summaryrefslogtreecommitdiff
path: root/lang
AgeCommit message (Collapse)AuthorFilesLines
2005-01-05Changes 8.4.9:adam4-30/+29
* Fixed segfault in info vars trivial matching branch (new in 8.4.8) * Fixed the treatment of backslashes in file join on Windows * Improved readdir_r detection and usage on unix * Fixed potential unix fileevent issue on 64-bit systems * Remove file normalize on tcl_findLibrary search path uniqification added in 8.4.8 * Ensure tilde paths are not returned specially by 'glob' * Fixed clipping of partially transparent images on buttons on unix to avoid X error * Fix Tk_PhotoPut(Zoomed)Block overlay compositing of partially transparent areas on blank targets
2005-01-04Set LC_ALL to "C". Otherwise, the package doesn't build due tominskim1-2/+2
warnings from perl. See PR pkg/24211 for related discussion.
2005-01-04Enable SQLitemanu2-4/+4
2005-01-04Fix instances of insecure use of /tmp that is subject to symlink attacksjlam5-5/+88
due to race conditions [CAN-2004-0976]. Also fix builds on *BSD boxes without a hostname set. Bump the PKGREVISION to 1.
2005-01-03Drop maintainership of perl-5.6.x. We should move pkgsrc to usingjlam1-2/+2
perl58 for everything, and remove this package altogether.
2005-01-02Use PKGDIRMODE and DOCMODE instead of 555 and 444, respectively.minskim1-4/+4
Bump PKGREVISION because the default value of PKGDIRMODE is 755, which is different from 555.
2005-01-02Use PKGDIRMODE and DOCMODE instead of 555 and 444, respectively.minskim1-4/+4
Bump PKGREVISION because the default value of PKGDIRMODE is 755, which is different from 555.
2004-12-30Honor LDFLAGS when building shared libraries on Linux. Previously,minskim8-45/+61
python2[34]{,-pth} failed on Linux when Berkeley DB from pkgsrc was used.
2004-12-30Use the same configure flags for NetBSD 1.[56] as for 1.[56].x (i.e.kristerw1-3/+4
adjust the tests to understand that 1.[56] is not written as 1.[56].0). Bump PKGREVISION. This solves PR pkg/28499.
2004-12-29Update lang/perl58 to perl-5.8.6. Selected hanges from version 5.8.5jlam5-30/+39
include: * By default, platforms that have native threads will build a threaded perl. Note that you will likely have to rebuild your Perl modules after this update unless your Perl is already threaded. * The perl interpreter is now more tolerant of UTF-16-encoded scripts. * Several core modules were updated. * Perl has a new -dt command-line flag, which enables threads support in the debugger. * "foreach" on threads::shared array used to be able to crash Perl. This bug has now been fixed. * A regexp in "STDOUT"'s destructor used to coredump, because the regexp pad was already freed. This has been fixed. * Using "delete" on an array no longer leaks memory. A "pop" of an item from a shared array reference no longer causes a leak. * "eval_sv()" failing a taint test could corrupt the stack - this has been fixed. * On platforms with 64 bit pointers numeric comparison operators used to erroneously compare the addresses of references that are overloaded, rather than using the overloaded values. This has been fixed. * From now on all applications embedding perl will behave as if perl were compiled with -DPERL_USE_SAFE_PUTENV.
2004-12-29Add and enable py24-html-docs.minskim1-1/+2
2004-12-29Import py24-html-docs.minskim4-0/+37
This is HTML documentation for Python 2.4.
