New in 2.7:
* Support for platforms that follow POSIX:2008, not POSIX:2001.
* Support for MirBSD 10.
* Support for IRIX 5.3. Contributed by Eric Blake.
* On Linux platforms, libsigsegv now prefers the POSIX way of defining the
signal handler over than the traditional one, when both are supported.
As a consequence, on Linux/i386 and other Linux platforms, the type
'stackoverflow_context_t' is now typedefed to 'ucontext_t *' rather than
'struct sigcontext *'.
XXX: Needs m4>=1.4.6, but we use USE_TOOLS+=gm4:run, and I don't see
a way to specify a version there. Please someone fix that :)
On the other hand, 1.4.6 was added to pkgsrc in 09/2006, so I hope
everyone has it by now.
* Major changes in Autoconf 2.64 (2009-07-26) [stable]
Released by Eric Blake, based on git versions 2.63b.*.
** Autoconf now requires GNU M4 1.4.6 or later. Earlier versions of M4
have a bug in regular expression handling that interferes with some
of the speedups provided since Autoconf 2.63. GNU M4 1.4.13 or
later is recommended.
** AS_IF and AS_CASE have been taught to avoid syntax errors even when
given arguments that expand to just whitespace.
** The following documented autoconf macros are new:
AC_ERLANG_SUBST_ERTS_VER
** The autoheader tool now understands m4 macro arguments passed to
AC_DEFINE and AC_DEFINE_UNQUOTED.
** Ensure AT_CHECK can support commands that include a # given with
proper m4 quoting. For shell comments, this is a new feature; for
non-shell comments, this fixes a regression introduced in 2.63b.
Additionally, AT_CHECK correctly supplies shell escapes for
metacharacters occurring in m4 macro expansions within the expected
stdout and stderr parameters.
** The macro AT_CHECK now understands the concept of hard failure. If
a test exits with an unexpected status 99, cleanup actions for the
test are inhibited and the test is treated as a failure regardless
of AT_XFAIL_IF. It also understands the new directives
ignore-nolog, stdout-nolog, and stderr-nolog.
** The following documented autotest macros are new:
AT_CHECK_UNQUOTED AT_FAIL_IF AT_SKIP_IF
** The following documented m4sugar macros are new:
m4_argn m4_copy_force m4_default_nblank m4_default_nblank_quoted
m4_ifblank m4_ifnblank m4_rename_force
** The autoconf testsuite now exercises all Erlang macros.
* Major changes in Autoconf 2.63b (2009-03-31) [beta]
Released by Eric Blake, based on git versions 2.63.*.
** The manual is now shipped under the terms of the GNU FDL 1.3.
** AC_REQUIRE now detects the case of an outer macro which first expands
then later indirectly requires the same inner macro. Previously,
this case led to silent out-of-order expansion (bug present since
2.50); it now issues a syntax warning, and duplicates the expansion
of the inner macro to guarantee dependencies have been met. See
the manual for advice on how to refactor macros in order to avoid
the bug in earlier autoconf versions and avoid increased script
size in the current version.
** AC_DEFUN_ONCE has improved semantics. Previously, a macro declared
with AC_DEFUN_ONCE warned on a second invocation; and out-of-order
expansion was still possible. Now, dependencies are guaranteed,
and subsequent invocations are a silent no-op. This makes
AC_DEFUN_ONCE an ideal macro for silencing AC_REQUIRE warnings.
** The following macros are now defined with AC_DEFUN_ONCE. This means
a subtle change in semantics; previously, an AC_DEFUN macro could
expand one of these macros multiple times or surround the macro
inside shell conditional text to bypass the effects of these
macros, but now the macro will expand exactly once, and prior to
the start of any enclosing AC_DEFUN macro:
AC_CANONICAL_BUILD AC_CANONICAL_HOST AC_CANONICAL_TARGET
AC_HEADER_ASSERT AC_PROG_INSTALL AC_PROG_MKDIR_P
AC_USE_SYSTEM_EXTENSIONS
** AC_LANG_ERLANG works once again (regression introduced in 2.61a).
** AC_HEADER_ASSERT is fixed so that './configure --enable-assert' no
longer mistakenly disables assertions.
** AC_INIT now takes an optional fifth parameter that can be used to
set AC_PACKAGE_URL, a URL for the package's home page; the URL is
used in `configure --help' and is also available via AC_DEFINE.
