Database (Py_UNICODE_TOLOWER, Py_UNICODE_ISDECIMAL, and others) now accept
and return characters from the full Unicode range (Py_UCS4).
The differences from Python code are few:
- unicodedata.numeric(), unicodedata.decimal() and unicodedata.digit()
now return the correct value for large code points
- repr() may consider more characters as printable.
* Use the current filesystem encoding instead of always using utf-8
* Enable the test on Mac OS X
* Use TESTFN_UNENCODABLE and TESTFN_UNICODE instead of arbitrary filenames
* To decode a filename, use strict error handler instead surrogateescape for
mbcs encoding (on Windows)
* Use TESTFN_UNENCODABLE (if available) for the directory name
Skip the test if no non-ascii filename can be created.
------------------------------------------------------------------------
r79925 | nick.coghlan | 2010-04-10 16:24:36 +0200 (sam. 10 avril 2010)
Try to turn some buildbots green by allowing test_multiprocessing to
pass even if it hits the sys.exc_clear code in the threading module, and
improve the test coverage by making the ctypes dependencies a bit more
granular (two of the cited ctypes objects don't exist on my system)
------------------------------------------------------------------------
* On non-Windows OSes: the constructor accepts bytes filenames
and use surrogateescape for unicode filenames
* On Windows: use GetFileAttributesW() instead of GetFileAttributesA()
* Use 0xff byte on non-Windows OSes
* mbcs is now really strict by default: i closed the issue #850997, so use the
filesystem encoding and not Latin-1
* Rename TESTFN_UNICODE_UNENCODEABLE to TESTFN_UNENCODEABLE
original read request had been satisfied, which can block indefinitely
when the underlying raw IO channel is e.g. a socket. Report and original
patch by Jason V. Miller.
------------------------------------------------------------------------
r81149 | florent.xicluna | 2010-05-13 23:40:01 +0200 (jeu. 13 mai 2010) | 2 lignes
Better test skipping, with message in the log.
------------------------------------------------------------------------
r81150 | florent.xicluna | 2010-05-13 23:41:05 +0200 (jeu. 13 mai 2010) | 2 lignes
Improve test feedback to troubleshoot issue #8423 on OS X.
------------------------------------------------------------------------
r81151 | florent.xicluna | 2010-05-14 01:46:48 +0200 (ven. 14 mai 2010) | 2 lignes
Revert changeset r81150 which helped diagnose issue #8423 on some OS X buildbot.
------------------------------------------------------------------------
r82056 | florent.xicluna | 2010-06-17 22:30:56 +0200 (jeu. 17 juin 2010) | 2 lignes
Add few words about test.test_genericpath.CommonTest
------------------------------------------------------------------------
------------------------------------------------------------------------
r80411 | florent.xicluna | 2010-04-23 19:59:10 +0200 (ven. 23 avril 2010) | 2 lignes
Remove ImportWarnings filters. They become obsolete after r79310, issue #8205.
------------------------------------------------------------------------
r80412 | florent.xicluna | 2010-04-23 20:10:12 +0200 (ven. 23 avril 2010) | 2 lignes
Fix the "regrtest -s" switch.
------------------------------------------------------------------------
r81140 | florent.xicluna | 2010-05-13 19:05:29 +0200 (jeu. 13 mai 2010) | 3 lignes
Add sensible information about the OS X platform to diagnose issue #8423:
test_pep277 fails on "x86 Tiger" buildbot but not on "PPC Tiger".
------------------------------------------------------------------------
r81141 | florent.xicluna | 2010-05-13 20:16:06 +0200 (jeu. 13 mai 2010) | 2 lignes
Revert the additional OS X information (r81140). Keep the endianness information.
------------------------------------------------------------------------
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.
An argparse option parser created with a prefix_chars that did not
include a '-' would happily add -h and --help options, and then throw
an error when it tried to format the help because the - was an invalid
prefix character. This patch makes it use the first character of
prefix_chars as the character for the help options if and only if '-'
is not one of the valid prefix_chars.
Fix by Theodore Turocy, unit tests by Catherine Devlin.
where 'id -G' and posix.getgroups() returned the same information,
but one of the sources contains duplicate information. Rewrite the
check using sets instead of lists.
output with no type specifier failed to match the str output:
- format(complex(-0.0, 2.0), '-') omitted the real part from the output,
- format(complex(0.0, 2.0), '-') included a sign and parentheses.
Cmd used to blindly chop off the last character of every input line. If
the input reached EOF and there was no final new line, it would truncate
the last character of the last command. This fix instead strips trailing
\r\n from the input lines. While this is a small behavior change, it
should not break any working code, since feeding a '\r\n' terminated
file to Cmd would previously leave the \r's on the lines, resulting
in failed command execution.
