Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
This package would build outside a bulk build, but a few flaws in
the makefile prevented it from building in all cases, bulk builds
being one such exception. It should build ok now.
|
|
This fixes build failure on Linux and others.
|
|
on Linux and others.
|
|
Remove it from the default list for the rest.
|
|
Long options have been removed from all programs.
Use standard single-letter POSIX-compatible options.
lmdbg-sym:
- Doesn't accept progname as an argument anymore.
Use -P option instead.
- option --with-so was replaced with -s.
SORRY FOR INCOMPATIBILITIES WITH PREVIOUS RELEASES.
----------------------------------
lmdbg:
- -M option was added. lmdbg(1) is now a meta tools which is able
to not only collecting memory leaks. See the manual page for
details.
- -n option was added
- -T and -B options were added
- Regression tests were added for -M
A lot of improvements and grammar fixes in manual pages.
Thanks to Nikolai Krot for this.
lmdbg-stat, tests/test.sh:
- Fixes for Solaris
lmdbg-run:
- options -T and -B were added to lmdbg-run.
- option -n was added. It disables logging on startup.
Sending SIGUSR1 signal to the process enables it.
-f option was added, synonym for -p.
lmdbg-grep.1:
- documentation for "bytes" variable
If set, LMDBG_MODULES_AWK and LMDBG_STRIP_AWK environment variables
are used as a path to AWK interpreter by "lmdbg-modules" and
"lmdbg-strip" programs respectively.
More regression tests
|
|
DragonFly needs the same socklet_t modification as the other BSDs.
|
|
|
|
There are several unreferenced functions in cqual which a modern gcc
complains about using -Werror. After trying to fix a few of them, it
was easier just to stop considering warnings as errors.
|
|
|
|
|
|
clisp-syscalls still doesn't build (it's broken on NetBSD too), but
it is known now that it will need gettext-lib if/when the other issues
are resolved.
|
|
|
|
DragonFly X86_64 can't build 32-bit binaries as the Makefile requests.
|
|
This is a regularly-scheduled bug-fix release.
alias: shortcut command matching show shadowing works properly (issue3104)
bundlerepo: don't write branch cache to disk
clone: don't save user's password in .hg/hgrc (issue3122)
convert/bzr: correctly handle divergent nested renames (issue3089)
convert/bzr: fix test of divergent nested renames for bzr 1.x
convert/svn: fix URL quoting issue with svn 1.7
convert: handle trailing slashes in filemap better (issue3124)
diff: '\ No newline at end of file' is also not part of the header
diff: --ignore-blank-lines was too enthusiastic
dirstate: fix case-folding identity for traditional Unix
graft: disallow grafting grafted csets in specific situations (issue3091)
graft: preserve original source in subsequent grafts
hook: flush stdout before redirecting to stderr
import: fix parent selection when importing merges
largefiles: ensure destination directory exists before findfile links to there
largefiles: file storage should be relative to repo, not relative to cwd
largefiles: fix 'hg clone . ../foo' OSError abort
largefiles: fix rename (issue3093)
largefiles: treat status of cache missed largefiles as "missing" correctly
largefiles: use "ui.configlist()" to get largefiles.patterns configuration
largefiles: write .hg/largefiles/ files atomically
makedate: wrong timezone offset if DST rules changed this year (issue2511)
mdiff/patch: fix bad hunk handling for unified diffs with zero context
posix: add extended support for OS X path folding
posix: fix findexe() to check for file type and access
rebase: ensure target is not taken as external (issue3085)
rebase: treat nullmerge as a special case in rebasestate (issue3046)
revset: balance %l or-expressions (issue3129)
revset: follow(nosuchfile) should give an empty set (issue3114)
subrepo: fix git status false positive (issue3109)
subrepo: use correct paths for subrepos with ..-relative paths on windows
update: don't clobber untracked files with wrong casing
url: handle file://localhost/c:/foo "correctly"
|
|
Change from previous:
0.364 04/12/2010
- Allow skipping the rest of a test using
plan skip_all => $reason
|
|
|
|
Bump PKGREVISION
|
|
|
|
OpenPGM is an open source implementation of the Pragmatic General
Multicast (PGM) specification in RFC 3208 available at www.ietf.org.
PGM is a reliable and scalable multicast protocol that enables
receivers to detect loss, request retransmission of lost data, or
notify an application of unrecoverable loss. PGM is a
receiver-reliable protocol, which means the receiver is responsible
for ensuring all data is received, absolving the sender of reception
responsibility. PGM runs over a best effort datagram service,
currently OpenPGM uses IP multicast but could be implemented above
switched fabrics such as InfiniBand.
PGM is appropriate for applications that require duplicate-free
multicast data delivery from multiple sources to multiple receivers.
PGM does not support acknowledged delivery, nor does it guarantee
ordering of packets from multiple senders.
PGM is primarly used on internal networks to help integrate disparate
systems through a common communication platform. A lack of IPv4
multicast-enabled infrastructure leads to limited capability for
internet applications, IPv6 promotes multicast to be a part of the
core functionality of IP but may still be disabled on core routers.