** Autotest testsuites accept an option --jobs[=N] for parallel testing.
This feature is still in testing, and may not work on every
platform, help in improving it would be appreciated.
** Autotest testsuites do not attempt to write startup error messages
to the log file before that is opened (regression introduced in 2.63).
** Configure scripts now use shell functions. This feature leads to
smaller configure files and faster execution.
** Present But Cannot Be Compiled: Autoconf will now proceed with
the compiler's result if a header is present but cannot be compiled.
The warning is still printed, and you should really fix it by
providing a fourth parameter to AC_CHECK_HEADER/AC_CHECK_HEADERS.
** Autoreconf added aclocal to the set of programs affected by the
`autoreconf -I dir' option.
** The following documented m4sugar macros are new:
m4_chomp m4_chomp_all m4_cleardivert m4_curry m4_default_quoted
m4_esyscmd_s m4_map_args m4_map_args_pair m4_map_args_sep
m4_map_args_w m4_set_map m4_set_map_sep m4_stack_foreach
m4_stack_foreach_lifo m4_stack_foreach_sep
m4_stack_foreach_sep_lifo
** The following m4sugar macros are documented now, but in some cases
with slightly different semantics than what the previous
undocumented version had:
m4_copy m4_dumpdefs m4_rename m4_version_prereq
** The m4sugar macro m4_expand has been taught to handle unterminated
comments and shell case statements. As a result, it is used
internally in more places, such as AC_DEFINE and AT_CHECK. Most
uses of AC_DEFINE and AT_CHECK should not behave any differently;
however, it may be necessary to add double-quoting around
unbalanced `(' where single-quoting used to be sufficient.
** The following documented m4sh macros are new:
AS_INIT_GENERATED AS_LINENO_PREPARE AS_ME_PREPARE AS_SET_STATUS
AS_VAR_APPEND AS_VAR_ARITH AS_VAR_COPY
** The following m4sh macros are documented now, but in some cases
with slightly different semantics than what the previous
undocumented version had:
AS_ECHO AS_ECHO_N AS_ESCAPE AS_EXIT AS_LITERAL_IF AS_UNSET
AS_VAR_IF AS_VAR_POPDEF AS_VAR_PUSHDEF AS_VAR_SET AS_VAR_SET_IF
AS_VAR_TEST_SET AS_VERSION_COMPARE
** The m4sh macros AS_IF and AS_CASE can now be used in shell lists.
The responsibility for supplying a trailing newline now belongs to
the call site, but since most users did not add dnl, this generally
results in fewer empty lines in configure.
pkgsrc changes:
- Adding license information
- Remove workarounds for the crap of the last version
Update changes:
from 1.25: Localize special variables so that the exit status
from waitpid doesn't leak out, causing exit status
to be incorrect (RT33440, fixed by Brad Cavanagh).
pkgsrc changes:
- Adjusting dependency information according to META.yml
Upstream changes:
[1.103] Released on 2009-08-03
Fix configure_require prerequisite on Module::Build 0.34_02.
[1.102] Released on 2009-08-03
Bug fixes:
* Works with PPI 1.205. Yay for 5.10 support!
* Variables::RequireLexicalLoopIterators didn't work correctly on foreach
loops with labels.
[1.101_003] Released on 2009-07-22
[1.101_002] Released on 2009-07-21
[1.101_001] Released on 2009-07-21
Changes summarized into 1.102 above. For exact details, see Changes on
BackPAN.
pkgsrc changes:
- Adding license information
- Adjusting dependencies according to comments in Changes and required
modules with version checking in PAR/dist.pm
Upstream changes:
By: smueller on 2009/07/31
* Do not use Archive::Zip if its version is 1.28.
* This is 0.46.
to 0.994
pkgsrc changes:
- Adding license information
- Adjusting dependencies according to META.yml
Upstream changce since 0.992:
[Changes for 0.994 - Jul 23, 2009]
- Fix for the PAR::Heavy fix to the INC priority handling.
[Changes for 0.993 - Jul 19, 2009]
- The priority (fallback=>0) repositories should look at @PAR_INC
for the loaded PARs instead of @PAR_INC_LAST.
- Don't reload from a downloaded .par file after installing it
via "upgrade".
