summaryrefslogtreecommitdiffstats
path: root/Python
Commit message (Collapse)AuthorAgeFilesLines
* Fix issue #1590864, multiple threads and fork() can cause deadlocks, byThomas Wouters2009-09-161-19/+12
| | | | | | | | | | | | | | | | | | | | | | acquiring the import lock around fork() calls. This prevents other threads from having that lock while the fork happens, and is the recommended way of dealing with such issues. There are two other locks we care about, the GIL and the Thread Local Storage lock. The GIL is obviously held when calling Python functions like os.fork(), and the TLS lock is explicitly reallocated instead, while also deleting now-orphaned TLS data. This only fixes calls to os.fork(), not extension modules or embedding programs calling C's fork() directly. Solving that requires a new set of API functions, and possibly a rewrite of the Python/thread_*.c mess. Add a warning explaining the problem to the documentation in the mean time. This also changes behaviour a little on AIX. Before, AIX (but only AIX) was getting the import lock reallocated, seemingly to avoid this very same problem. This is not the right approach, because the import lock is a re-entrant one, and reallocating would do the wrong thing when forking while holding the import lock. Will backport to 2.6, minus the tiny AIX behaviour change.
* revert unintended changesBenjamin Peterson2009-09-091-86/+8
|
* tabbifyBenjamin Peterson2009-09-091-8/+86
|
* revert r74699 since it loses useful error informationBenjamin Peterson2009-09-071-2/+13
|
* PyObject_GetIter can set an error for its self just fineBenjamin Peterson2009-09-061-13/+2
|
* better col_offsets for "for" statements with tuple unpacking #6704Benjamin Peterson2009-08-151-11/+12
| | | | Patch from Frank Wierzbicki.
* - Issue #6624: yArg_ParseTuple with "s" format when parsing argument withSean Reifscheider2009-08-011-1/+1
| | | | NUL: Bogus TypeError detail string.
* Better name a variable: 'buf' seems to imply a mutable buffer.Amaury Forgeot d'Arc2009-07-251-7/+7
|
* the Slice in x[::] has to have step as None to help the interpreterBenjamin Peterson2009-07-201-1/+15
|
* Issue #6415: Fixed warnings.warn sagfault on bad formatted string.Hirokazu Yamamoto2009-07-171-0/+2
|
* Update issue 6070 patch to match the patch that was actually testedR. David Murray2009-07-071-1/+3
| | | | on Windows.
* Issue 6070: when creating a compiled file, after copying the mode bits, onR. David Murray2009-07-071-1/+3
| | | | | posix zap the execute bit in case it was set on the .py file, since the compiled files are not directly executable on posix. Patch by Marco N.
* when print() gets unicode arguments, sep and end should be unicode by ↵Benjamin Peterson2009-07-021-18/+74
| | | | default #4618
* small optimization: avoid popping the current block until we have toBenjamin Peterson2009-07-011-5/+5
|
* Fix error handling in PyCode_Optimize, by Alexander Schremmer at EuroPython ↵Georg Brandl2009-06-291-6/+9
| | | | sprint.
* Update docstrings for sys.getdlopenflags() and sys.setdlopenflags().Alexandre Vassalotti2009-06-291-10/+12
|
* use stack macrosBenjamin Peterson2009-06-281-2/+2
|
* add two generic macros for peeking and setting in the stackBenjamin Peterson2009-06-281-0/+2
|
* fix grammarBenjamin Peterson2009-06-231-1/+1
|
* remove tmpname support since it's no longer usedBenjamin Peterson2009-06-212-28/+0
|
* don't need to add the name 'lambda' as assignedBenjamin Peterson2009-06-211-2/+1
|
* remove unused 'encoding' member from the compiler structBenjamin Peterson2009-06-171-4/+0
|
* update ast versionBenjamin Peterson2009-06-131-2/+2
|
* when no module is given in a 'from' relative import, make ImportFrom.module NULLBenjamin Peterson2009-06-133-20/+20
|
* give a better error message when deleting ()Benjamin Peterson2009-06-131-4/+7
|
* set Print.values to NULL if there are no valuesBenjamin Peterson2009-06-131-9/+12
|
* allow importing from a module named None if it has an 'as' clauseBenjamin Peterson2009-06-131-4/+6
|
* prevent import statements from assigning to NoneBenjamin Peterson2009-06-131-11/+26
|
* keep the slice.step field as NULL if no step expression is givenBenjamin Peterson2009-06-131-8/+1
|
* Add const qualifier to PyErr_SetFromErrnoWithFilename and toAlexandre Vassalotti2009-06-121-2/+2
| | | | PyErr_SetFromErrnoWithUnicodeFilename.
