and only had
  the XS implementation
- uniq() also produced wrong results on 5.8.0
  (thanks to Slaven Rezic for spotting it and suggesting
  a workaround)
- the test-suite triggered a bug in 5.6.x perls
- the test-suite now tests both the XS- and Perl-implementation
- a wrong example in the perldocs fixed (Ron Savage)


0.07	2004-12-01

- new functions:
  after, after_incl, before, before_incl, indexes
  lastval, firstval, pairwise, each_array, natatime,
  mesh (all from Eric J. Roodes' List::MoreUtil).


0.06	2004-11-14

- new function 'apply' on behalf of Brian McCauley
  (<nobull AT mail DOT com>)


0.05	2004-09-18

- merged in insert_after() and insert_after_string() from
  List::Utils which is now obsolete
  (thanks to James Keenan <jkeen AT verizon DOT net> and
  Terrence Brannon <terry AT hcoop DOT net>)


0.04	2004-07-10

- renamed to List::SomeUtils on suggestion by
  Steve Purkis <spurkis AT quiup DOT com>


0.03	2004-07-09

- some compilers don't like the stale goto labels
  without any statement following. Fixed.
  (Robert Rothenberg <wlkngowl AT i-2000 DOT com>)


0.02	2004-07-08

- added Perl implementations of all functions 
  as a fallback 
  (Adam Kennedy <adam AT phase-n DOT com>)


0.01	2004-07-05

- original version; created by h2xs 1.23 with options
  -b 5.5.3 -A -n List::Any