* Replace dramatic footnote with in-line comment about possible round-off errors in logarithms of large numbers.
* Add comments to the pure python code equivalent.
* replace floor() with int() in the mathematical equivalent so the type is correct (should be an int, not a float).
* add abs() to the mathematical equivalent so that it matches the previous line that it is supposed to be equivalent to.
* make one combined example with a negative input.
Fix an issue in the tokenizer, where a file is opened by fd, but the underlying PyFileIO object wasn created with the closefd attribute true.
Also fix error handling for close() int _fileio.c . It was incorrect, looking for a negative refcount, and so errors weren't raised. This is why this issue wasn't caught.
There is a second reason why it isn't seen: Class IOBase in io.py has a try:/except: around the close() funtion in the __del__() method. This also masks these error conditions.
This issue was discovered by removing the _set_invalid_parameter_handler() fiddling, thus enabling the C runtime checks on windows.
Add missing Py_CHARMASK when calling isspace().
Found by enabling runtime tests on windows, by disabling the _set_invalid_parameter_handler() fiddling.
Merged revisions 67427,67431,67433,67435,67630,67652,67656-67657,67674-67675,67678-67679,67705-67706,67716,67723,67765-67771,67774,67776,67778 via svnmerge from
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r67427 | benjamin.peterson | 2008-11-28 16:07:41 -0600 (Fri, 28 Nov 2008) | 1 line
fix spelling in comment
........
r67431 | benjamin.peterson | 2008-11-28 17:14:08 -0600 (Fri, 28 Nov 2008) | 1 line
add a scripts directory; move things to it
........
r67433 | benjamin.peterson | 2008-11-28 17:18:48 -0600 (Fri, 28 Nov 2008) | 1 line
run svneol.py
........
r67435 | benjamin.peterson | 2008-11-28 17:25:03 -0600 (Fri, 28 Nov 2008) | 1 line
rename pre/post_order_mapping to pre/post_order_heads
........
r67630 | alexandre.vassalotti | 2008-12-06 21:51:56 -0600 (Sat, 06 Dec 2008) | 2 lines
Fix typo in the urllib2.HTTPDigestAuthHandler fixer.
........
r67652 | armin.ronacher | 2008-12-07 15:39:43 -0600 (Sun, 07 Dec 2008) | 5 lines
Added a fixer that cleans up a tuple argument to isinstance after the tokens
in it were fixed. This is mainly used to remove double occurrences of
tokens as a leftover of the long -> int / unicode -> str conversion.
........
r67656 | armin.ronacher | 2008-12-07 16:54:16 -0600 (Sun, 07 Dec 2008) | 3 lines
2to3: intern and reduce fixes now add the imports if missing. Because that is a common task the fixer_util module now has a function "touch_import" that adds imports if missing.
........
r67674 | benjamin.peterson | 2008-12-08 19:58:11 -0600 (Mon, 08 Dec 2008) | 1 line
copy permission bits when making backup files #4602
........
r67675 | benjamin.peterson | 2008-12-08 19:59:11 -0600 (Mon, 08 Dec 2008) | 1 line
add forgotten import
........
r67678 | benjamin.peterson | 2008-12-08 20:08:30 -0600 (Mon, 08 Dec 2008) | 1 line
fix #4602 for real
........
r67679 | armin.ronacher | 2008-12-09 00:54:03 -0600 (Tue, 09 Dec 2008) | 3 lines
Removed redudant code from the 2to3 long fixer. This fixes #4590.
........
r67705 | benjamin.peterson | 2008-12-11 13:04:08 -0600 (Thu, 11 Dec 2008) | 1 line
put trailers after a range call after the list()
........
r67706 | benjamin.peterson | 2008-12-11 13:17:57 -0600 (Thu, 11 Dec 2008) | 1 line
add html related modules to the fix_imports mapping
........
r67716 | benjamin.peterson | 2008-12-11 22:16:47 -0600 (Thu, 11 Dec 2008) | 1 line
consolidate tests
........
r67723 | benjamin.peterson | 2008-12-12 19:49:31 -0600 (Fri, 12 Dec 2008) | 1 line
fix name
........
r67765 | benjamin.peterson | 2008-12-14 14:05:05 -0600 (Sun, 14 Dec 2008) | 1 line
run fix_isinstance after fix_long and fix_unicode
........
r67766 | benjamin.peterson | 2008-12-14 14:13:05 -0600 (Sun, 14 Dec 2008) | 1 line
use run_order instead of order
........
r67767 | benjamin.peterson | 2008-12-14 14:28:12 -0600 (Sun, 14 Dec 2008) | 1 line
don't retain parenthesis if there is only one item left
........
r67768 | benjamin.peterson | 2008-12-14 14:32:30 -0600 (Sun, 14 Dec 2008) | 1 line
use insert_child()
........
r67769 | benjamin.peterson | 2008-12-14 14:59:10 -0600 (Sun, 14 Dec 2008) | 1 line
parenthesize doesn't belong in pygram or FixerBase
........
r67770 | alexandre.vassalotti | 2008-12-14 15:15:36 -0600 (Sun, 14 Dec 2008) | 2 lines
Fix typo: html.paser -> html.parser.
