| 108f137 | Tim Peters | 27 August 2004, 05:36:07 UTC | test_bug1001011(): Verify that s.join([t]) is t for (s, t) in (str, str), (unicode, unicode), and (str, unicode). For (unicode, str), verify that it's *not* t (the result is promoted to unicode instead). Also verify that when t is a subclass of str or unicode that "the right thing" happens. | 27 August 2004, 05:36:07 UTC |
| 894c512 | Tim Peters | 27 August 2004, 05:08:36 UTC | PyUnicode_Join(): Missed a spot where I intended a cast from size_t to int. I sure wish MS would gripe about that! Whatever, note that the statement above it guarantees that the cast loses no info. | 27 August 2004, 05:08:36 UTC |
| 269ab62 | Brett Cannon | 27 August 2004, 05:00:22 UTC | Fix the spelling of Fredrik Lundh's last name (unless there really is a "Fredrik Lund" who contributed the code in question). | 27 August 2004, 05:00:22 UTC |
| 12a4baf | Edward Loper | 27 August 2004, 04:29:23 UTC | Remove unused method _OutputRedirectingPdb.resume | 27 August 2004, 04:29:23 UTC |
| 2de91ba | Edward Loper | 27 August 2004, 02:07:46 UTC | - Removed redundant call to expandtabs in DocTestParesr. - Improvements to interactive debugging support: - Changed the replacement pdb.set_trace to redirect stdout to the real stdout *only* during interactive debugging; stdout from code continues to go to the fake stdout. - When the interactive debugger gets to the end of an example, automatically continue. - Use a replacement linecache.getlines that will return source lines from doctest examples; this makes the source available to the debugger for interactive debugging. - In test_doctest, use a specialized _FakeOutput class instead of a temporary file to fake stdin for the interactive interpreter. | 27 August 2004, 02:07:46 UTC |
| 8ce9f16 | Tim Peters | 27 August 2004, 01:49:32 UTC | PyUnicode_Join(): Two primary aims: 1. u1.join([u2]) is u2 2. Be more careful about C-level int overflow. Since PySequence_Fast() isn't needed to achieve #1, it's not used -- but the code could sure be simpler if it were. | 27 August 2004, 01:49:32 UTC |
| 00f8da7 | Edward Loper | 26 August 2004, 18:05:07 UTC | - Added DocTestParser.parse(), which parses a docstring into Examples and intervening text strings. - Removed DocTestParser.get_program(): use script_from_examples() instead. - Fixed bug in DocTestParser._INDENT_RE - Fixed bug in DocTestParser._min_indent - Moved _want_comment() to the utility function section | 26 August 2004, 18:05:07 UTC |
| 57d88e5 | Walter Dörwald | 26 August 2004, 16:53:04 UTC | Move test_bug1001011() to string_tests.MixinStrUnicodeTest so that it can be used for str and unicode. Drop the test for "".join([s]) is s because this is an implementation detail (and doesn't work for unicode) | 26 August 2004, 16:53:04 UTC |
| 592ef61 | Johannes Gijsbers | 26 August 2004, 10:40:12 UTC | Correct incorrect "it's" to "its". | 26 August 2004, 10:40:12 UTC |
| e7edcb8 | Tim Peters | 26 August 2004, 05:44:27 UTC | output_difference(): In fancy-diff cases, the way this split expected & actual output into lines created spurious empty lines at the ends of each. Those matched, but the fancy diffs had surprising line counts (1 larger than expected), and tests kept having to slam <BLANKLINE> into the expected output to account for this. Using the splitlines() string method with keepends=True instead accomplishes what was intended directly. | 26 August 2004, 05:44:27 UTC |