I wrote the unit test in preparation for a PyOhio TeachMe session
run by Catherine Devlin, and we can thank Catherine and the PyOhio
session attendees for the fix. I've added Catherine to the Acks file
for organizing and leading the TeachMe session, out of which we will
hopefully get some new contributors.
* added basic test for basic commands
* removed duplication of command docs, and moved them to their implementation
* unified and useful display of exceptions
* output messages and errors using overridable methods (also fixes#1503502)
been increased from 100 to 500 and the cache replacement policy has changed
from simply clearing the entire cache on overflow to randomly forgetting 20%
of the existing cached compiled regular expressions. This is a performance
win for applications that use a lot of regular expressions and limits the
impact of the performance hit anytime the cache is exceeded.
with the fact that getgroups(2) might return
more that MAX_GROUPS on OSX.
See the issue (and python-dev archives) for the
gory details. Summarized: OSX behaves rather oddly
and Apple says this is intentional.
the Crash Reporter on OSX test_subprocess will trigger
the reporter.
This patch prints a warning when the Crash Reporter will
get triggered intentionally, which should avoid confusing
people.
using WindowsError in a try/except. Only add WindowsError to the list of
exceptions to catch when we are actually running on Windows.
Additionally, add a call that was left out in test_posixpath.
Thanks Amaury, Antoine, and Jason.
Added Windows support for os.symlink when run on Windows 6.0 or greater,
aka Vista. Previous Windows versions will raise NotImplementedError
when trying to symlink.
Includes numerous test updates and additions to test_os, including
a symlink_support module because of the fact that privilege escalation
is required in order to run the tests to ensure that the user is able
to create symlinks. By default, accounts do not have the required
privilege, so the escalation code will have to be exposed later (or
documented on how to do so). I'll be following up with that work next.
Note that the tests use ctypes, which was agreed on during the PyCon
language summit.
This required moving the class from importlib/abc.py into
importlib/_bootstrap.py and jiggering some code to work better with the class.
This included changing how the file finder worked to better meet import
semantics. This also led to fixing importlib to handle the empty string from
sys.path as import currently does (and making me wish we didn't support that
instead just required people to insert '.' instead to represent cwd).
It also required making the new set_data abstractmethod create
any needed subdirectories implicitly thanks to __pycache__ (it was either this
or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
method or have set_data with no data arg mean to create a directory).
Lastly, as an optimization the file loaders cache the file path where the
finder found something to use for loading (this is thanks to having a
sourceless loader separate from the source loader to simplify the code and
cut out stat calls).
Unfortunately test_runpy assumed a loader would always work for a module, even
if you changed from underneath it what it was expected to work with. By simply
dropping the previous loader in test_runpy so the proper loader can be returned
by the finder fixed the failure.
At this point importlib deviates from import on two points:
1. The exception raised when trying to import a file is different (import does
an explicit file check to print a special message, importlib just says the path
cannot be imported as if it was just some module name).
2. the co_filename on a code object is not being set to where bytecode was
actually loaded from instead of where the marshalled code object originally
came from (a solution for this has already been agreed upon on python-dev but has
not been implemented yet; issue8611).
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82492 | victor.stinner | 2010-07-03 15:36:19 +0200 (sam., 03 juil. 2010) | 3 lines
Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
ensure that the input string length is a multiple of the frame size
........
1) #8271: when a byte sequence is invalid, only the start byte and all the
valid continuation bytes are now replaced by U+FFFD, instead of replacing
the number of bytes specified by the start byte.
See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (pages 94-95);
2) 5- and 6-bytes-long UTF-8 sequences are now considered invalid (no changes
in behavior);
3) Change the error messages "unexpected code byte" to "invalid start byte"
and "invalid data" to "invalid continuation byte";
4) Add an extensive set of tests in test_unicode;
5) Fix test_codeccallbacks because it was failing after this change.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82403 | benjamin.peterson | 2010-06-30 12:11:08 -0500 (Wed, 30 Jun 2010) | 1 line
mark test depending on ref counting
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r81499 | georg.brandl | 2010-05-24 16:29:07 -0500 (Mon, 24 May 2010) | 1 line
#8016: add the CP858 codec (approved by Benjamin). (Also add CP720 to the tests, it was missing there.)
........
r81506 | benjamin.peterson | 2010-05-24 17:04:53 -0500 (Mon, 24 May 2010) | 1 line
set svn:eol-style
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r81380 | brett.cannon | 2010-05-20 13:37:55 -0500 (Thu, 20 May 2010) | 8 lines
Turned out that if you used explicit relative import syntax
(e.g. from .os import sep) and it failed, import would still try the implicit
relative import semantics of an absolute import (from os import sep). That's
not right, so when level is negative, only do explicit relative import
semantics.