........
r67771 | benjamin.peterson | 2008-12-14 15:22:09 -0600 (Sun, 14 Dec 2008) | 1 line
altering .children needs to call changed()
........
r67774 | benjamin.peterson | 2008-12-14 15:55:38 -0600 (Sun, 14 Dec 2008) | 1 line
employ an evil hack to fix multiple names in the same import statement
........
r67776 | benjamin.peterson | 2008-12-14 16:22:38 -0600 (Sun, 14 Dec 2008) | 1 line
make a common mixin class for Test_imports and friends
........
r67778 | alexandre.vassalotti | 2008-12-14 17:48:20 -0600 (Sun, 14 Dec 2008) | 2 lines
Make fix_imports refactor multiple imports as.
........
................
Nick Coghlan [Sun, 14 Dec 2008 11:50:48 +0000 (11:50 +0000)]
Merged revisions 67750-67751 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r67750 | nick.coghlan | 2008-12-14 20:54:50 +1000 (Sun, 14 Dec 2008) | 1 line
Fix several issues relating to access to source code inside zipfiles. Initial work by Alexander Belopolsky. See Misc/NEWS in this checkin for details.
........
r67751 | nick.coghlan | 2008-12-14 21:09:40 +1000 (Sun, 14 Dec 2008) | 1 line
Issues #3167, #3682: tests for math.log and math.log10 were failing on
Solaris and OpenBSD. Fix this by handling special values and domain
errors directly in mathmodule.c, passing only positive nonspecial floats
to the system log/log10.
........
Mark Dickinson [Thu, 11 Dec 2008 18:03:03 +0000 (18:03 +0000)]
Issue #2173: fix build failure on OS X. device_encoding was returning an
empty string, causing an (invisible) LookupError on any attempt to write
to sys.stdout.
Issue #4084: Fix max, min, max_mag and min_mag Decimal methods to
give correct results in the case where one argument is a quiet NaN
and the other is a finite number that requires rounding.
Thanks Mark Dickinson.
........
Issue 4597: Fix several cases in EvalFrameEx where an exception could be
"raised" without setting x, err, or why to let the eval loop know.
........
r67685 | jeffrey.yasskin | 2008-12-09 23:35:02 -0800 (Tue, 09 Dec 2008) | 2 lines
Facundo Batista [Thu, 11 Dec 2008 04:20:07 +0000 (04:20 +0000)]
Issue #4084: Fix max, min, max_mag and min_mag Decimal methods to
give correct results in the case where one argument is a quiet NaN
and the other is a finite number that requires rounding.
Thanks Mark Dickinson.
Georg Brandl [Sun, 7 Dec 2008 15:30:06 +0000 (15:30 +0000)]
Merged revisions 67511,67536-67537,67543 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r67511 | vinay.sajip | 2008-12-04 00:22:58 +0100 (Thu, 04 Dec 2008) | 1 line
Issue #4384: Added logging integration with warnings module using captureWarnings(). This change includes a NullHandler which does nothing; it will be of use to library developers who want to avoid the "No handlers could be found for logger XXX" message which can appear if the library user doesn't configure logging.
........
r67536 | gregory.p.smith | 2008-12-04 21:21:09 +0100 (Thu, 04 Dec 2008) | 3 lines
Adds a subprocess.check_call_output() function to return the output from a
process on success or raise an exception on error.
........
r67537 | vinay.sajip | 2008-12-04 21:32:18 +0100 (Thu, 04 Dec 2008) | 1 line
Took Nick Coghlan's advice about importing warnings globally in logging, to avoid the possibility of race conditions: "This could deadlock if a thread spawned as a side effect of importing a module happens to trigger a warning. warnings is pulled into sys.modules as part of the interpreter startup - having a global 'import warnings' shouldn't have any real effect on logging's import time."
........
r67543 | gregory.p.smith | 2008-12-05 03:27:01 +0100 (Fri, 05 Dec 2008) | 2 lines
rename the new check_call_output to check_output. its less ugly.
........
Georg Brandl [Sun, 7 Dec 2008 15:15:22 +0000 (15:15 +0000)]
Merged revisions 67463,67572,67576,67628 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r67463 | skip.montanaro | 2008-12-01 02:55:22 +0100 (Mon, 01 Dec 2008) | 1 line
typo in comment
........
r67572 | georg.brandl | 2008-12-05 10:23:14 +0100 (Fri, 05 Dec 2008) | 2 lines
#4458: recognize "-" as an argument, not a malformed option in gnu_getopt().
........
r67576 | georg.brandl | 2008-12-05 13:09:41 +0100 (Fri, 05 Dec 2008) | 2 lines
#4529: fix parser's validation for try-except-finally statements.
........
r67628 | skip.montanaro | 2008-12-07 03:16:00 +0100 (Sun, 07 Dec 2008) | 1 line
Georg Brandl [Sun, 7 Dec 2008 15:06:20 +0000 (15:06 +0000)]
Merged revisions 67531-67532,67538,67553-67554,67556-67557,67571,67574-67575,67579-67580,67591,67597,67608,67631 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
Speed up Python (according to pybench and 2to3-on-itself) by 1-2% by caching
whether any thread has tracing turned on, which saves one load instruction in
the fast_next_opcode path in PyEval_EvalFrameEx(). See issue 4477.
........
Move __import__ to the bottom of the functions list.
It doesn't make sense for such a fundamental document to have
the most obscure function listed at the top.
........
Tweak verbiage and markup for relative/absolute import. Backport candidate.
(I guess *all* changes to this file should be backported to the 3.0 branch.)
Issue #1055234: cgi.parse_header(): Fixed parsing of header parameters to
support unusual filenames (such as those containing semi-colons) in
Content-Disposition headers.
........
projects / python / log