- Band-aid fix for the loading priority of shared librares from
PAR files: Try PAR's first, the local stuff, then fallback-PARs.
- Initial support for running external perl scripts from a packaged
interpreter.
pkgsrc changes:
- Adjust license according to META.yml
- Adjust dependencies according to META.yml
Upstream changes:
3.56 Wed Jul 29 19:11:02 2009
- Changed label that was a keyword
pkgsrc changes:
- Adjusting license
- Adding dependency to MRO::Compat
Upstream changes:
0.15 Sun, Jul 26 18:02:05 2009 +0100
* Fix test which was failing in some cases and additional test cases.
* No other changes on the dev release.
0.14_01 Thu, Jul 16 20:04:04 2009 +0100
* Add TODO tests for role combination with method attributes, proving that this doesn't work.
* Add nasty hack to allow the application of roles with method attributes to anon classes to
work if the user does a special handwave. This is less than optimum, see big block comment
in the code. :/
pkgsrc changes:
- Use Module::Install as module type
Upstream changes:
[Changes for 0.93 - 2009-07-19]
* Implement caching of dependencies (Christoph Lamprecht)
[Changes for 0.92 - 2009-07-19]
* Fix bug with {type} being set to unexpected values in some cases (Christoph Lamprecht)
* Add tests for scan_chunk (Alexandr Ciornii)
* Add special case for parent.pm (Alexandr Ciornii)
* Fix for "use parent::something" (Alexandr Ciornii)
* Add special case for Catalyst.pm (Alexandr Ciornii)
pkgsrc changes:
- Adjusting license according to module's POD
Upstream changes:
1.24 (2009/07/08)
* (ms) Fixed bug with Log::Log4perl::Util::tmpfile_name which
surfaced on VMS, reported by Ben Humphreys.
* (ms) Fixed system-wide threshold to no longer lower appender
thresholds. Bug reported by Jean-Denis Muys.
* (ms) Added benchmark to determine impact of eval-free handlers
* (ms) Merged with eval_free branch. Now there are no more
eval("") statements left in the code, making it much easier
to debug. Performance on init() is about the same, performance
on init_and_watch() (noops and logged statements alike) is
25% slower but still in the range of 400,000/sec on my
1.80Ghz CPU.
pkgsrc changes:
- Adding license (perl5 license)
Upstream changes:
=================================
2009-07-27T16:08:44.238440Z v1_01
=================================
2009-07-27 16:08:19 (r22) by rcaputo; lib/Lexical/Persistence.pm M
Reorganize documentation links.
2009-07-27 04:48:07 (r21) by rcaputo
lib/Lexical/Persistence.pm M; Makefile.PL M
Added a machine-readable repository directory to the distribution.
Documented the bug tracker, repository, and other resource URLs.
2008-12-15 16:19:56 (r20) by rcaputo; Makefile.PL M
Keep UTF8 out of the README, because that messes up the SYNOPSIS.
pkgsrc changes:
- Adding license (unrolled perl license?) according to README file
- Updated dependecies according to META.yml
Upstream changes:
0.84 Mon 13 Jul 2009
- Add darwin to the list of known-bad platforms
0.83 Fri 10 Jul 2009
- Switch from command.com to cmd.exe so we work on 2008 or newer.
pkgsrc changes:
- Adding license (perl5 license)
Upstream changes:
1.06 Tue Aug 4 15:10:34 CDT 2009
[FIXES]
Fixed closing =cut in POD.
There are no functionality changes, but the lack of a closing
=cut meant that ack would not build properly.
1.04 Fri Jul 31 16:24:36 CDT 2009
[ENHANCEMENTS]
It's never been correct to call File::Next::files() as a method,
as File::Next->files(). Now, if you do, files() will die with
an error. This is also the case with dirs() and everything().
Thanks to Eric Lyons for reporting.
Tiny directory reading speedups.
[DOCUMENTATION]
Updated URLs for support sites.
Added a little note about the follow_symlinks=>0 being a speed
hit.
pkgsrc changes:
- Adjusting dependencies according to META.yml
Upstream changes:
1.09 Thu 30 Jul 2009
- 1.08 broke Windows. This unbreaks it again (ADAMK)
pkgsrc changes:
- Adding perl5 license
Upstream changes:
1.54
This is a "no-change" version bump because I pushed the v1.53 change
and then realized that MakeMaker.t was a bad name for a file that would
end up in core where the EUMM tests and the EUI tests are in the same
directory. This renames it to InstallWithMM.t.