2004-12-28- Fix an error where -lpython does not get defined on Darwin and causesadrianp1-1/+5
the compile of python modules to fail. - Fixes PR# 28605 - Ok'ed recht@
2004-12-27use ${PKG_SYSCONFDIR} rather than hardcoding ${PREFIX}/etcjdolecek1-2/+2
Fixes PR pkg/28718 by Amitai Schlair
2004-12-26Update to 1.6.7:wiz4-36/+27
Changes since Guile 1.6.6 (changes in 1.6.7): * Changes to the distribution ** A build problem has been fixed. Previously, on some systems, the build would fail when libguile-ltdl couldn't be found during the build. This should now be fixed. * Changes to Scheme functions and syntax ** array-map! and array-map-in-order! now correctly require at least one source A mistake caused a call with just one source array to be rejected, this has been fixed. ** string->number and octal constant bignums An incorrect bignum size calculation has been fixed, this caused overflow errors in string->number on bases other than 2, 10 and 16, including octal literal constants in code or the reader. ** SRFI-1 alist-delete equality argument order fixed. In the srfi-1 module alist-delete and alist-delete!, the order of the arguments to the "=" procedure now matches the SRFI-1 specification. ** SRFI-13 string-any and string-every tail calls string-any and string-every now make a tail call to their predicate function on reaching the last character in the string, per the SRFI-13 specification.
2004-12-24Looks like the rpath problem is a bit more serious than I initially thoughtjmmv2-1/+36
(it currently affects packages using X11, but I guess it will also cause problems when using pkgviews, for example). So, instead of fixing it on a package basis, create some wrappers in the buildlink directory that parse CFLAGS and LDFLAGS contents and convert them to ocaml flags (basically, prepend -ccopt to each of them).
2004-12-24Adjust dependencies after ocaml has been splitted into three differentjmmv1-2/+2
packages: ocaml, ocaml-graphics and labltk. Bump PKGREVISION.
2004-12-24Do not depend on Tk nor X11. Two new packages will be added to providejmmv9-394/+93
the missing bits, namely labltk and ocaml-graphics, respectively. To simplify this, add a Makefile.common. Bump PKGREVISION to 8. Per discussion with wiz@ a while ago.
2004-12-24Remove patch to non-used file; patch-ae's content exists intaca2-22/+1
devel/ruby-curses package. No functional change to ruby18 package itself.
2004-12-21py-basicproperty-0.6.3adarcy4-0/+202
The BasicProperty/BasicTypes system provides a mechanism for intelligent property-based modeling of problem domains in Python 2.2+. The idea behind property-based domain modeling is to simplify the definition of objects/data structures so that concerns such as initialization, data-type checking and coercion, run-time introspection, and domain specific storage/retrieval operations can be abstracted out from the modeling operations.
2004-12-21Add py-basicpropertydarcy1-1/+2
2004-12-20actually bump PKGREVISION of perl58 and BUILDLINK_RECOMMENDED to addressgrant2-4/+4
PR pkg/28619. this should have been part of the PKGREVISION bump of p5-*.
2004-12-20remove workaround for Darwin missing poll.h as a more correct fix isgrant1-4/+2
applied globally now.
2004-12-20since perl is now built with threads on most platforms, the perl archlibgrant1-1/+2
module directory has changed (eg. "darwin-2level" vs. "darwin-thread-multi-2level"). binary packages of perl modules need to be distinguishable between being built against threaded perl and unthreaded perl, so bump the PKGREVISION of all perl module packages and introduce BUILDLINK_RECOMMENDED for perl as perl>=5.8.5nb5 so the correct dependencies are registered and the binary packages are distinct. addresses PR pkg/28619 from H. Todd Fujinaka.
2004-12-19Add back ONLY_FOR_PLATFORM that was removed by mistake when the packagekristerw1-1/+3
was updated in r1.17.
2004-12-19fix missed from previous: Darwin 7.x doesn't appear to want thegrant2-8/+20
"-u PyMac_Error" when linking the shared library.
2004-12-19all of these use dlopen().grant4-4/+8
2004-12-19Darwin build fixes:grant3-14/+25
- ignore poll() - needs dlopen.buildlink3.mk - pass -fPIC -fno-common to CC for objects being linked into a shared library
2004-12-19Create a modifier PERL5_OPTIONS that specifies properties of the perljlam2-1/+33
installation needed when including perl5/buildlink3.mk. The only option currently supported is "threads", which implies that a perl that supports threads is required. The requirement is checked at pre-install time using an INSTALL script template by both the package build and the binary package. Add PERL5_OPTIONS+=threads to both devel/p5-SDL and graphics/p5-GD since those modules require a perl that supports threads.