Support of Source-Specific Multicast (SSM) allows for improved WAN
deployment by allowing end-point router filtering of unwanted source
traffic.
|
|
Changes from previous:
0.03 2010-11-02
Improve reinstatement of $1, $2 so they don't leak outside of the
scope of your blocks
Remove a couple stray references to named rules
0.02 2010-10-24
Reinstate the back-compat that Path-Dispatcher 1.00 broke by setting
$1, $2, etc. for you
|
|
Changes from previous:
1.04 Thu Sep 1 2011
Explicitly depend on Test::Fatal (reported by Jérôme Quelin [rt.cpan.org #70666])
Convert all tests from Test::Exception to Test::Fatal
1.03 Tue Aug 30 2011
Provisional support for arbitrary payloads, not just code blocks, as results
See https://github.com/sartak/path-dispatcher/blob/master/t/200-payload.t
Eventually blocks will be deprecated, but not for a while. Internally blocks are
rewritten as payloads so if you're doing deep magic with PD you may be exposed to
that implementation detail.
1.02 Tue Nov 2 2010
Various improvements to the handling of $match->leftover with undef
1.01 Sun Oct 24 2010
Skip t/026-named-captures.t if you don't have 5.10.1
Give matches a ->parent for inspecting an Under prefix rule's
captures etc
1.00 Sun Oct 17 2010
Bump to 1.0 to indicate back compat breakage
Add match->named($key) for pulling out a named capture
Use Try::Tiny to avoid swallowing exceptions (rafl)
0.16 Sun Oct 17 2010
** BACKWARDS INCOMPATIBLE CHANGE: Pass the match object into rule
** blocks rather than assigning $1, $2, $3. Use ->pos(1), (2), (3),
** etc. instead. This cleans up the code a lot and makes
** Path-Dispatcher more flexible.
Add support for named captures
Make all attributes read-only. If you need something to be read-write,
I am open to such a change. Just let me know!
Make _dispatcher_rule private
Remove trace/readable_attribute/name support. This wants to be
implemented in a different, better, more comprehensive way.
Add documentation for shell tab completion
Plenty other updated documentation
|
|
From Opolo Remo in private mail.
Bump PKGREVISION.
|
|
* Version 1.23 (released 2011-11-25) [stable]
** stringprep.h: Now #include's sys/types.h instead of unistd.h for ssize_t.
Some systems (e.g., Mingw with MSVC 9) does not have unistd.h.
** idn-free.h: Protect prototypes with 'extern "C"' marker.
Reported by Bittner Ede <bittner.ede@euronetrt.hu>.
** doc: Update link to experimental TLD tables.
The new link is <https://github.com/gnuthor/tldchk>.
** Update gnulib files and translations.
** QA: Improved cyclo output. Update GTK-DOC files. Various bugfixes.
** API and ABI is backwards compatible with the previous version.
|
|
Tested on x86_64 Linux with USE_BUILTIN.libuuid=(yes|no).
|
|
of a faked up pkg-config file.
|
|
pkgsrc/mk/buildlink3/bsd.builtin.mk and change the fake uuid.pc so its
similar to what's used in other builtins (and a little bit simpler).
|
|
|
|
|
|
Changes from 1.1.0 to 1.2.0:
o Add AVX2 instructions (rev 11 of Intel AVX reference) (#227).
o Allow 64-bit LFS/LGS/LSS.
o Improve LAR instruction support (#224).
o Default win64 .xdata to nobase, add support for ..imagebase (#135).
o Fix TIMES relocation handling.
o Fix no-suffix push and pop in GAS mode (#212).
|
|
0.65 Mon Sep 20 03:32:46 2010
- Remove unnecessary test boilerplate that caused trouble in the perl core.
|
|
|
|
Release 8.20 21-Oct-2011
------------------------
The main change in this release is the inclusion of Zoltan Herczeg's
just-in-time compiler support, which can be accessed by building PCRE with
--enable-jit. Large performance benefits can be had in many situations. 8.20
also fixes an unfortunate bug that was introduced in 8.13 as well as tidying up
a number of infelicities and differences from Perl.
|
|
Changes from previous:
0.25 2011-08-17
- Add timeit() function to time individual behaviours.
0.24 2011-06-18
- Update dependencies to pick up multiple bug fixes, new behaviours,
in various Test:: libraries.
|
|
|
|
|
|
Tested on current and Linux.
|
|
Tested on netBSD current and Linux.
|
|
NetBSD current.
|
|
|
|
Tested on NetBSD current and Linux.
|
|
Tested on NetBSD current, Linux and Solaris-10.
|
|
Observed in PR pkg/45553.
|
|
|
|
|
|
|
|
|
|
Bump PKGREVISION to 3. Thanks Dmitry Matveev.
|
|
bootstrapped environment on NetBSD, MKPROFILE defaults to no, but
using the base system's make library it defaults to yes. This seems
like a bug, but I'm not sure where to fix it so I'm going to stick to
working around it for the moment.
|
|
not on other platforms.
|