1.53
Final stage of the divorce from EUMM. Now the EUMM related tests are no
longer shared. Build.pl and Build.t go, and there shall be peace on earth.
At least until somebody patches EUMM/t/basic.t for something EUI related...
Thanks to M. Schwern for helping me work this one out. Cheers man.
1.52_03
Missed the t/Installed.t test from core. Bumped version number to allow
a new distro to be released.
1.52_02
Make _chmod verbose message use octal modes, thanks to BDFOY
Further changes from core, including lastest test file infrastructure
from EUMM.
Fixed a number of problems in ExtUtils::Installed, for various reasons
this includes a version bump to 1.999_001, which will eventually become
version 2.0. These problems related to finding modules that were installed
with either INSTALL_BASE or PREFIX. Hopefully this resolves these issues.
1.52_01 (core only release)
Changes from Core:
commit 3d55b451d9544fbd4c27c33287b76bee30328830
Author: John Malmberg
Date: Sun Feb 15 09:25:10 2009 -0600
ExtUtils::Install VMS extended character set support
Preview from https://rt.cpan.org/Ticket/Display.html?id=42149
pkgsrc changes:
- Adding license (perl5 license)
Upstream changes:
1.04 Sun 12 Jul 2009
- Upgrading to Module::Install::DSL 0.91
- Sometimes there is no Config_heavy.pl
- Updating dependencies to rid myself of memory leaks
pkgsrc changes:
- Setting license to perl5 license
Upstream changes:
2009-07-28 John Peacock <jpeacock@cpan.org>
Little jog to make sure META.yml does't contain UNIVERSAL
2009-07-28 John Peacock <jpeacock@cpan.org>
Only replace use_ok() if running with Test::More < 0.48
2009-07-28 John Peacock <jpeacock@cpan.org>
Provide replacement use_ok to make the 02derived.t tests pass.
2009-07-28 John Peacock <jpeacock@cpan.org>
Fix for https://rt.cpan.org/Ticket/Display.html?id=48268
2009-07-26 John Peacock <jpeacock@cpan.org>
Script to run through all of the Perl releases in one go
2009-07-26 John Peacock <jpeacock@cpan.org>
Release 0.77 to CPAN without the warning change, for release with 5.10.1
2009-07-24 John Peacock <jpeacock@cpan.org>
Finally complete the POD rewrite. Change the behavior to throw a
warning if you try and use a v-string without a leading 'v' in a
version object declaration (this may get pulled) and rewrite the POD
to follow the New World Order.
2009-07-24 John Peacock <jpeacock@cpan.org>
Resolves https://rt.cpan.org/Public/Bug/Display.html?id=48135
2009-07-22 John Peacock <jpeacock@cpan.org>
Forgot to bump this. Always run tests before committing!
2009-07-22 John Peacock <jpeacock@cpan.org>
Revised version::Internals POD; bump $VERSION for potential last CPAN
release.
2009-07-22 John Peacock <jpeacock@cpan.org>
WIP for version::Internals
2009-07-21 John Peacock <jpeacock@cpan.org>
Don't need to load the class in order to check whether it contains package or assignments.
2009-07-18 John Peacock <jpeacock@cpan.org>
Rename Extended to Dotted-Decimal
2009-07-18 John Peacock <jpeacock@cpan.org>
Rename "Numeric" to "Decimal"
2009-07-18 John Peacock <jpeacock@cpan.org>
Fix for RT#47980. Don't check $@ if you haven't actually done the eval().
2009-07-16 John Peacock <jpeacock@cpan.org>
Apply David Golden's suggested changes with some minor massaging.
2009-07-15 John Peacock <jpeacock@cpan.org>
Another CPAN alpha release.
2009-07-15 John Peacock <jpeacock@cpan.org>
Start reworking the Internals documentation into something useful.
2009-07-14 John Peacock <jpeacock@cpan.org>
Tweakage of POD
2009-07-10 John Peacock <jpeacock@cpan.org>
Neglected to delete this directory from the repo
2009-06-29 John Peacock <jpeacock@cpan.org>
Allow the pure Perl version module to be installed on Perl 5.10.0,
and trump the core code.
