- remove build/IFiles.pm from the shipped dist, as it is meant to be
auto-generated at install time (was added in 0.26)
- Support native B::OP::parent with 5.21.2 -DPERL_OP_PARENT (Reini Urban)
- fix t/utils/40walk.t with 5.6 (Reini Urban)
- Add $op->_parent method as fallback to our parent method since 5.21.2.
Do it for all packages that
* mention perl, or
* have a directory name starting with p5-*, or
* depend on a package starting with p5-
like last time, for 5.18, where this didn't lead to complaints.
Let me know if you have any this time.
a) refer 'perl' in their Makefile, or
b) have a directory name of p5-*, or
c) have any dependency on any p5-* package
Like last time, where this caused no complaints.
0.15nb1 to 0.17.
Upstream changes:
- Use FUNCLIST to declare exported symbols (Alexandr Ciornii)
0.16 Fri Jul 08 2011
- EU::Depends version. The CPAN built version specified only 0.301
but Windows computers needed 0.302 so their CPAN clients were
broken.
* Try harder to add ExtUtils::Depends to META.yml's configure_requires
Changes 0.14:
* Predeclare carp and croak with proper signature
* Add ExtUtils::Depends to META.yml's configure_requires
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:
0.11 Tue Jan 12 2010
- Repackaged using environment variables COPYFILE_DISABLE=1 and COPYFILE_EXTENDED_ATTRIBUTES_DISABLE=1
- Silenced PROTOTYPES warning on compilation
- Update manifest
0.10 Sun Dec 20 2009
- Fixed B::OP::siblings
- Updated docs
0.09 Sun Dec 06 21:36:00 PDT 2009
- Add ExtUtils::CBuilder to prequisite list to hopefully quench any
bad builds attempted on perl-5.
Upstream changes:
0.07 Tue Nov 11 23:05:.. PDT 2008
- Fixed a refactoring error in walkallops_simple. walkallops* was broken.
- Avoided base.pm because that library is kind of nasty
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=...").
Pkgsrc changes:
o Package now uses C
Upstream changes:
0.05_09 Sat Apr 6 16:09:.. CEST 2008
- Allow for nested disjunctions in opgrep patterns.
0.05_08 Sat Apr 5 22:58:.. CEST 2008
- I-am-a-muppet.
- Removed spurious use of Data::Compare.
0.05_07 Tue Apr 4 18:15:.. CEST 2008
- opgrep() can now capture sub-ops so if you plan to
extract information from or modify a deeply nested
op-tree, you don't have to do the tree-walking
yourself.
0.05_06 Tue Apr 3 18:15:.. CEST 2008
- Add the as_opgrep_pattern() method to B::OP for
dumping op trees as patterns for matching with opgrep.
- Add warnings.
- Documentation fixes.
0.05_05 Tue Apr 1 18:15:.. CEST 2008
- Fix POD error and update README.
0.05_04 Tue Apr 1 18:10:.. CEST 2008
- Fix bug in opgrep() alternations.
- Add the "kids" keyword to the opgrep syntax.
- Add the "dump" keyword to the opgrep syntax for debugging.
- Fixed the Change log (yes, this file) for the last
version since the comment on an opgrep() change was
incorrect.
0.05_03 Sun Mar 30 11:13:.. CEST 2008
- Fix various bugs in opgrep(). Did that work in 0.05_02 at all?
- Introduce op_or() to match one of many possible conditions.
- Add a longer example to opgrep() documentation.
- Fix POD markup.
- Add a clear license statement to the docs.
- Use a string as version -- otherwise it might not be flagged
as a development release!
0.05_02 Thu Dec 1 ..:..:.. ... 2005
- ->parent works for all tests now.
- Added more tests.
0.05_01 Fri Oct 28 23:00:00 CST 2005
- Added tests. That's the focus of my work right now.
- opgrep() now acceps \@conditions
assertions may be code references
- Lots of changes to ->parent. This is in progress. It fails when
the parent of a node does not exist in the execution pathway of
a program. The unused null() ops rooting some conditionals, etc.
This function fails tests.
- Added a function recalc_sub_cache so anon_subs, all_starts, and
all_roots' cache of functions can be cleared.
- Added functions ->siblings, ->ancestors, ->descendants, and
_uniq_ops.
