BUILDLINK_DEPENDS.zlib instead of adding to it.
fixes a problem where a buildlink dependency would be added on both
zlib>=1.1.4nb1 and zlib>=1.2.1, resulting in various buildlink
breakage.
the normal case when BUILDLINK_DEPENDS.<pkg> isn't specified, it receives
a value only once due to the multiple inclusion protection in the
bulldlink3.mk files. In the case where a package includes several
buildlink3.mk files that each want a slightly different version of another
dependency, having BUILDLINK_DEPENDS.<pkg> be a list allows for the
strictest <pkg> dependency to be matched.
Changes in 0.15.1:
------------------
* Require GnuTLS >= 1.0.0
Changes in 0.15:
----------------
* Better Win32 instructions.
* Fixed bug when disconnecting and then connecting again using the same
LmConnection.
* Documentation typos and errors fixed by Ross Burton.
* Better SSL support, now gives the application the possibility to decide
if it wants to connect on bogus SSL certificates.
* The SSL API changed a bit, take a look at the API docs.
* Fixed timestamp error on FreeBSD.
installs are something like databases that need to be handled automatically
by the package system during installation of modules, not by the end user.
Bump PKGREVISION to 1. (As a side effect, the package gets simplified a lot).
There are no packages in pkgsrc installing pango modules at the moment.
When they come, we can easily add a framework to handle them, as has been
done for x11/gtk2.
other platforms (a YES should have been a NO, plus it was in the wrong
place!). This should fix PR 24129.
* Add more sophisticated ncurses version detection for platforms that
actually have real ncurses in the base system. This should be a win
for FreeBSD systems.
* Downgrade the required ncurses to 5.0. The previous bump done in
revision 1.8 wasn't necessary since ncurses never had a dependency on
libiconv.
need are placed in the LIBS variable rather than OTHERLDFLAGS.
Other Perl modules use LIBS, and lang/perl5/module.mk clobbers
OTHERLDFLAGS.
Thanks to Johnny Lam for suggesting the fix.
updated package update bumped the zlib shared lib major, which required
that BUILDLINK_DEPENDS.zlib be bumped as well. Rather then requiring
zlib>=1.2.1 for packages that can use the built-in zlib on *BSD systems,
allow those built-in versions to satisfy zlib>=1.1.4nb1 dependencies,
and only require the latest version if no suitable zlib is found.
Bug fixed in 1.8.2:
* A (well known) portability bug slipped in the changes made to
install-sh in Automake 1.8.1. The broken install-sh would refuse to
install anything on Tru64.
* Fix install rules for conditionally built python files. (This never
really worked.)
Bug fixed in 1.8.1:
* Bugs introduced by 1.8:
- Fix Config.pm import error with old Perl versions (at least
5.005_03). One symptom is that aclocal could not find its macro
directory.
- Automake 1.8 used `mkdir -m 0755 -p --' to ensure that directories
created by `make install' are always world readable, even if the
installer happens to have an overly restrictive umask (e.g. 077).
This was a mistake and has been reverted. There are at least two
reasons why we must not use `-m 0755':
- it causes special bits like SGID to be ignored,
- it may be too restrictive (some setups expect 775 directories).
- Fix aclocal to honor definitions located in files which have been
m4_included manually. aclocal 1.8 had been updated to check
m4_included files for new requirements, but forgot that these
m4_included files can also provide new definitions.
Note that if you have such a setup, we recommend you get rid of
it. In the past, there was a reason to m4_include files manually:
aclocal used to duplicate entire M4 files into aclocal.m4, even
files that were distributed. Some packages were therefore
m4_including the distributed file directly, and playing some
tricks to ensure aclocal would not copy that file to aclocal.m4,
in order to limit the amount of duplication. Since aclocal 1.8.x
will precisely output m4_includes for local M4 files, we recommend
that you clean up your setup, removing all manual m4_includes and
letting aclocal output them.
- Output detailed menus in the Info version if the Automake manual,
so that Emacs can locate the indexes.
