summaryrefslogtreecommitdiff
path: root/databases/py-bsddb3/buildlink3.mk
AgeCommit message (Collapse)AuthorFilesLines
2010-06-02Changes 5.0.0:adam1-5/+6
* Support for Berkeley DB 5.0. * Drop support for Python 3.0. * Now you can use TMPDIR env variable to override default test directory ("/tmp"). * Versioning of C API. If you use the code from C, please check the bsddb_api->api_version number against PYBSDDB_API_VERSION macro. * In C code, the bsddb_api->dbsequence_type component is always available, even if the Berkeley DB version used doesn't support sequences. In that case, the component will be NULL. * In C code, "DBSequenceObject_Check()" macro always exists, even if the Berkeley DB version used doesn't suport sequences. In that case, the test macro always returns "false". * For a long time, the API has been accesible via C using "_bsddb.api" or "_pybsddb.api". If you are using Python >=2.7, you acquire access to that API via the new Capsule protocol (see "bsddb.h"). If you use the C API and upgrade to Python 2.7 and up, you must update the access code (see "bsddb.h"). The Capsule protocol is not supported in Python 3.0, but pybsddb 5.0.x doesn't support Python 3.0 anymore. * Capsule support was buggy. The string passed in to PyCapsule_New() must outlive the capsule. * Solve an "Overflow" warning in the testsuite running under python 2.3. * When doing a complete full-matrix test, any warning will be considered an error.
2009-03-20Simply and speed up buildlink3.mk files and processing.joerg1-13/+6
This changes the buildlink3.mk files to use an include guard for the recursive include. The use of BUILDLINK_DEPTH, BUILDLINK_DEPENDS, BUILDLINK_PACKAGES and BUILDLINK_ORDER is handled by a single new variable BUILDLINK_TREE. Each buildlink3.mk file adds a pair of enter/exit marker, which can be used to reconstruct the tree and to determine first level includes. Avoiding := for large variables (BUILDLINK_ORDER) speeds up parse time as += has linear complexity. The include guard reduces system time by avoiding reading files over and over again. For complex packages this reduces both %user and %sys time to half of the former time.
2008-09-18bump BUILDLINK_API_DEPENDS to the new revisiondrochner1-3/+3
(The dependency is already there, due to the propagated db4 dependency. Just make it explicit.)
2008-09-06Recursive PKGREVISION/ABI-depends bump for db4 4.6->4.7 update (shlibwiz1-2/+2
name change).
2008-01-12db4 update related revision bumpadam1-2/+2
2007-06-08PKGREVISION bump for db4 shlib name change.wiz1-2/+2
2006-08-08we can't have PYTHON_VERSIONS_ACCEPTED in buildlink3 files becausedrochner1-4/+1
this might conflict with definitions in Makefiles noticed by Jeremy C. Reed
2006-07-08Change the format of BUILDLINK_ORDER to contain depth information as well,jlam1-2/+2
and add a new helper target and script, "show-buildlink3", that outputs a listing of the buildlink3.mk files included as well as the depth at which they are included. For example, "make show-buildlink3" in fonts/Xft2 displays: zlib fontconfig iconv zlib freetype2 expat freetype2 Xrender renderproto
2006-07-08Track information in a new variable BUILDLINK_ORDER that informs usjlam1-1/+2
of the order in which buildlink3.mk files are (recursively) included by a package Makefile.
2006-06-16Fixed pkglint warnings.rillig1-8/+8
2006-04-12Aligned the last line of the buildlink3.mk files with the first line, sorillig1-2/+2
that they look nicer.
2006-04-06Over 1200 files touched but no revisions bumped :)reed1-3/+3
RECOMMENDED is removed. It becomes ABI_DEPENDS. BUILDLINK_RECOMMENDED.foo becomes BUILDLINK_ABI_DEPENDS.foo. BUILDLINK_DEPENDS.foo becomes BUILDLINK_API_DEPENDS.foo. BUILDLINK_DEPENDS does not change. IGNORE_RECOMMENDED (which defaulted to "no") becomes USE_ABI_DEPENDS which defaults to "yes". Added to obsolete.mk checking for IGNORE_RECOMMENDED. I did not manually go through and fix any aesthetic tab/spacing issues. I have tested the above patch on DragonFly building and packaging subversion and pkglint and their many dependencies. I have also tested USE_ABI_DEPENDS=no on my NetBSD workstation (where I have used IGNORE_RECOMMENDED for a long time). I have been an active user of IGNORE_RECOMMENDED since it was available. As suggested, I removed the documentation sentences suggesting bumping for "security" issues. As discussed on tech-pkg. I will commit to revbump, pkglint, pkg_install, createbuildlink separately. Note that if you use wip, it will fail! I will commit to pkgsrc-wip later (within day).
2006-02-05Recursive revision bump / recommended bump for gettext ABI change.joerg1-1/+2
2006-01-06DEPENDS and PKGREVISION bumps because db4's library name changed.wiz1-2/+2
2005-01-24sync PYTHON_VERSIONS_ACCEPTED value with Makefile.seb1-2/+2
2004-05-11On NetBSD current with native threads it seems this python module can'tseb1-0/+26
be imported by a non-thread aware python without it core dumping. Hence restrict accepted python packages to threaded ones if native threads are supported by the platform. Bump PKGREVISION for that. Provide a buildlink3.mk file mostly to help bl3ified packages to pick up the right python packages with respect to this threading issue.