Commit graph

43198 commits

Author SHA1 Message Date
Antoine Pitrou
679be99a85 Fix (presumably) test_hash under big-endian systems (PPC). 2012-02-22 03:33:56 +01:00
Antoine Pitrou
658d1963e1 Additional debug info in case of failure 2012-02-22 02:46:58 +01:00
Antoine Pitrou
037615e1ef unload() should be sufficient 2012-02-22 02:30:09 +01:00
Antoine Pitrou
abe72d7eb3 Try to debug sporadic test failures 2012-02-22 01:11:31 +01:00
Antoine Pitrou
93291534b7 Avoid py3k warnings related to sort() of unrelated types. 2012-02-22 00:28:46 +01:00
Victor Stinner
65532111b4 site: don't import traceback at startup to speed up Python startup 2012-02-21 22:10:16 +01:00
Benjamin Peterson
c9f54cf512 enable hash randomization by default 2012-02-21 16:08:05 -05:00
Antoine Pitrou
0607f731f2 Fix flaky os.urandom test. 2012-02-21 22:02:04 +01:00
Antoine Pitrou
bc499d26a8 Actually run the new tests. 2012-02-21 19:18:10 +01:00
Antoine Pitrou
07c65884c6 Add a test that memoryviews have hash randomization enabled. 2012-02-21 19:14:26 +01:00
Benjamin Peterson
e39ccef64c carefully don't depend on any dict order 2012-02-21 09:07:40 -05:00
Benjamin Peterson
b6596c516b remove extra zero 2012-02-21 08:08:29 -05:00
Petri Lehtinen
4ab701b2d3 sqlite3: Fix 64-bit integer handling in user functions on 32-bit architectures
Closes #8033.
2012-02-21 14:04:46 +02:00
Petri Lehtinen
8b24506534 Merge branch '3.2'
Closes #8033.
2012-02-21 13:59:40 +02:00
Petri Lehtinen
4fe85abab9 sqlite3: Fix 64-bit integer handling in user functions on 32-bit architectures
Closes #8033.
2012-02-21 13:49:50 +02:00
Ezio Melotti
307da2b070 Merge the HTMLParser fix with 3.2. 2012-02-21 09:29:10 +02:00
Ezio Melotti
29877e8e04 HTMLParser is now able to handle slashes in the start tag. 2012-02-21 09:25:00 +02:00
Ezio Melotti
36b7361fe7 HTMLParser is now able to handle slashes in the start tag. 2012-02-21 09:22:16 +02:00
Benjamin Peterson
c7bbd7be94 don't rely on dict order 2012-02-21 00:15:10 -05:00
Benjamin Peterson
c59df7d4e9 a frozenset is better here 2012-02-20 23:13:20 -05:00
Benjamin Peterson
740593416a use set 2012-02-20 23:11:19 -05:00
Benjamin Peterson
f093724845 don't rely on dict order 2012-02-20 23:06:22 -05:00
Benjamin Peterson
72949bd511 this was supposed to die 2012-02-20 22:33:33 -05:00
Benjamin Peterson
71f660e00f update to Unicode 6.1 2012-02-20 22:24:29 -05:00
Benjamin Peterson
11fa11bc95 fix test_gdb under hash randomization 2012-02-20 21:55:32 -05:00
Benjamin Peterson
5bc92e0824 don't rely on dict order here 2012-02-20 21:47:54 -05:00
Benjamin Peterson
aee9dfba4a merge 2.6 with hash randomization fix 2012-02-20 21:44:56 -05:00
Barry Warsaw
b19fb2462e Whitespace normalization 2012-02-20 20:44:15 -05:00
Barry Warsaw
1e13eb084f - Issue #13703: oCERT-2011-003: add -R command-line option and PYTHONHASHSEED
environment variable, to provide an opt-in way to protect against denial of
  service attacks due to hash collisions within the dict and set types.  Patch
  by David Malcolm, based on work by Victor Stinner.
2012-02-20 20:42:21 -05:00
Georg Brandl
802505d05e Merge more test vs hashing-order fixes from 3.2. 2012-02-21 00:34:05 +01:00
Georg Brandl
2fb477c0f0 Merge 3.2: Issue #13703 plus some related test suite fixes. 2012-02-21 00:33:36 +01:00
Georg Brandl
178e5ea305 Disable an inspect test: it depends on dict ordering which parameter is reported as duplicate. 2012-02-21 00:32:36 +01:00
Antoine Pitrou
95a97b7c0c Delete the iterator, which could accidentally keep a temporary reference to the yielded element. 2012-02-21 00:00:06 +01:00
Antoine Pitrou
e9f637b062 Make "regrtest -j" "-R"-aware 2012-02-20 23:49:07 +01:00
Georg Brandl
e5a0e0a75f Fix obscure failures of datetime-related tests due to the datetime tests failing to restore the system state completely after testing the pure-Python versions. 2012-02-20 23:37:36 +01:00
Georg Brandl
16684eb624 Fix dbm_gnu test relying on set order. 2012-02-20 22:48:06 +01:00
Georg Brandl
f47b20f0b0 Fix use of deprecated assertRegexpMatches method. 2012-02-20 22:08:27 +01:00
Antoine Pitrou
b5c793a0b3 Issue #14063: fix test_importlib failure under OS X case-insensitive filesystems
(regression)
2012-02-20 22:06:59 +01:00
Georg Brandl
9571155ae4 Remove setting hash seed to regrtest's random seed and re-execv()ing: this doesn't preserve Python flags and fails from a temp directory. 2012-02-20 22:06:02 +01:00
Georg Brandl
7c573f7a07 Fix dbm_gnu test relying on set order. 2012-02-20 22:48:06 +01:00
Georg Brandl
61470246d0 Remove setting hash seed to regrtest's random seed and re-execv()ing: this doesn't preserve Python flags and fails from a temp directory. 2012-02-20 22:06:02 +01:00
Georg Brandl
27fe226eb1 Another test_dis dict order dependency. 2012-02-20 22:03:28 +01:00
Georg Brandl
b9dafe618f Merge with 3.1. 2012-02-20 21:43:25 +01:00
Georg Brandl
a108227c47 Fix test_dis dependency on dict order. 2012-02-20 21:41:03 +01:00
Georg Brandl
c425a94899 Fix use of deprecated assert_ method. 2012-02-20 21:37:22 +01:00
Georg Brandl
242631da86 Fix "sys.path modified" warning in test_strlit, by not replacing sys.path itself, only its contents. 2012-02-20 21:36:28 +01:00
Georg Brandl
a86b262d1e Fix bad inheritance in test_subprocess that led to a number of tests being executed twice. 2012-02-20 21:34:57 +01:00
Georg Brandl
09a7c72cad Merge from 3.1: Issue #13703: add a way to randomize the hash values of basic types (str, bytes, datetime)
in order to make algorithmic complexity attacks on (e.g.) web apps much more complicated.

The environment variable PYTHONHASHSEED and the new command line flag -R control this
behavior.
2012-02-20 21:31:46 +01:00
Benjamin Peterson
34b345b888 don't rely on the order of module clearing 2012-02-20 15:20:37 -05:00
Benjamin Peterson
a3ed11bd34 don't rely on the order of module clearing 2012-02-20 15:20:37 -05:00