- configure.ac and configure were listed twice in DIST_COMMON (an
internal variable where Automake lists configury files to
distribute). This was harmless, but unaesthetic.
- Use `chmod a-w' instead of `chmod -w' as the latter honors umask.
This was an issue only in the Automake package itself, not in
its output.
- Automake assumed that all AC_CONFIG_LINKS arguments had the form
DEST:SRC. This was wrong, as some packages do
AC_CONFIG_LINKS($computedlinks). This version no longer abort in
that situation.
- Contrary to mkinstalldirs, $(mkdir_p) was expecting exactly one
argument. This caused two kinds of failures:
- Rules installing data in a conditionally defined directory
failed when that directory was undefined. In this case no
argument was supplied.
- `make installdirs' failed, because several directories were
passed to $(mkdir_p). This was an issue only on platform
were $(mkdir_p) is implemented with `install-sh -d'.
$(mkdir_p) as been changed to accept 0 or more arguments, as
mkinstalldirs did.
* Long-standing bugs:
- Fix an unexpected diagnostic occurring when users attempt
to override some internal variables that Automake appends to.
- aclocal now scans configure.ac for macro definitions (PR/319).
- Fix a portability issue with OSF1/Tru64 Make. If a directory
distributes files which are outside itself (this usually occurs
when using AC_CONFIG_AUX_DIR([../dir]) to use auxiliary files
from a parent package), then `make distcheck' fails due to an
optimization performed by OSF1/Tru64 Make in its VPATH handling.
(tests/subpkg2.test failure)
- Fix another portability issue with Sun and OSF1/Tru64 Make.
In a VPATH-build configuration, `make install' would install
nobase_ files to wrong locations.
- Fix a Perl `uninitialized value' diagnostic occurring when
automake complains that a Texinfo file does not have a
@setfilename statement.
- Erase config.status.lineno during `make distclean'. This file
can be created by config.status. Automake already knew about
configure.lineno, but forgot config.status.lineno.
- Distribute all files, even those which are built and installed
conditionally. This change affects files listed in conditionally
defined *_HEADERS and *_PYTHON variable (unless they are nodist_*)
as well as those listed in conditionally defined dist_*_DATA,
dist_*_JAVA, dist_*_LISP, and dist_*_SCRIPTS variables.
- Fix AM_PATH_LISPDIR to avoid \? in sed regular expressions; it
doesn't conform to POSIX.
- Normalize help strings for configure variables and options added
by Automake macros.
* Anticipation:
- Check for python2.4 in AM_PATH_PYTHON.
* Spurious failures in test suite:
- tests/libtool5.test, tests/ltcond.test, tests/ltcond2.test,
tests/ltconv.test: fix failures with CVS Libtool.
- tests/aclocal6.test: fix failure if autom4te.cache is disabled.
- tests/txinfo24.test, tests/txinfo25.test, tests/txinfo28.test:
fix failures with old Texinfo versions.
Added HOMEPAGE.
Changes since 2.28:
2.40 Tue Dec 30 20:38:59 CST 2003
[FIXES]
* Test::Harness::Straps should now properly quote on VMS.
[ENHANCEMENTS]
* prove now takes a -l option to add lib/ to @INC. Now when you're
building a module, you don't have to do a make before you run
the prove. Thanks to David Wheeler for the idea.
[INTERNALS]
* Internal functions corestatus() and canonfailed() prepended with
underscores, to indicate such.
* Gratuitous text-only changes in Test::Harness::Iterator.
* All tests now do their use_ok() in a BEGIN block. Some of the
use_ok() calls were too much of a hassle to put into a BEGIN block,
so I changed them to regular use calls.
2.38 Mon Nov 24 22:36:18 CST 2003
Released. See changes below.
2.37_03 Tue Nov 18 23:51:38 CST 2003
[ENHANCEMENTS]
* prove -V now shows the Perl version being used.
* Now there's a HARNESS_DEBUG flag that shows diagnostics as the
harness runs the tests. This is different from HARNESS_VERBOSE,
which shows test output, but not information about the harness
itself.