2009-06-28 John Peacock <jpeacock@cpan.org>
Better heuristic for deciding when to rebless
2009-06-28 John Peacock <jpeacock@cpan.org>
Remove old file from MANIFEST
2009-06-28 John Peacock <jpeacock@cpan.org>
Convert this test to make its own Empty class
2009-06-28 John Peacock <jpeacock@cpan.org>
Missed a couple more MAGIC NUMBERS in the tests.
2009-06-28 John Peacock <jpeacock@cpan.org>
Tests all pass now in 5.005_04 in XS mode too!
Need to adapt the pure Perl release to work with 5.10.0 as well.
2009-06-27 John Peacock <jpeacock@cpan.org>
Whitespace differences from blead
2009-06-27 John Peacock <jpeacock@cpan.org>
Sync changes from bleadperl to vutil.c
2009-06-26 John Peacock <jpeacock@cpan.org>
Finally have all tests passing in 5.10.0!
2009-06-23 John Peacock <jpeacock@cpan.org>
Disable all tests for 5.10.0 for the moment, so we can release
as an alpha and not falsely claim success or failure.
2009-06-13 John Peacock <jpeacock@cpan.org>
Fix RT#46921 - locale and eval action at a distance. All tests pass
except on 5.10.0 (because the core code is broken).
2009-06-13 John Peacock <jpeacock@cpan.org>
Fix mistaken regex to convert large exponential numbers to non-exponential
form before scanning. Resolves:
https://rt.cpan.org/Ticket/Display.html?id=45241
2009-05-27 John Peacock <jpeacock@cpan.org>
More fiddling with the revised POD.
2009-05-20 John Peacock <jpeacock@cpan.org>
Resolve some more misleading warnings from vpp.pm in 5.005 and 5.6
2009-05-20 John Peacock <jpeacock@cpan.org>
Misplaced #endif caused 5.005_04 and 5.6.x to fail tests. Add
documentation for is_qv.
2009-05-18 John Peacock <jpeacock@cpan.org>
Make all test pm files use File::Temp. Change qv() to be both method and
function. All tests pass using Build.PL from perl 5.6.x forward, but
a couple of test failures using Makefile.PL in 5.6.x and 5.005_04.
2009-05-16 John Peacock <jpeacock@cpan.org>
Tests pass but POD is incomplete
2009-05-10 John Peacock <jpeacock@cpan.org>
All tests pass in all Perl's and in XS and pure Perl. Perl 5.005_04 still
throws lots of stupid warnings in pure Perl; can't help it apparently.
2009-05-10 John Peacock <jpeacock@cpan.org>
Now tests all pass on 5.005 as well (two warnings I can't prevent)
2009-05-10 John Peacock <jpeacock@cpan.org>
Now all three test files are 100% for 5.8.x and 5.10.x, but 01 and 02 fail
with 5.6.x and 5.005
2009-05-09 John Peacock <jpeacock@cpan.org>
Begin massive reorg/redesign. Tests 01 and 03 are 100%; 02 needs work.
pkgsrc changes:
- Setting license to perl5 license
- Setting module type to Module::Build
Upstream changes:
0.30 5 August, 2009
Applied changes from Florian Ragwitz to ensure that we preserve the
name of the package the alias is defined in.
pkgsrc changes:
- Adding license (perl5 license, according to module's doc)
- Adjust dependencies according to META.yml
- Removing obsoleted deletion of formerly included Carp::Clan
Upstream changes:
Version 5.6 28.07.2009
+ Made the module MacOS X compatible
+ Made some tiny changes to the documentation
Version 5.5 was skipped due to an unauthorized upload by someone else
pkgsrc changes:
- Re-add dependency to devel/p5-Storable - desired version will be in
Perl CORE not before 5.10.1
Upstream changes:
Version 6.7 08.08.2009
+ Replaced STORABLE_thaw by STORABLE_attach
pkgsrc changes:
- Adjust dependencies according to META.yml
- Add license (perl5 license)
Upstream changes:
1.206 Sun 9 Aug 2009
Summary:
- This is an optimisation release (1-2% speed up)
(Using information uncovered by a Devel::NYTProf 3 alpha)
Details:
- Removing som superfluous 1; returns
- Using defined and ref to avoid highly excessive calls
to PPI::Util::TRUE
1.205 Mon 3 Aug 2009
Summary:
- This is a production release
Details:
- No changes from 1.204_07
1.204_07 Fri 31 Jul 2009
Summary:
- Minor tweaks
Details:
- Allow ::For and ::List to return true to ->isa(::ForLoop)
and do a once-per-process warning when we do.