| 4085f03 | Martin v. Löwis | 26 August 2004, 05:44:02 UTC | Add missing executable option to DummyCommand. | 26 August 2004, 05:44:02 UTC |
| 66cb018 | Tim Peters | 26 August 2004, 05:23:19 UTC | Whitespace normalization. | 26 August 2004, 05:23:19 UTC |
| 5b799c1 | Tim Peters | 26 August 2004, 05:21:59 UTC | _do_a_fancy_diff(): Pay no attention to the ellipses behind the curtain. While a fancy diff can be confusing in the presence of ellipses, so far I'm finding (2-0-0) that it's much more a major aid in narrowing down the possibilities when an ellipsis-slinging test fails. So we no longer refuse to do a fancy diff just because of ellipses. This isn't ideal; it's just better. | 26 August 2004, 05:21:59 UTC |
| f33683f | Tim Peters | 26 August 2004, 04:52:46 UTC | Fine tune a word. | 26 August 2004, 04:52:46 UTC |
| cf53355 | Tim Peters | 26 August 2004, 04:50:38 UTC | Thinko repair. | 26 August 2004, 04:50:38 UTC |
| a07bcd4 | Tim Peters | 26 August 2004, 04:47:31 UTC | Reorg of exception section. Now that there are fewer details needing explanation, it's easier to push the remaining insufferably anal details into a "fine print" section at the bottom. | 26 August 2004, 04:47:31 UTC |
| 770acc2 | Raymond Hettinger | 26 August 2004, 04:29:47 UTC | Remove unnecessary line. | 26 August 2004, 04:29:47 UTC |
| f4fd79c | Raymond Hettinger | 26 August 2004, 03:11:56 UTC | Small wording fixups. | 26 August 2004, 03:11:56 UTC |
| ca9111e | Edward Loper | 26 August 2004, 03:00:24 UTC | Changed OutputChecker.output_difference to expect an Example object, rather than an expected output string. This gives the output_difference method access to more information, such as the indentation of the example, which might be useful. | 26 August 2004, 03:00:24 UTC |
| a89f88d | Edward Loper | 26 August 2004, 02:45:51 UTC | Added REPORT_ONLY_FIRST_FAILURE flag, which supresses output after the first failing example in each test. | 26 August 2004, 02:45:51 UTC |
| cc8a4f6 | Brett Cannon | 26 August 2004, 01:44:07 UTC | When building with --disable-toolbox-glue under Darwin, skip building any Mac-specific modules. Before all modules were compiled but would fail thanks to a dependence on the code included when Python was built without the compiler flag. Closes bug #991962. | 26 August 2004, 01:44:07 UTC |
| 71f55af | Edward Loper | 26 August 2004, 01:41:51 UTC | Renamed UNIFIED_DIFF->REPORT_UDIFF; CONTEXT_DIFF->REPORT_CDIFF; and NDIFF_DIFF->REPORT_NDIFF. This establishes the naming convention that all reporting options should begin with "REPORT_" (since reporting options are a different class from output comparison options; but they are both set in optionflags). | 26 August 2004, 01:41:51 UTC |
| 5662929 | Edward Loper | 26 August 2004, 01:31:56 UTC | Shortened diff output for unified & context diffs | 26 August 2004, 01:31:56 UTC |
| aacf083 | Edward Loper | 26 August 2004, 01:19:50 UTC | - Changed the output of report_start() and report_unexpected_exception() to be more consistent with report_failure() - If `want` or `got` is empty, then print "Expected nothing\n" or "Got nothing\n" rather than "Expected:\n" or "Got:\n" - Got rid of _tag_msg | 26 August 2004, 01:19:50 UTC |
| 0e44807 | Tim Peters | 26 August 2004, 01:02:08 UTC | Restored half of a \versionadded only half of which should have been deleted. | 26 August 2004, 01:02:08 UTC |
| e8bbfe4 | Dave Cole | 26 August 2004, 00:51:16 UTC | Patch #1015012. Improve markup and punctuation in libsocket.tex | 26 August 2004, 00:51:16 UTC |