* Added _command_line() to the Strap API.
[FIXES]
* Bad interaction with Module::Build: The strap was only checking
$ENV{HARNESS_PERL_SWITCHES} for definedness, but not emptiness.
It now also strips any leading or trailing whitesapce from the
switches.
* Test::Harness and prove only quote those parms that actually need
to be quoted: Have some whitespace and aren't already quoted.
2.36 Fri Nov 14 09:24:44 CST 2003
[FIXES]
* t/prove-includes.t properly ignores PROVE_SWITCHES that you may
already have set.
2.35_02 Thu Nov 13 09:57:36 CST 2003
[ENHANCEMENTS]
* prove's --blib now works just like the blib pragma.
2.35_01 Wed Nov 12 23:08:45 CST 2003
[FIXES]
* Fixed taint-handling and path preservation under MacOS. Thanks to
Schwern for the patch and the tests.
* Preserves case of -t or -T in the shebang line of the test.
[ENHANCEMENTS]
* Added -t to prove analogous to Perl's -t. Removed the --taint
switch.
* prove can take default options from the PROVE_SWITCHES variable.
* Added HARNESS_PERL to allow you to specify the Perl interpreter
to run the tests as.
* prove's --perl switch sets the HARNESS_PERL on the fly for you.
* Quotes the switches and filename in the subprogram. This helps
with filenames with spaces that are subject to shell mangling.
2.34 Sat Nov 8 22:09:15 CST 2003
[FIXES]
* Allowed prove to run on Perl versions < 5.6.0.
[ENHANCEMENTS]
* Command-line switches to prove may now be stacked.
* Added check for proper Pod::Usage version.
* "make clean" does a better job of cleaning up after itself.
2.32 Fri Nov 7 09:41:21 CST 2003
Test::Harness now includes a powerful development tool to help
programmers work with automated tests. The prove utility runs
test files against the harness, like a "make test", but with many
advantages:
* prove is designed as a development tool
Perl users typically run the test harness through a makefile via
"make test". That's fine for module distributions, but it's
suboptimal for a test/code/debug development cycle.
* prove is granular
prove lets your run against only the files you want to check.
Running "prove t/live/ t/master.t" checks every *.t in t/live, plus
t/master.t.
* prove has an easy verbose mode
To get full test program output from "make test", you must set
"HARNESS_VERBOSE" in the environment. prove has a "-v" option.
* prove can run under taint mode
prove's "-T" runs your tests under "perl -T".
* prove can shuffle tests
You can use prove's "--shuffle" option to try to excite problems
that don't show up when tests are run in the same order every time.
* Not everything is a module
More and more users are using Perl's testing tools outside the
context of a module distribution, and may not even use a makefile at
all.
Prove requires Pod::Usage, which is standard after Perl 5.004.
I'm very excited about prove, and hope that developers will begin
adopting it to their coding cycles. I welcome your comments at
andy@petdance.com.
There are also some minor bug fixes in Test::Harness itself, listed
below in the 2.31_* notes.
2.31_05 Thu Nov 6 14:56:22 CST 2003
[FIXES]
- If a MacPerl script had a shebang with -T, the -T wouldn't get
passed as a switch.
- Removed the -T on three *.t files, which didn't need them, and
which were causing problems.
- Conditionally installs bin/prove, depending on whether Pod::Usage
is available, which prove needs.
- Removed old leftover code from Makefile.PL.
2.31_04 Mon Nov 3 23:36:06 CST 2003
Minor tweaks here and there, almost ready to release.
2.31_03 Mon Nov 3 08:50:36 CST 2003
[FEATURES]
- prove is almost feature-complete. Removed the handling of
--exclude for excluding certain tests. It may go back in the
future.
- prove -d is now debug. Dry is prove -D.
2.31_02 Fri Oct 31 23:46:03 CST 2003
[FEATURES]
- Added many more switches to prove: -d for dry run, and -b for
blib.
[FIXES]
- T:H:Straps now recognizes MSWin32 in $^0.
