| 841815e | cvs2svn | 09 October 2000, 23:43:55 UTC | This commit was manufactured by cvs2svn to create tag 'r20c1'. | 09 October 2000, 23:43:55 UTC |
| cfc4178 | Tim Peters | 09 October 2000, 23:43:55 UTC | When the classes in wave.py opened files themselves, their .close() methods didn't bother to close the files. This caused the new test_wave test to fail under Windows, as Windows won't let you delete a file that's open. Fixed that by ensuring the wave read & write classes' .close() and __del__ methods close files that were opened by their constructors. | 09 October 2000, 23:43:55 UTC |
| e8d2f55 | Guido van Rossum | 09 October 2000, 22:14:43 UTC | The usual. | 09 October 2000, 22:14:43 UTC |
| eebb1c6 | Jeremy Hylton | 09 October 2000, 21:48:02 UTC | The Usual. | 09 October 2000, 21:48:02 UTC |
| cc46bd2 | Jeremy Hylton | 09 October 2000, 21:34:51 UTC | the usual | 09 October 2000, 21:34:51 UTC |
| 6040aaa | Jeremy Hylton | 09 October 2000, 21:27:22 UTC | add note explaining what a release candidate is | 09 October 2000, 21:27:22 UTC |
| 95fdec4 | Fred Drake | 09 October 2000, 21:26:12 UTC | This file is no longer used. | 09 October 2000, 21:26:12 UTC |
| 46446d6 | Tim Peters | 09 October 2000, 21:19:31 UTC | Repaired IDLE Unicode bug description. Added tokenize.py bugfix info. | 09 October 2000, 21:19:31 UTC |
| ff71839 | Fred Drake | 09 October 2000, 20:22:28 UTC | Another generated file... | 09 October 2000, 20:22:28 UTC |
| c9a5cc1 | Fred Drake | 09 October 2000, 20:18:32 UTC | Remove macro that mentions DL_EXPORT_HEADER; it caused a spurious warning from autoheader and is not useful anymore. (Approved by Donn Cave.) | 09 October 2000, 20:18:32 UTC |
| 56fe7f8 | Guido van Rossum | 09 October 2000, 20:07:00 UTC | WAVE test + bugfix contributor. | 09 October 2000, 20:07:00 UTC |
| 16b198e | Guido van Rossum | 09 October 2000, 20:06:50 UTC | wave test output | 09 October 2000, 20:06:50 UTC |
| a00afc1 | Guido van Rossum | 09 October 2000, 20:05:59 UTC | Simple test suite for wave.py by Jean-Claude Rimbault (with some changes to avoid using assert). | 09 October 2000, 20:05:59 UTC |
| 13a3069 | Fred Drake | 09 October 2000, 20:04:16 UTC | Paul Prescod <paul@prescod.net>: Correct the chaining between siblings. | 09 October 2000, 20:04:16 UTC |
| eca576c | Guido van Rossum | 09 October 2000, 20:01:53 UTC | Fix by Jean-Claude Rimbault [ Bug #116271 ] -- the WAVE header was never written properly because the '4' length indicators for the 's' format characters were missing. | 09 October 2000, 20:01:53 UTC |
| ebe7302 | Fred Drake | 09 October 2000, 19:57:39 UTC | Move the test for confirmation that all nodes have been freed into the driver code, so that each test gets this; it had been done inconsistently. Remove the lines that set the variables holding dom objects to None; not needed since the interpreter cleans up locals on function return. | 09 October 2000, 19:57:39 UTC |
| 5c2c604 | Guido van Rossum | 09 October 2000, 19:52:41 UTC | And another. | 09 October 2000, 19:52:41 UTC |