| 3203efb | Dave Cole | 26 August 2004, 00:37:31 UTC | Patch #1014930. Expose current parse location to XMLParser. | 26 August 2004, 00:37:31 UTC |
| 0d58e2b | Raymond Hettinger | 26 August 2004, 00:21:13 UTC | Minor improvements to the template code. * Add comment bars segregating this code from the rest. * Improve readability of the re pattern with indentation and comments on the same line. * Replace the groupdict() and get() pair with a direct call to group() which does the same thing. | 26 August 2004, 00:21:13 UTC |
| a6b6832 | Edward Loper | 26 August 2004, 00:05:43 UTC | Added an "exc_msg" attribute to Example (containing the expected exception message, or None if no exception is expected); and moved exception parsing from DocTestRunner to DocTestParser. This is architecturally cleaner, since it moves all parsing work to DocTestParser; and it should make it easier for code outside DocTestRunner (notably debugging code) to properly handle expected exceptions. | 26 August 2004, 00:05:43 UTC |
| c5625ba | Trent Mick | 25 August 2004, 23:59:39 UTC | Re-generate configure with recent configure.in change (for patch #1016224). | 25 August 2004, 23:59:39 UTC |
| af16e8c | Trent Mick | 25 August 2004, 23:55:59 UTC | Apply patch #1016224: configure.in change to allow build of _bsddb module on AIX 5.1. | 25 August 2004, 23:55:59 UTC |
| 19b1958 | Edward Loper | 25 August 2004, 23:07:03 UTC | Only recognize the expected output as an exception if it *starts* with a traceback message. I.e., examples that raise exceptions may no longer generate pre-exception output. This restores the behavior of doctest in python 2.3. The ability to check pre-exception output is being removed because it makes the documentation simpler; and because there are very few use cases for it. | 25 August 2004, 23:07:03 UTC |
| d2afee4 | Raymond Hettinger | 25 August 2004, 19:42:12 UTC | Fix docstring typo. | 25 August 2004, 19:42:12 UTC |
| 1792bfb | Raymond Hettinger | 25 August 2004, 17:19:38 UTC | Bypass peepholing of code with lineno tables having intervals >= 255. Allows the lineno fixup code to remain simple and not have to deal with multibyte codings. * Add an assertion to that effect. * Remove the XXX comment on the subject. | 25 August 2004, 17:19:38 UTC |
| 65d3c05 | Raymond Hettinger | 25 August 2004, 15:15:56 UTC | Fix typo in comment and add clarification. | 25 August 2004, 15:15:56 UTC |
| e3a1b48 | Fred Drake | 25 August 2004, 14:01:32 UTC | clean up markup for --executable description | 25 August 2004, 14:01:32 UTC |
| 87c98b2 | Andrew M. Kuchling | 25 August 2004, 13:38:46 UTC | Add various items | 25 August 2004, 13:38:46 UTC |
| cae9e67 | Martin v. Löwis | 25 August 2004, 13:04:53 UTC | Patch #970015: Replace - by _ in version and release. | 25 August 2004, 13:04:53 UTC |
| cf52524 | Martin v. Löwis | 25 August 2004, 13:00:34 UTC | Patch #970019: Include version and release in the BuildRoot. | 25 August 2004, 13:00:34 UTC |
| 9f5c0c4 | Martin v. Löwis | 25 August 2004, 11:37:43 UTC | Patch #736857, #736859: Add -e option to build_scripts. | 25 August 2004, 11:37:43 UTC |
| be83737 | Martin v. Löwis | 25 August 2004, 11:24:42 UTC | Patch #798244: More urllib2 examples. | 25 August 2004, 11:24:42 UTC |
| c11d6f1 | Martin v. Löwis | 25 August 2004, 10:52:58 UTC | Patch #1014992: Never return more than a line from readline. Will backport to 2.3. | 25 August 2004, 10:52:58 UTC |
| 8d494f3 | Martin v. Löwis | 25 August 2004, 10:42:41 UTC | Patch #1015021: Stop claiming that coerce can return None. Will backport to 2.3. | 25 August 2004, 10:42:41 UTC |