Fixes issue #7902. Thanks to Meador Inge for the patch.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r81465 | georg.brandl | 2010-05-22 06:29:19 -0500 (Sat, 22 May 2010) | 2 lines
Issue #3924: Ignore cookies with invalid "version" field in cookielib.
........
r81466 | georg.brandl | 2010-05-22 06:31:16 -0500 (Sat, 22 May 2010) | 1 line
Underscore the name of an internal utility function.
........
r81468 | georg.brandl | 2010-05-22 06:43:25 -0500 (Sat, 22 May 2010) | 1 line
#8635: document enumerate() start parameter in docstring.
........
r81679 | benjamin.peterson | 2010-06-03 16:21:03 -0500 (Thu, 03 Jun 2010) | 1 line
use a set for membership testing
........
r81735 | michael.foord | 2010-06-05 06:46:59 -0500 (Sat, 05 Jun 2010) | 1 line
Extract error message truncating into a method (unittest.TestCase._truncateMessage).
........
r81760 | michael.foord | 2010-06-05 14:38:42 -0500 (Sat, 05 Jun 2010) | 1 line
Issue 8302. SkipTest exception is setUpClass or setUpModule is now reported as a skip rather than an error.
........
r81868 | benjamin.peterson | 2010-06-09 14:45:04 -0500 (Wed, 09 Jun 2010) | 1 line
fix code formatting
........
r82183 | benjamin.peterson | 2010-06-23 15:29:26 -0500 (Wed, 23 Jun 2010) | 1 line
cpython only gc tests
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82233 | r.david.murray | 2010-06-25 20:06:44 -0400 (Fri, 25 Jun 2010) | 2 lines
#4640: add a test to optparse that proves issue is invalid.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82229 | benjamin.peterson | 2010-06-25 17:58:47 -0500 (Fri, 25 Jun 2010) | 1 line
mark implementation detail as such
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82225 | benjamin.peterson | 2010-06-25 15:34:01 -0500 (Fri, 25 Jun 2010) | 1 line
mark implementation detail as such
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82165 | benjamin.peterson | 2010-06-22 15:26:20 -0500 (Tue, 22 Jun 2010) | 1 line
must force gc here
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82075 | jean-paul.calderone | 2010-06-18 16:00:17 -0400 (Fri, 18 Jun 2010) | 12 lines
Revert r60115
This revision introduced quoting for strings containing | based
on a misunderstanding of the commonly used quoting rules used
on Windows.
| is interpreted by cmd.exe, not by the MS C runtime argv initializer.
It only needs to be quoted if it is part of an argument passed through
cmd.exe.
See issue1300, issue7839, and issue8972.
........
dealing with unary minus applied to a constant. The removed code was
mutating the ST, causing a second compilation to fail. (The peephole
optimizer already takes care of optimizing this case, so there's no
lost optimization opportunity here.)
svn+ssh://pythondev@svn.python.org/python/trunk
........
r82039 | r.david.murray | 2010-06-16 21:36:52 -0400 (Wed, 16 Jun 2010) | 10 lines
#8720: fix inspect regression by teaching getsourcefile about linecache.
The fix for issue 4050 caused a regression: before that fix, source
lines in the linecache would eventually be found by inspect. After the
fix inspect reports an error earlier, and the source isn't found.
The fix for the fix is to have getsourcefile look in the linecache for
the file and return the psuedo-filename if the source is there, just as
it already returns it if there is a PEP 302 loader.
........
mode raises unicode errors. The encoder only supports "strict" and "replace"
error handlers, the decoder only supports "strict" and "ignore" error handlers.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r81967 | mark.dickinson | 2010-06-13 11:50:29 +0100 (Sun, 13 Jun 2010) | 4 lines
Issue #8986: erfc was raising OverflowError on Linux for arguments in
the (approximate) range (-27.3, 30.0), as a result of an escaped errno
value.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r80578 | nick.coghlan | 2010-04-29 00:29:06 +1000 (Thu, 29 Apr 2010) | 1 line
Issue 7490: make IGNORE_EXCEPTION_DETAIL also ignore details of the module containing the exception under test (original patch by Lennart Regebro)
........
Victor Stinner
Amaury Forgeot d'Arc
Antoine Pitrou
Giampaolo Rodolà
Nick Coghlan
Ezio Melotti
Benjamin Peterson
Florent Xicluna
Alexander Belopolsky
Richard Jones
Raymond Hettinger
Brian Curtin
Eric Smith
Tim Golden
Mark Dickinson
Fred Drake
Senthil Kumaran
R. David Murray
Georg Brandl
Ronald Oussoren
Gregory P. Smith
Andrew M. Kuchling
Michael Foord
Brett Cannon
Martin v. Löwis
Stefan Krah
Jean-Paul Calderone