* remove check for case handled in sub-functionBenjamin Peterson2009-06-111-3/+0
|
* revert r73361Benjamin Peterson2009-06-111-0/+4
|
* remove duplicate checkBenjamin Peterson2009-06-111-4/+0
|
* remove error checks already done in set_context()Benjamin Peterson2009-06-081-25/+0
|
* remove useless assertionBenjamin Peterson2009-06-081-1/+0
|
* Issue #5330: C functions called with keyword arguments were not reported byAntoine Pitrou2009-05-301-2/+16
| | | | the various profiling modules (profile, cProfile). Patch by Hagen Fürstenau.
* Fix nearly all compilation warnings under Apple gcc-4.0. Tested with OPT="-gJeffrey Yasskin2009-05-291-12/+0
| | | | | | -Wall -Wstrict-prototypes -Werror" in both --with-pydebug mode and --without. There's still a batch of non-prototype warnings in Xlib.h that I don't know how to fix.
* further hint to where the open docs really arePhilip Jenvey2009-05-281-1/+1
|
* teach the peepholer about SETUP_WITHBenjamin Peterson2009-05-261-0/+3
|
* Allow multiple context managers in one with statement, as proposedGeorg Brandl2009-05-252-39/+61
| | | | | | | in http://codereview.appspot.com/53094 and accepted by Guido. The construct is transformed into multiple With AST nodes so that there should be no problems with the semantics.
* take into account the fact that SETUP_WITH pushes a finally blockBenjamin Peterson2009-05-251-1/+1
|
* add a SETUP_WITH opcodeBenjamin Peterson2009-05-253-57/+57
| | | | | It speeds up the with statement and correctly looks up the special methods involved.
* handle errors from _PyObject_LookupSpecial when __get__ failsBenjamin Peterson2009-05-251-4/+6
|
* Issue #6042:Jeffrey Yasskin2009-05-232-65/+14
| | | | | | | | | | | | | | | lnotab-based tracing is very complicated and isn't documented very well. There were at least 3 comment blocks purporting to document co_lnotab, and none did a very good job. This patch unifies them into Objects/lnotab_notes.txt which tries to completely capture the current state of affairs. I also discovered that we've attached 2 layers of patches to the basic tracing scheme. The first layer avoids jumping to instructions that don't start a line, to avoid problems in if statements and while loops. The second layer discovered that jumps backward do need to trace at instructions that don't start a line, so it added extra lnotab entries for 'while' and 'for' loops, and added a special case for backward jumps within the same line. I replaced these patches by just treating forward and backward jumps differently.
* support building with subversion 1.7 #6094Benjamin Peterson2009-05-231-1/+1
|
* Issue #3527: Removed Py_WIN_WIDE_FILENAMES which is not used any more.Hirokazu Yamamoto2009-05-171-6/+2
|
* Move news item to correct section, remove spurious 'see below'R. David Murray2009-05-131-1/+1
| | | | from docstring.
* Issue 5994: add docstrings to marshal.R. David Murray2009-05-131-5/+81
|
* don't ignore exceptions from _PyObject_LengthHintBenjamin Peterson2009-05-091-1/+3
|
* convert some more special methods to use _PyObject_LookupSpecialBenjamin Peterson2009-05-091-12/+7
|
generated by cgit v0.12 at 2024-10-23 20:45:56 (GMT)