| b92b7ed | Martin v. Löwis | 25 August 2004, 06:24:58 UTC | Back out 1.289, which breaks platforms needing addrinfo.h, and 1.293, 1.298, and 1.300, which have tried to fix this for specific platforms. | 25 August 2004, 06:24:58 UTC |
| ef0a82b | Raymond Hettinger | 25 August 2004, 03:18:29 UTC | Simplify chains of conditional jumps. (Suggested by Neal Norwitz.) | 25 August 2004, 03:18:29 UTC |
| 08b07de | Barry Warsaw | 25 August 2004, 03:09:58 UTC | Ah whoops, we didn't turn string into a package (thanks Neal!) | 25 August 2004, 03:09:58 UTC |
| 8bee761 | Barry Warsaw | 25 August 2004, 02:22:30 UTC | PEP 292 classes Template and SafeTemplate are added to the string module. This patch includes test cases and documentation updates, as well as NEWS file updates. This patch also updates the sre modules so that they don't import the string module, breaking direct circular imports. | 25 August 2004, 02:22:30 UTC |
| c885443 | Tim Peters | 25 August 2004, 02:14:08 UTC | Stop producing or using OverflowWarning. PEP 237 thought this would happen in 2.3, but nobody noticed it still was getting generated (the warning was disabled by default). OverflowWarning and PyExc_OverflowWarning should be removed for 2.5, and left notes all over saying so. | 25 August 2004, 02:14:08 UTC |
| 1fa040b | Neal Norwitz | 25 August 2004, 01:20:18 UTC | SF #1015517, get readline to compile with older compilers | 25 August 2004, 01:20:18 UTC |
| eb619bb | Mark Hammond | 24 August 2004, 22:24:08 UTC | Fix for [ 1010677 ] thread Module Breaks PyGILState_Ensure(), and a test case. When booting a new thread, use the PyGILState API to manage the GIL. | 24 August 2004, 22:24:08 UTC |
| 8107ca4 | Skip Montanaro | 24 August 2004, 14:26:43 UTC | Keep option parser from gobbling up the filename to be profiled and the flags it accepts. It's too late to change optparse's default behavior now, but I find the default setting of allow_interspersed_args very weird. | 24 August 2004, 14:26:43 UTC |
| fb15417 | Vinay Sajip | 24 August 2004, 09:36:23 UTC | Fixed bug in DatagramHandler.send() | 24 August 2004, 09:36:23 UTC |
| a12fa14 | Raymond Hettinger | 24 August 2004, 04:34:16 UTC | Incorporate review comments courtesy of Neal Norwitz: * Perform the code length check earlier. * Eliminate the extra PyMem_Free() upon hitting an EXTENDED_ARG. * Assert that the NOP count used in jump retargeting matches the NOPs eliminated in the final step. * Add an XXX note to indicate that more work is being to done to handle linenotab with intervals > 255. | 24 August 2004, 04:34:16 UTC |
| fd2d1f7 | Raymond Hettinger | 23 August 2004, 23:37:48 UTC | SF Patch #1013667: Cleanup Peepholer Output * Make a pass to eliminate NOPs. Produce code that is more readable, more compact, and a tiny bit faster. Makes the peepholer more flexible in the scope of allowable transformations. * With Guido's okay, bumped up the magic number so that this patch gets widely exercised before the alpha goes out. | 23 August 2004, 23:37:48 UTC |
| 08158a0 | Raymond Hettinger | 23 August 2004, 23:30:20 UTC | Add Nick Coghlan for his patch solving an issue with joining string subclasses. | 23 August 2004, 23:30:20 UTC |
| 674f241 | Raymond Hettinger | 23 August 2004, 23:23:54 UTC | SF Patch #1007087: Return new string for single subclass joins (Bug #1001011) (Patch contributed by Nick Coghlan.) Now joining string subtypes will always return a string. Formerly, if there were only one item, it was returned unchanged. | 23 August 2004, 23:23:54 UTC |