| 5839e58 | Guido van Rossum | 09 October 2000, 19:52:35 UTC | Checking in three Darwin-specific patches. Tony Lownds: [ Patch #101816 ] Fixes shared modules on Mac OS X 1. Mac OS X is recognized by the Next-ish host recognition code as "Darwin/1.2" 2. When specifying just --with-dyld, modules can compile as shared 3. --with-dyld and --with-next-framework, modules can compile as shared 4. --with-suffix=.exe, and Lib/plat-darwin1.2 is being made, the regen script invokes python as python.exe [I had to reformat this patch a bit to make it work. Please test!] Dan Wolfe: [ Patch #101823 ] Fix Darwin POSIX Thread redefinition The patch below fixes the redefinition problem in Darwin with _POSIX_THREADS. I'm not sure if this is the correct long term fix but for now it fixes the problem and the fix is specific to Darwin. Dan Wolfe: [ Patch #101824 ] On Darwin, remove unrecognized option `-OPT:Olimit=0' After many, many, many compiles, I finally got itchy of this warning cluttering up the output... so I scratched (Darwin configs only) and it's gone! :-) | 09 October 2000, 19:52:35 UTC |
| 32e20ff | Jeremy Hylton | 09 October 2000, 19:48:11 UTC | typo | 09 October 2000, 19:48:11 UTC |
| 95d53d2 | Guido van Rossum | 09 October 2000, 19:34:12 UTC | Use python$EXE instead of python, for Darwin. (Patch by Tony Lownds. (#101816) [Note: I'm not sure that this is really the right fix. Surely Darwin doesn't require you to say "python.exe" everywhere??? Even Windows doesn't! Or am I misunderstanding the point?] | 09 October 2000, 19:34:12 UTC |
| fdddb31 | Guido van Rossum | 09 October 2000, 19:31:40 UTC | For Darwin, export EXE (needed by Lib/plat-generic/regen checkin, to follow). Adapted from a patch by Tony Lownds. (#101816) | 09 October 2000, 19:31:40 UTC |
| fded5ed | Guido van Rossum | 09 October 2000, 19:29:35 UTC | One more name. | 09 October 2000, 19:29:35 UTC |
| 87d4a03 | Fred Drake | 09 October 2000, 18:56:23 UTC | Do not forget to build the acks.html file when building "all"! | 09 October 2000, 18:56:23 UTC |
| 1092120 | Jeremy Hylton | 09 October 2000, 18:34:12 UTC | added better description of BeOS changes from Donn Cave | 09 October 2000, 18:34:12 UTC |
| ed9e644 | Jeremy Hylton | 09 October 2000, 18:26:42 UTC | Summary of changes between 2.0b2 and 2.0c1 | 09 October 2000, 18:26:42 UTC |
| 30e7877 | Jeremy Hylton | 09 October 2000, 18:26:16 UTC | new name | 09 October 2000, 18:26:16 UTC |
| 17a2b64 | Fred Drake | 09 October 2000, 18:12:29 UTC | Work around annoyances in LaTeX2HTML. | 09 October 2000, 18:12:29 UTC |
| 955e03b | Fred Drake | 09 October 2000, 18:11:24 UTC | Push xmllib to the end of the markup chapter since it is deprecated. | 09 October 2000, 18:11:24 UTC |
| ecbd2aa | Fred Drake | 09 October 2000, 18:08:56 UTC | Another name. | 09 October 2000, 18:08:56 UTC |
| 5790be1 | Fred Drake | 09 October 2000, 17:06:13 UTC | Donn Cave <donn@oz.net>: Update for BeOS. This closes SourceForge patch #101774. Also fix typo in a comment. | 09 October 2000, 17:06:13 UTC |
| a2cb78f | Fred Drake | 09 October 2000, 17:01:03 UTC | Donn Cave <donn@oz.net>: Removed DL_EXPORT_HEADER -- only needed on BeOS, and not needed there anymore. This closes SourceForge patch #101775. | 09 October 2000, 17:01:03 UTC |
