| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
| |
Now the _collections is statically built, the build dependencies are in proper
order and build works fine.
Commit Log from r83874:
Issue 9396. Apply functools.lru_cache in the place of the
random flushing cache in the re module.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The problem is that the re module is imported by sysconfig
and re needs functools which uses collections.OrderedDict()
but the _collectionsmodule.c code is not yet constructed
at this point in the build.
The likely best solution will be to include _collections
as part of the static build before the rest of the
boot-strapping.
|
|
|
|
| |
random flushing cache in the re module.
|
| |
|
|
|
|
|
|
| |
Updated to solution to use the functools.lru_cache().
Restores the API so that purge() is not needed
(because the cache never gets big).
|
|
|
|
| |
Just move the code and some variables.
|
|
|
|
|
| |
socket into the closed state without closing the underlying file
descriptor.
|
|
|
|
| |
#6231
|
| |
|
|
|
|
| |
Fix a regression introduced in test_sys by r83778.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
setup_context() replaces .pyc or .pyo filename suffix by .py, but it
didn't work if the filename contains a non-ascii character because the
function used the wrong unit for the length (number of characters
instead of the number of bytes).
With this patch, it uses unicode filenames instead of bytes filenames,
to fix the bug and to be fully unicode compliant.
|
|
|
|
| |
array.
|
|
|
|
| |
shutdown.
|
|
|
|
| |
Use ``encoding="unicode"`` to generate a Unicode string.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
82779,82855,83740,83789-83791,83797-83801,83803,83811,83827,83844 via svnmerge from
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r82779 | benjamin.peterson | 2010-07-10 14:45:08 -0500 (Sat, 10 Jul 2010) | 1 line
typo in attribute name #9217
........
r82855 | benjamin.peterson | 2010-07-13 16:27:38 -0500 (Tue, 13 Jul 2010) | 1 line
remove more extraneous commas #9245
........
r83740 | alexandre.vassalotti | 2010-08-05 01:58:36 -0500 (Thu, 05 Aug 2010) | 4 lines
Issue 5077: Update fixer for the other functions gone from the operator module.
Patch by Meador Inge.
........
r83789 | benjamin.peterson | 2010-08-07 17:45:14 -0500 (Sat, 07 Aug 2010) | 1 line
cleanup and use unicode consistently
........
r83790 | benjamin.peterson | 2010-08-07 17:52:06 -0500 (Sat, 07 Aug 2010) | 1 line
unicode literal
........
r83791 | benjamin.peterson | 2010-08-07 17:52:55 -0500 (Sat, 07 Aug 2010) | 1 line
.get() is pointless here
........
r83797 | benjamin.peterson | 2010-08-07 18:54:51 -0500 (Sat, 07 Aug 2010) | 1 line
add a function to find how a node is indented
........
r83798 | benjamin.peterson | 2010-08-07 18:55:28 -0500 (Sat, 07 Aug 2010) | 1 line
when splitting import statements, use correct indentation #9386
........
r83799 | benjamin.peterson | 2010-08-07 18:57:43 -0500 (Sat, 07 Aug 2010) | 1 line
double quotes
........
r83800 | benjamin.peterson | 2010-08-07 18:58:52 -0500 (Sat, 07 Aug 2010) | 1 line
add another test
........
r83801 | benjamin.peterson | 2010-08-07 19:02:10 -0500 (Sat, 07 Aug 2010) | 1 line
cleanup; style-nits
........
r83803 | benjamin.peterson | 2010-08-07 19:05:08 -0500 (Sat, 07 Aug 2010) | 1 line
slightly more explicit
........
r83811 | benjamin.peterson | 2010-08-07 22:56:44 -0500 (Sat, 07 Aug 2010) | 4 lines
Fix node.pre_order() to call the right method on its children.
This was a rather tragic copy-paste error.
........
r83827 | benjamin.peterson | 2010-08-08 08:12:48 -0500 (Sun, 08 Aug 2010) | 1 line
cause test to actually run and fix it
........
r83844 | benjamin.peterson | 2010-08-08 13:46:37 -0500 (Sun, 08 Aug 2010) | 1 line
fix whitespace
........
|
| |
|
| |
|
| |
|
|
|
|
| |
PROT_WRITE and PROT_EXEC.
|
| |
|
| |
|
| |
|
|
|
|
| |
patch from Gabriel Genellina
|
|
|
|
| |
urllib2 with urllib.request in comments.
|
| |
|
| |
|
|
|
|
| |
This is how it should work,wget and curl work like this way too. Old behavior was wrong.
|
|
|
|
| |
instead
|
| |
|
|
|
|
| |
name in a tuple subclass.
|
|
|
|
| |
at the top and by list concrete classes before abstract base classes.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
* Call Py_MakePendingCalls() before converting the filename from
wchar_t* to char*
* Use PyUnicode_AsUTF8String() instead of _PyUnicode_AsString()
|
|
|
|
| |
ncurses.
|
|
|
|
| |
Use PyUnicode_AsUTF8String() instead of _PyUnicode_AsString()
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
rather than do it by platform -- the proper skips are in place already.
|
| |
|
|
|
|
| |
occur, as a result of signed integer overflow.
|