- Fixed a bug in Class::XSAccessor prototype.
1.204_06 Wed 22 Jul 2009
Summary:
- API Change
Details:
- Changing PPI::Structure::ForLoop to PPI::Structure::For
1.204_05 Tue 21 Jul 2009
Summary:
- Bug fixes in preparation for production release
Details:
- There is no longer any real reason to bundle the testing modules
except as a potential source of more bugs.
- Removed quantifier ? on zero-length ^ in /^?for(?:each)?\z/
- Run-time load PPI::Document instal of compile-time loading it
- Tweak a few load orders to get PPI::Util loaded earlier.
- Fixed location access methods on PPI::Element
- New PPI::Statement::Include::version_literal() method.
1.204_04 Thu 16 Jul 2009
Summary:
- Dependency tweaks
Details:
- Because we bundle Test::ClassAPI, we need to explicitly match its
dependencies. Bumped Params::Util to 1.00.
- Bumped a couple of deps a couple of revisions to get better XS.
1.204_03 Tue 14 Jul 2009
Summary:
- More bug fixing, clean up, and optimisation
- Cleaning up contributed APIs
- Adding some demonstration classes
Details:
- Implemented PPI::Transform::UpdateCopyright
- Removed the use of 'use base'
- Various minor simplifications
- Renamed PPI::Statement::Switch to ::Given
- Renamed PPI::Structure::WhenMatch to ::When
- Converted the Lexer internals to use exception-based error
handling.
- Take advantage of the removal of all those "or return undef"
to simplify the Lexer code, remove variable declarations, and
inline calls to several hot-code-path functions. The Lexer
should be significantly faster (FSDO "significant").
- The v6 key on Tokenizer broke support for Perl 5.6
(perl thought it was a numeric v-string)
1.204_02 Sun 10 May 2009
Summary:
- Various bug fixing and stabilisation work
- It's a perl 5.10 extravaganza!
Details:
- Updated Module::Install to 0.87
- Added Test::NoWarnings to the test suite
- Added support for qw{foo} in addition to for ('foo')
- Added support for vstrings again
- Now supports the 5.10 "state" keyword.
(As far as PPI is concerned it's a synonym for "my")
- Now supports switch statements.
- Now supports the smart match operator (~~).
- Now supports keeping track of line numbers and file names as
affected by the #line directive.
- Now supports UNITCHECK blocks.
- Statement::Include::module_version() implemented.
- Statement::Include::arguments() implemented.
- Statement::Variable::symbols() implemented.
- Token::QuoteLike::Words::literal() implemented.
- Token::Quote::Double::simplify() fixed.
- Element line_number(), column_number(), visual_column_number(),
logical_line_number(), and logical_filename() implemented.
- Support for Unicode byte order marks (PPI::Token::BOM) added.
- Token::Word::method_call() implemented.
- Element::descendant_of() and Element::ancestor_of() implemented.
- Statement::specialized() implemented.
- Now can handle files named "0".
(Perl::Critic got a complaint about this)
- foreach loop variables can be declared using "our".
- Much more comprehensive testing of compound statement detection.
1.204_01 Sun 18 May 2008
Summary:
- Unicode cleanup and bug fixing
- Taking the opportunity to do some house cleaning while the
code base is relatively stable, before things get crazy again.
Details:
- For completeness sake, add support for empty documents
- Moved capability detection into PPI::Util
- POD test script now skips on install properly
- Removed 200 lines of old dead "rawinput" code from PPI::Tokenizer
- 100% of PPI::Tokenizer is now exception-driven
- Workaround for "RT#35917 - charsets.t eats all available VM"
(unicode bug in 5.8.6, works in 5.8.8)
- Temporarily disable round-trip testing of 14_charset.t
The API of the fcall struct was changed for ANSI/C99 compliance.
A bug affecting the offset of consecutive writes was fixed. The
server's maximum message size is honored. ixpc's command line
parsing was fixed. Build problems on Solaris were fixed. The append
command was added to ixpc. wstat server support was added.
do not treat a failed rcscheckout() as system failure
A failed checkout is not necessarily caused by a solid system error, but
can be also caused by a faulty RCS file. Do not use rb_sys_fail(), since
it will abort() if errno == 0. Instead, simply raise a RuntimeError.