- RT#3811: Could do regex matching on garbage in _is_test().
Fixed by Yves Orton
- RT#3827: Strips backslashes from and normalizes @INC entries
for Win32. Fixed by Yves Orton.
[INTERNALS]
- Added $self->{_is_macos} to the T:H:Strap object.
- t/test-harness.t sorts its test results, rather than relying on
internal key order.
2.31_01
[FEATURES]
- Added "prove" script to run a test or set of tests through the
harness. Thanks to Curtis Poe for the foundation.
[DOCUMENTATION]
- Fixed POD problem in Test::Harness::Assert
2.30 Thu Aug 14 20:04:00 CDT 2003
No functional changes in this version. It's only to make some doc
tweaks, and bump up the version number in T:H:Straps.
[DOCUMENTATION]
- Changed Schwern to Andy as the maintainer.
- Incorporated the TODO file into Harness.pm proper.
- Cleaned up formatting in Test::Harness::Straps.
2.29 Wed Jul 17 14:08:00 CDT 2003
- Released as 2.29.
2.28_91 Sun Jul 13 00:10:00 CDT 2003
[ENHANCEMENTS]
- Added support for HARNESS_OK_SLOW. This will make a significant
speedup for slower connections.
- Folded in some changes from bleadperl that spiff up the
failure reports.
[INTERNALS]
- Added some isa_ok() checks to the tests.
- All Test::Harness* modules are used by use_ok()
- Fixed the prototype for the canonfailed() function, not that
it matters since it's never called without parens.
2.28_90 Sat Jul 05 20:21:00 CDT 2003
[ENHANCEMENTS]
- Now, when you run a test harnessed, the numbers don't fly by one
at a time, one update per second. This significantly speeds
up the run time for running thousands of tests. *COUGH*
Regexp::Common *COUGH*
Changes:
BUG FIXES:
* When checking if a process exists, accept EPERM as indicating it
exists but isn't ours. This allows watching the directory of a
different user, subject only to filesystem permissions.
Reported by Ernst Bachmann.
* Correct a bug introduced in 2.12 that made distccd fail to start
when run by root because setgid failed. Fix from Wayne Davison.
* --without-gnome, --without-gtk now turn those options off rather
than on.
DOCUMENTATION:
* New man page for distccmon-text, by Shri Shrikumar and Martin
Pool.
(branching 13 January 2004, from /branches/1.0-stabilization)
http://svn.collab.net/repos/svn/tags/0.36.0
User-visible changes:
* add cancellation suport to svnadmin and svnlook (r8222)
* runtime 'store-password' option renamed to 'store-auth-creds' (r8014)
* 'svn blame' changes:
- now shows correct revision info (r8035-6)
- responds to cancellation better (r8129)
* svnserve changes:
- added '--inetd' option; now required to speak with stdin/stdout (r8205)
- added '--listen-port' and '--listen-host' options (r8001-2)
- removed '-u' option (r8003)
- ignore SIGPIPE (no more repos lockups when you terminate a pipe) (r8140)
* lots of Book work (many newly-documented Apache and svnserve topics)
Developer-visible changes:
* bugfix: svnserve network crash (r8142)
* bugfix: return result_rev from svn_client_checkout correctly (r8096)
* bugfix: fs history harvesting code (r8154)
* bugfix: memory leak in mod_dav_svn (r8223)
* bugfixes in edge-cases of status and update (r8114-5)
* make 'svn blame' work with 18n and uri-escaped filenames (r8023, 8030, 8040)
* small bugfixes to authentication system (r8006, r8235)
* standardize error message formatting (r8218)
* load RA modules as foo.so.0, not foo.so (r8098)
* various core API changes:
- use constructor for svn_client_cxt_t (r8053-4)
- anchor/target may use NULL for target (r8216)
- stop using apr_ symbols (r8219)
- rename to 'svn_repos_authz_func_t' (r8213)
- add pool parameter to finish_report and abort_report (r8215)
* numerous changes to Perl and Java bindings, to keep up with C API.