| dd50cb7 | Tim Peters | 23 August 2004, 22:42:55 UTC | The attempt to shut up deprecation warnings for doctest's own use of is_private in its tests failed if doctest.py was run directly. Now it works. | 23 August 2004, 22:42:55 UTC |
| 4de7c5c | Tim Peters | 23 August 2004, 22:38:05 UTC | test_DocTestFinder(): This test failed when test_doctest was run directly, due to assuming a filename specific to running tests "the normal way". +ELLIPSIS to the rescue! | 23 August 2004, 22:38:05 UTC |
| a7def72 | Tim Peters | 23 August 2004, 22:13:22 UTC | Moved some test cases from doctest to test_doctest. | 23 August 2004, 22:13:22 UTC |
| 3fa8c20 | Tim Peters | 23 August 2004, 21:43:39 UTC | Misc cleanups. | 23 August 2004, 21:43:39 UTC |
| b6a04d6 | Tim Peters | 23 August 2004, 21:37:56 UTC | debug_script(): I changed this in haste before to take out the use of NamedTemporaryFile (which can't work for this function's purposes on Windows). Leaving temp files behind wasn't a great idea either, though, so try to clean up. At least the test suite no longer leaves any of these guys behind now. | 23 August 2004, 21:37:56 UTC |
| 31bd529 | Martin v. Löwis | 23 August 2004, 20:42:35 UTC | Patch #1014237: Consistently return booleans throughout. | 23 August 2004, 20:42:35 UTC |
| 07fda7e | Dave Cole | 23 August 2004, 05:16:23 UTC | Updated the socketpair() docstring and documentation to explain that the default famility is AF_UNIX if defined for the platform, otherwise the default is AF_INET. | 23 August 2004, 05:16:23 UTC |
| 0fc8575 | Dave Cole | 23 August 2004, 04:54:53 UTC | Removed unnecessary calls to signal() to ignore SIGPIPE. SIGPIPE is ignored in initsigs() inside pythonrun.c. | 23 August 2004, 04:54:53 UTC |
| 656f7e4 | Tim Peters | 23 August 2004, 00:26:42 UTC | Removed redundant versionadded{} for NDIFF_DIFF. Virtually everything in this section is new in 2.4, and that's all mentioned already in versionadded{} thingies at the end of the section. It hurts readability to have them after every line <wink>. | 23 August 2004, 00:26:42 UTC |
| 8bd5cb3 | Neal Norwitz | 22 August 2004, 21:48:37 UTC | Add version info | 22 August 2004, 21:48:37 UTC |
| c234a52 | Martin v. Löwis | 22 August 2004, 21:28:33 UTC | Flush bz2 data even if nothing had been written so far. Fixes #1013882. Will backport to 2.3. | 22 August 2004, 21:28:33 UTC |
| 70aa1f2 | Martin v. Löwis | 22 August 2004, 21:09:15 UTC | Fix repr for negative imaginary part. Fixes #1013908. | 22 August 2004, 21:09:15 UTC |
| f82a9de | Tim Peters | 22 August 2004, 20:51:53 UTC | Start deferring to the LaTeX docs for details. I'd like to move the docstrings toward being a lot shorter, and telling the whole truth in the manual instead. This change is an example: the manual has detailed explanations of the option names now, so it's Bad to repeat them in the docstring (two detailed descriptions are certain to get out of synch). Just listing the names has memory-jogging benefits, though, so that's still helpful in the docstring. | 22 August 2004, 20:51:53 UTC |
| c6cbab0 | Tim Peters | 22 August 2004, 19:43:28 UTC | Added NDIFF_DIFF option. | 22 August 2004, 19:43:28 UTC |
| 94607dd | Tim Peters | 22 August 2004, 19:42:56 UTC | Whitespace normalization. | 22 August 2004, 19:42:56 UTC |
| 75dc5e1 | Tim Peters | 22 August 2004, 17:50:45 UTC | Type in docstring. | 22 August 2004, 17:50:45 UTC |