Reported-by: Brad King
check deltas for correctness before applying
We can not blindly trust the RCS file to contain correct deltas. We have
to verify that the chunks (especially deletes) make sense and operate on
existing lines.
Reported-by: Brad King
pkgsrc changes:
- Adding recommended dependencies to pkgsrc and this package
Upstream changes:
4.91 Thu Aug 6 15:42:45 CEST 2009
- AE::Handle::starttls could get out of sync when the read buffer
already contains some TLS handshake.
- AE::Handle did not properly free the TLS session, causing bigger
memory leaks in Net::SSLeay than expected (reported by toaster).
- AE::Socket::tcp_connect will now artificially delay invoking
the callback to avoid returning after invoking the callback.
- convert many internal watcher uses to AE API.
4.9 Sat Aug 1 11:07:01 CEST 2009
- Glib silently fails when registering a timeout with
a negative value, so avoid that.
- call condvar callback immediately when it is set
after the condition is already signalled.
- check rbuf_max condition only after trying to consume
data.
- normalise signal numbers to names when using the ae's
signal handling, but do not document this (yet).
- pure perl signal emulation did not properly set nonblocking
mode on the signal pipe on win32 (this is unlikely to have caused
any issues).
- new module: AnyEvent::Debug.
- AnyEvent::Strict now checks that a signal was specified
by name (not name or number).
- reduce memory footprint in typical cases by ~50kb by
only compiling signal code when necessary.
- add AnyEvent::Handle->rbuf_max.
- grab the AE namespace for future new API, implement stubs for
said future/alternative API.
- new function: AnyEvent::Socket::format_hostport.
4.881 Tue Jul 28 12:51:53 CEST 2009
- work around a bug in local in pre-5.10 perls, causing
AnyEvent::Handle to recurse when it should not
(analyzed by elmex).
4.88 Tue Jul 28 04:04:37 CEST 2009
- re-bless the handle into a dummy package after calling
AnyEvent::Handle::destroy, so the user does not need to check
for errors after every push_write etc.
- do not attempt to run t/02_signals.t on obviously broken
platforms.
4.87 Sun Jul 26 02:06:16 CEST 2009
- do not attempt to linger when there is no longer a valid fh.
- enforce tls mode to be either accept or connect, do not simply
segfault in Net::SSLeay.
- AnyEvent::Handle can now call tcp_connect itself (new parameters
connect, on_prepare, on_connect and on_connect_error). Updated
tutorial accordingly.
- add AnyEvent::Impl::Irssi backend.
4.86 Mon Jul 20 23:52:29 CEST 2009
- since the verbose warning is not enough, explicitly document
that versions before 1.33 of Net::SSLeay are not secure.
- work around signal handling races in Event and (...) Event::Lib.
- try to align signal-race timer to full-second boundaries.
- work around Tk not liking negative timeouts.
- don't complain of different grades of environmental unfriendlyness
in IO::Async.
4.85 Sat Jul 18 06:16:14 CEST 2009
- nail the signal race problem in perl once and for all
(see $AnyEvent::MAX_SIGNAL_LATENCY).
- take advantage of Async::Interrupt if it is available.
- load Time::HiRes and Guard modules on demand only.
- add optional/recommended modules section to AnyEvent
documentation.
- reduce memory usage considerably (and reduce startup penalty)
by not using "strict", "warnings" and "overload" modules.
- work around buggy windows/openbsd perls and provide EBADMSG
and EPROTO ourselves when missing.
- improve perl 5.6 compatibility of the core event loop.
- made Net::SSLeay version 1.33 a soft requirement.
4.83 Fri Jul 17 16:56:26 CEST 2009
- implement AnyEvent::Socket::getprotobyname.
- AnyEvent::CondVar's will now detect recursive blocking
waits and will croak, as too many people fall into
this trap.
- AnyEvent::Handle will now call ->destroy on itself after
executing the on_error callback, instead of doing some
half-baked internal shutdown, for fatal errors.
- clarify on_eof behaviour w.r.t. the read queue and
on_read callbacks.
- ignore some possible spurious wake-ups in tcp_connect.