2004-12-18Native threading on VAX isn't ready for primetime in NetBSD>=2.0.jlam1-1/+12
Build Perl without threads-support until VAX native threads are known to work. The hack was added to the Makefile instead of to hacks.mk since it sets a variable which is used within the package Makefile.
2004-12-18g/c MAINTAINER (use one from Makefile.common)jdolecek1-2/+1
2004-12-18claim MAINTAINERjdolecek1-2/+2
2004-12-18claim MAINTAINERjdolecek1-2/+2
2004-12-17Update php5 to 5.0.3. This maintenance release addressing several veryjdolecek4-8/+13
serious security issues, as well as bunch of non-critical bug fixes. All PHP5 users are strongly encouraged to upgrade to this version. Detailed change list at: http://www.php.net/ChangeLog-5.php#5.0.3
2004-12-17move DISTINFO_FILE setting from Makefile.common to Makefile.php,jdolecek2-3/+4
so that it PECL extension packages could override it properly
2004-12-16Don't always set DIST_SUBDIR but only if RUBY_DIST_SUBDIR isn't empty.taca1-1/+3
This change dosen't affect any current ruby based packages.
2004-12-15Remove duplicate "the" in comment.wiz2-4/+4
2004-12-14"-ldl" is needed to build on Linux.minskim1-1/+2
2004-12-12completely ignore system php.ini for all pear-related operations; thisjdolecek4-5/+16
is necessary to avoid being subject to e.g. open_basedir or safe_mode settings
2004-12-12abort do-install if pear install failsjdolecek1-2/+2
2004-12-12use PLIST_SRC+= so that setting PLIST_SRC setting in extension Makefilejdolecek1-2/+2
would have an effect; fixes php4-apc
2004-12-12Change ALL_TARGET to BUILD_TARGET.kristerw1-2/+2
2004-12-11The Open Group says that localtime_r() doesn't need to ensure thatjlam3-3/+22
the tz* related variables are initialized when it is invoked. In fact, Solaris does do this while NetBSD 2.0 does not. Explicitly call tzset() before calling localtime_r() to initialize the local timezone data. This fixes the output of: perl -e '$t = localtime($^T); print "$t\n";' ; date to always print the time in the local timezone twice instead of the first time in UTC and the second in the local timezone. Bump the PKGREVISION.
2004-12-11Add buildlink3.mk file, needed by wip/darcs.wiz1-0/+23
Closes PR 26624 by Roland Illig.
2004-12-10On NetBSD/alpha, building perl with -mieee causes perl to not passjlam1-1/+12
the test for integer.pm (pkg/28498). Until this is fixed in either NetBSD, GCC or perl, strip out -mieee from the compiler command line. This "fixes" pkg/28498.
2004-12-10On VAX, feeding a base "NaN" to nawk causes nawk to core dump sincejlam1-1/+17
it tries to interpret it as a number, which causes an FP exception. Fix this by replacing "NaN" with "*NaN*" which hides the problem. This makes perl Configure and build correctly on NetBSD-2.0/vax. Fix provided by John Klos in private email.
2004-12-10Don't bother with using perl5/module.mk and PERL5_PACKLIST to generatejlam2-25/+26
the PLIST and use custom post-install code instead. This avoids needing to guess at the correct value PERL5_ARCHLIB before perl has been configured and built. This should fix PR pkg/28433.
2004-12-10Nuke bsd.perl.mk... we compute these values where they're needed injlam2-13/+3
perl5/buildlink3.mk already, so no need to generate this file to be included by bsd.pkg.mk anymore. Bump the PKGREVISION.
2004-12-10This is configured with -with-thread. Include pthread.buildlink3.mkuebayasi1-1/+2
here explicitly.
2004-12-07I haven't maintained this in a very, very long time.groo1-2/+2
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-04 06:00:47 +0000