| c504915 | Tim Peters | 22 August 2004, 17:34:58 UTC | _parse_example(): Simplified new code to preserve trailing spaces before final newline. Anything to get rid of "l" as a variable name <0.5 wink>. | 22 August 2004, 17:34:58 UTC |
| 2dd2a28 | Martin v. Löwis | 22 August 2004, 17:10:12 UTC | Add acknowledgements to ExitDialog. | 22 August 2004, 17:10:12 UTC |
| e09bd93 | Martin v. Löwis | 22 August 2004, 16:13:26 UTC | Patch #997284: Allow pydoc to work with XP Themes (.manifest file) Will backport to 2.3. | 22 August 2004, 16:13:26 UTC |
| ac37f3d | Martin v. Löwis | 22 August 2004, 16:08:04 UTC | Patch #1009389: Make __credits__ a Unicode object. | 22 August 2004, 16:08:04 UTC |
| 12237b3 | Martin v. Löwis | 22 August 2004, 16:04:50 UTC | Replace yield with sequence class. Fixes #1009803. | 22 August 2004, 16:04:50 UTC |
| d3b0bab | Raymond Hettinger | 22 August 2004, 15:24:33 UTC | SF bug #1013760: Error in Chapter 4 of Tutorial Fit a nit. | 22 August 2004, 15:24:33 UTC |
| 07a349c | Jim Fulton | 22 August 2004, 14:10:00 UTC | Bugs fixed: - Test filenames sometimes had trailing .pyc or .pyo sufixes (when module __file__ did). - Trailing spaces spaces in expected output were dropped. New default failure format: - Separation of examples from file info makes examples easier to see - More vertical separation, improving readability - Emacs-recognized file info (also closer to Python exception format) | 22 August 2004, 14:10:00 UTC |
| 8ffe9ab | Martin v. Löwis | 22 August 2004, 13:34:34 UTC | Move msi from sandbox to Tools. | 22 August 2004, 13:34:34 UTC |
| 183403a | Kurt B. Kaiser | 22 August 2004, 05:14:32 UTC | 1. If user passes a non-existant filename on the commandline, just open a new file, don't raise a dialog. IDLEfork 954928. 2. Refactor EditorWindow.wakeup() to WindowList.ListedToplevel.wakeup() and clarify that the Toplevel of an EditorWindow is a WindowList.ListedToplevel. 3. Make a number of improvements to keyboard focus binding. Improve window raising, especially in the debugger. IDLEfork Bug 763524 (GvR list). 4. Bump idlever to 1.1a3 M Debugger.py M EditorWindow.py M FileList.py M NEWS.txt M PyShell.py M WindowList.py M idlever.py | 22 August 2004, 05:14:32 UTC |
| e594bee | Tim Peters | 22 August 2004, 01:47:51 UTC | _ellipsis_match(): Removed special-casing of "...\n". The semantics are non-obvious either way because the newline character "is invisible", but it's still there all the same, and it's easier to explain/predict if that reality is left alone. | 22 August 2004, 01:47:51 UTC |
| 5268c39 | Neal Norwitz | 21 August 2004, 15:13:52 UTC | sf #1009373, #1005936. fix underscores in index entries in PDF files. I couldn't test this, but it didn't break anything and the patch reported fixed the problem. Bugfix candidate. | 21 August 2004, 15:13:52 UTC |
| 6abc685 | Johannes Gijsbers | 21 August 2004, 12:30:26 UTC | Patch #880621: the last message of a Babyl mailbox ends in '\037' instead of '\037\014\n' (see http://quimby.gnus.org/notes/BABYL) so look for that as well, so that applications won't get '\037' as the last line of the last message. | 21 August 2004, 12:30:26 UTC |
| 6d63a8d | Johannes Gijsbers | 21 August 2004, 10:43:29 UTC | Patch #1011123: Use urllib.quote() instead of cgi.escape() for encoding the href attribute in list_directory(). This fixes the links for legal Unix filenames such as 'a"b'. | 21 August 2004, 10:43:29 UTC |