4.82 Sat Jul 11 00:34:55 CEST 2009
- POE and Event backends didn't accept some callable objects as
callbacks.
- use Config module instead of POSIX module to detect signal names
in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
module doesn't even have all POSIX signals :/.
- use more workarounds around the many refcnt/corruption bugs in
Event::Lib.
- work around a race condition in perl's select, causing t/03_child.t
to rarely fail.
4.81 Thu Jul 9 10:30:30 CEST 2009
- AnyEvent::Handle didn't properly diagnose write errors
(it expected -1 from syswrite, how lame... :).
- support file descriptors in addition to file handles
in AnyEvent->io.
- new env variables: PERL_ANYEVENT_RESOLV_CONF,
PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
and PERL_ANYEVENT_CA_PATH.
- provide a sensible synopsis section for AnyEvent::TLS.
- add a "supported backends" section to the manpage.
- added simple io watcher test to testsuite, using a
portable_socketpair.
- tried to improve the stability of the Event::Lib backend,
YMMV.
This module implements so-called "guards". A guard is something (usually an
object) that "guards" a resource, ensuring that it is cleaned up when
expected.
Specifically, this module supports two different types of guards: guard
objects, which execute a given code block when destroyed, and scoped guards,
which are tied to the scope exit.
for devel/p5-AnyEvent.
This module implements a single feature only of interest to advanced perl
modules, namely asynchronous interruptions (think "UNIX signals", which are
very similar).
Sometimes, modules wish to run code asynchronously (in another thread, or
from a signal handler), and then signal the perl interpreter on certain
events. One common way is to write some data to a pipe and use an event
handling toolkit to watch for I/O events. Another way is to send a signal.
Those methods are slow, and in the case of a pipe, also not asynchronous -
it won't interrupt a running perl interpreter.
This module implements asynchronous notifications that enable you to signal
running perl code from another thread, asynchronously, and sometimes even
without using a single syscall.
This module provides an interface to libev
(http://software.schmorp.de/pkg/libev.html). While the documentation is
comprehensive, one might also consult the documentation of libev itself
(http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod or perldoc EV::libev)
for more subtle details on watcher semantics or some discussion on the
available backends, or how to force a specific backend with LIBEV_FLAGS,
or just about in any case because it has much more detailed information.
This module is very fast and scalable. It is actually so fast that you can
use it through the AnyEvent module, stay portable to other event loops (if
you don't rely on any watcher types not available through it) and still be
faster than with any other event loop currently supported in Perl.
scheduled import of p5-EV-3.7
This module implements some sane defaults for Perl programs, as defined
by two typical (or not so typical - use your common sense) specimens of
Perl coders.
module AnyEvent).
Libev is modelled (very losely) after libevent and the Event perl
module, but is faster, scales better and is more correct, and also more
featureful. And also smaller. Yay.
Some of the specialties of libev not commonly found elsewhere are:
- extensive and detailed, readable documentation (not doxygen garbage).
- fully supports fork, can detect fork in various ways and automatically
re-arms kernel mechanisms that do not support fork.
- highly optimised select, poll, epoll, kqueue and event ports backends.
- filesystem object (path) watching (with optional linux inotify support).
- wallclock-based times (using absolute time, cron-like).
- relative timers/timeouts (handle time jumps).
- fast intra-thread communication between multiple
event loops (with optional fast linux eventfd backend).
- extremely easy to embed.
- very small codebase, no bloated library.
- fully extensible by being able to plug into the event loop,
integrate other event loops, integrate other event loop users.
- very little memory use (small watchers, small event loop data).
- optional C++ interface allowing method and function callbacks
at no extra memory or runtime overhead.
- optional Perl interface with similar characteristics (capable
of running Glib/Gtk2 on libev, interfaces with Net::SNMP and
libadns).
- support for other languages (multiple C++ interfaces, D, Ruby,
Python) available from third-parties.
Examples of programs that embed libev: the EV perl module,
rxvt-unicode, gvpe (GNU Virtual Private Ethernet), the Deliantra MMORPG
server (http://www.deliantra.net/), Rubinius (a next-generation Ruby
VM), the Ebb web server, the Rev event toolkit.
TryCatch implements first class try catch semantics for Perl, without
source filters. The main benefits are:
- return from subroutines within the try blocks
- Moose type checking
