summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* SF patch #1467512, fix double free with triple quoted string in standard build.Neal Norwitz2006-04-109-37/+45
| | | | | | This was the result of inconsistent use of PyMem_* and PyObject_* allocators. By changing to use PyObject_* allocator almost everywhere, this removes the inconsistency.
* Get rid of warnings about using chars as subscriptsNeal Norwitz2006-04-102-3/+3
| | | | on Alpha (and possibly other platforms) by using Py_CHARMASK().
* Fix some warnings on HP-UX when using cc/aCCNeal Norwitz2006-04-104-10/+11
|
* I wonder if we can be too graceful? One oughta be enough. :-)Neal Norwitz2006-04-101-2/+1
|
* Python on OS X 10.3 and above now uses dlopen() (via dynload_shlib.c)Anthony Baxter2006-04-097-8/+17
| | | | | | to load extension modules and now provides the dl module. As a result, sys.setdlopenflags() now works correctly on these systems. (SF patch #1454844)
* Patch #1466993: remove wrong comment in socket.pyGeorg Brandl2006-04-091-8/+0
|
* Fix tests so they pass in -R modeNeal Norwitz2006-04-093-7/+18
|
* Get test_optparse to pass in -R mode by stop changing state (COLUMNS env var)Neal Norwitz2006-04-091-0/+7
|
* Convert test_compare to use unittest. Hopefully we can find outNeal Norwitz2006-04-092-124/+28
| | | | why this breaks on openbsd sometimes.
* Fix problem (not checking return result) reported by CoverityNeal Norwitz2006-04-091-1/+4
|
* Fix a couple of strings that were no-ops. urllib.open_file was a docstringNeal Norwitz2006-04-092-2/+2
| | | | in 2.4, so put it back. The string in telnetlib looks like a comment.
* glob('anything*/') would fail because isdir is in os.path, not os.Neal Norwitz2006-04-093-1/+11
|
* test_timeout(): This test was added during Bug Day, but disabledTim Peters2006-04-081-20/+36
| | | | | | | | | soon after because the gmail address it connects to started timing out on all the buildbot slaves. Rewrote the test to produce a warning message (instead of failing) when the address times out. Also removed the special case for Windows -- this test started to work on Windows as soon as bug 1462352 was fixed.
* Define WIN32, to get proper struct alignment in WinSock2.h.Martin v. Löwis2006-04-081-2/+2
| | | | Fixes #1462352. Will backport.
* Fix bsddb.db.DBError derived exceptions so they can be unpickled.Gregory P. Smith2006-04-084-6/+100
| | | | | Also adds some backwards compatibility when compiling _bsddb.c on earlier python versions (needed for pybsddb).
* Don't try to build ctypes when configuration of libffi fails.Thomas Heller2006-04-071-2/+4
|
* Configure Modules/_ctypes/libffi just before building the _ctypes extension.Thomas Heller2006-04-071-2/+13
| | | | This avoids configuring it when it is not build by setup.py.
* Fix a few XXX markersAndrew M. Kuchling2006-04-071-8/+7
|
* Adjust compileall -x option to Makefile.pre.in. Fixes #1465093.Martin v. Löwis2006-04-071-1/+1
|
* Regenerate.Martin v. Löwis2006-04-071-1/+1
|
* missing 'self' from TextCalendar.prweek.Anthony Baxter2006-04-071-1/+1
|
* minor error in uudecode main error handlingAnthony Baxter2006-04-071-1/+1
|
* Patch #1462080: Conditionalize some NETLINK defines.Martin v. Löwis2006-04-061-0/+11
|
* Whitespace normalization.Tim Peters2006-04-061-2/+2
|
* Expose RTLD_LOCAL and RTLD_GLOBAL always from the _ctypes extension module.Thomas Heller2006-04-063-12/+15
| | | | | | | If RTLD_LOCAL is not #defined in any header file (Windows), set it to 0. If RTLD_GLOBAL is not #defined, set it equal to RTLD_LOCAL. This should fix ctypes on cygwin.
* Fix unfinished paragraph; markup fixAndrew M. Kuchling2006-04-061-4/+6
|
* Expand the PEP 353 section; various smaller changesAndrew M. Kuchling2006-04-061-14/+64
|
* Bug #1464658: make clear that PyList_GetItem doesn't take negative indices.Georg Brandl2006-04-061-2/+3
|
* Bug #1465619: rot13 codec decodes to a Unicode string.Georg Brandl2006-04-061-1/+1
|
* Bug #1465600: note encoding issue in ZipFile.write().Georg Brandl2006-04-061-4/+11
|
* Replace a few more 'char' types with 'signed char', to fix test failuresThomas Heller2006-04-061-5/+5
| | | | on platforms (ppc debian) where 'char' is unsigned.
* This change shouldn't have any functional effect. Coverity wasNeal Norwitz2006-04-061-1/+1
| | | | complaining because it seemed like parameters_iter could be NULL.
* Fix unchecked return result from Coverity.Neal Norwitz2006-04-061-0/+2
|
* Remove dead code (reported by HP compiler).Neal Norwitz2006-04-061-8/+5
| | | | Can probably be backported if anyone cares.
* Handle ssize_tNeal Norwitz2006-04-061-2/+2
|
* Handle error conditions from PyString_ConcatAndDel().Neal Norwitz2006-04-061-1/+6
|
* Handle ssize_tNeal Norwitz2006-04-061-3/+4
| | | | | | No need to INCREF then let PyString_ConcatAndDel() DECREF. Just use PyString_Concat(). Handle error condition if we can't concat.
* Hopefully this makes test_urllib2 have inconsistent leak resultsNeal Norwitz2006-04-062-2/+3
|
* I don't think we know of any tests that really leak anymoreNeal Norwitz2006-04-061-2/+2
| | | | (other than those in leakers).
* Fix for failure of test_urllib2 breaking test_mimetypes (SF bug 1464978)Anthony Baxter2006-04-062-156/+166
| | | | will backport.
* In some environments (under screen, in a chroot) curses doesn't supportAnthony Baxter2006-04-061-5/+7
| | | | | | | | mouse events. This makes the test fail. Catch that case and don't run the tests. Should make the debian/ubuntu buildbots that run in a chroot work again. Will backport to release24-maint.
* Bug #1451341: find fuzzy marks correctly.Georg Brandl2006-04-061-1/+1
|
* remove more cruft no longer neededFred Drake2006-04-061-4/+2
|
* remove much of the Python-version compatibility cruft; the minimum PythonFred Drake2006-04-064-141/+64
| | | | version this should support is Python 2.3
* revert incomplete changes committed by accidentFred Drake2006-04-061-3/+4
|
* update URL to reflect new websiteFred Drake2006-04-062-5/+4
|
* Explicitely use 'signed char', to avoid problems on platforms with unsigned ↵Thomas Heller2006-04-051-26/+26
| | | | char type.
* Explicitely use 'signed char', to avoid problems on platforms with unsigned ↵Thomas Heller2006-04-051-2/+2
| | | | char type.
* One test still fails on the ia64 debian box.Thomas Heller2006-04-051-1/+1
|
* Fixed error in comment for new PY_SSIZE_T_MIN.Tim Peters2006-04-051-1/+1
|
generated by cgit v0.12 at 2024-12-03 09:16:35 (GMT)