| 037b3ee | Tim Peters | 21 August 2004, 06:55:43 UTC | Patch 1012740: cStringIO's truncate doesn't truncate() left the stream position unchanged, which meant the "truncated" data didn't go away: >>> io.write('abc') >>> io.truncate(0) >>> io.write('xyz') >>> io.getvalue() 'abcxyz' Patch by Dima Dorfman. | 21 August 2004, 06:55:43 UTC |
| 7109b28 | Neal Norwitz | 20 August 2004, 23:13:26 UTC | Fix grammar, spotted by Hye-Shik Chang | 20 August 2004, 23:13:26 UTC |
| 8e3bec5 | Johannes Gijsbers | 20 August 2004, 14:38:56 UTC | Correct argument specifications of load_compiled and load_source: the file argument is optional on both (the format string is "ss|O!"). | 20 August 2004, 14:38:56 UTC |
| c890618 | Johannes Gijsbers | 20 August 2004, 14:37:05 UTC | Quote # as \# to make lib compile again. | 20 August 2004, 14:37:05 UTC |
| 37af986 | Martin v. Löwis | 20 August 2004, 07:31:37 UTC | Bump minimum Python version to 2.1. Fixes #1009803. | 20 August 2004, 07:31:37 UTC |
| 61d77e0 | Martin v. Löwis | 20 August 2004, 06:35:46 UTC | Replace tricky and/or with straight-forward if:else: | 20 August 2004, 06:35:46 UTC |
| 701abe7 | Martin v. Löwis | 20 August 2004, 06:26:59 UTC | Fail fatally if strdup fails. | 20 August 2004, 06:26:59 UTC |
| cd4d297 | Martin v. Löwis | 20 August 2004, 06:19:23 UTC | Remove debug print. | 20 August 2004, 06:19:23 UTC |
| a45cacf | Tim Peters | 20 August 2004, 03:47:14 UTC | Whitespace normalization. | 20 August 2004, 03:47:14 UTC |
| 8d7626c | Tim Peters | 20 August 2004, 03:27:12 UTC | Stab at SF 1010777: test_queue fails occasionally test_queue has failed occasionally for years, and there's more than one cause. The primary cause in the SF report appears to be that the test driver really needs entirely different code for thread tests that expect to raise exceptions than for thread tests that are testing non-exceptional blocking semantics. So gave them entirely different code, and added a ton of explanation. Another cause is that the blocking thread tests relied in several places on the difference between sleep(.1) and sleep(.2) being long enough for the trigger thread to do its stuff sot that the blocking thread could make progress. That's just not reliable on a loaded machine. Boosted the 0.2's to 10.0's instead, which should be long enough under any non-catastrophic system conditions. That doesn't make the test take longer to run, the 10.0 is just how long the blocking thread is *willing* to wait for the trigger thread to do something. But if the Queue module is plain broken, such tests will indeed take 10 seconds to fail now. For similar (heavy load) reasons, changed threaded-test termination to be willing to wait 10 seconds for the signal thread to end too. | 20 August 2004, 03:27:12 UTC |
| afe5297 | Tim Peters | 20 August 2004, 02:37:25 UTC | Semantic-neutral format and comment changes. | 20 August 2004, 02:37:25 UTC |
| 8623b36 | Neal Norwitz | 20 August 2004, 02:36:27 UTC | Try to improve grammar and use versionadded | 20 August 2004, 02:36:27 UTC |
| b0a04e1 | Tim Peters | 20 August 2004, 02:08:04 UTC | Gave _ellipsis_match() an attractive new leading underscore. | 20 August 2004, 02:08:04 UTC |
| 525b315 | Neal Norwitz | 20 August 2004, 01:52:42 UTC | Fix markup. versionchanged automatically adds a period, so remove the extra one. | 20 August 2004, 01:52:42 UTC |
| 27dbcf2 | Raymond Hettinger | 19 August 2004, 22:39:55 UTC | Establish policies with respect to 2.3 compatibilty and treated spec updates as bugfixes. | 19 August 2004, 22:39:55 UTC |