| 047fe99 | Fred Drake | 09 October 2000, 16:51:49 UTC | Donn Cave <donn@oz.net>: Revise BeOS support. This closes SourceForge patch #101776. | 09 October 2000, 16:51:49 UTC |
| 9e2e1e5 | Fred Drake | 09 October 2000, 16:48:09 UTC | Donn Cave <donn@oz.net>: Updated to work better with BeOS. This closes SourceForge patch #101777. | 09 October 2000, 16:48:09 UTC |
| d359022 | Fred Drake | 09 October 2000, 16:46:02 UTC | Updated version from Donn Cave <donn@oz.net>. This closes SourceForge patch #101778. | 09 October 2000, 16:46:02 UTC |
| 04f4943 | Martin v. Löwis | 09 October 2000, 16:45:54 UTC | _exceptions: Format a missing system id as <unknown>. expatreader: Use the error handler instead of raising exception directly. | 09 October 2000, 16:45:54 UTC |
| 3383792 | Fred Drake | 09 October 2000, 15:27:31 UTC | Denis S. Otkidach <ods@users.sourceforge.net>: Correct description of leapdays() function. This closes SourceForge patch #101840. | 09 October 2000, 15:27:31 UTC |
| 1e75e81 | Jeremy Hylton | 09 October 2000, 15:13:47 UTC | bump patchlevel to 2.0c1 | 09 October 2000, 15:13:47 UTC |
| b769e80 | Jeremy Hylton | 09 October 2000, 14:35:24 UTC | read in a .pyc file and disassemble the code objects | 09 October 2000, 14:35:24 UTC |
| d923831 | Andrew M. Kuchling | 09 October 2000, 14:18:10 UTC | Patch #101810: check whether zst.avail_out is non-zero when getting a Z_BUF_ERROR while decompressing. If it is, assume that this means the data being decompressed is bad and raise an exception, instead of just assuming that Z_BUF_ERROR always means that more space is required. | 09 October 2000, 14:18:10 UTC |
| 46735ad | Guido van Rossum | 09 October 2000, 12:42:04 UTC | Fixed leapdays(). From Patch #101841, by Denis S. Otkidach. | 09 October 2000, 12:42:04 UTC |
| e9f922f | Guido van Rossum | 08 October 2000, 19:48:46 UTC | Adapt test output to changed error message. | 08 October 2000, 19:48:46 UTC |
| 5931667 | Guido van Rossum | 08 October 2000, 19:47:47 UTC | Be consistent in the description of audio formats: <format> <width> "audio". Also add AFMT_S16_NE ("native-endian"). (Somehow there's no AFMT_U16_NE.) | 08 October 2000, 19:47:47 UTC |
| ce6292e | Guido van Rossum | 08 October 2000, 00:21:43 UTC | Correct output. | 08 October 2000, 00:21:43 UTC |
| 923ebe4 | Guido van Rossum | 08 October 2000, 00:20:20 UTC | It seems our sound cards can't play mulaw data. Use native-format 16-bit signed data instead. Hope this works for you; it works for me. | 08 October 2000, 00:20:20 UTC |
| 830b37b | Martin v. Löwis | 07 October 2000, 19:03:20 UTC | Don't use string methods to allow sharing this code with PyXML. | 07 October 2000, 19:03:20 UTC |
| 13ac992 | Fredrik Lundh | 07 October 2000, 17:38:23 UTC | Fixed too ambitious "nothing to repeat" check. Closes bug #114033. | 07 October 2000, 17:38:23 UTC |
| 5c0b43d | Fred Drake | 07 October 2000, 16:58:11 UTC | The test is good, but Jim forgot to check in the updated output. | 07 October 2000, 16:58:11 UTC |
| 4095101 | Guido van Rossum | 07 October 2000, 16:21:27 UTC | Put arguments to test -z in double quotes. Fixes Bug #116325. | 07 October 2000, 16:21:27 UTC |
| 7787841 | Fred Drake | 07 October 2000, 12:50:05 UTC | Hush the nanny. | 07 October 2000, 12:50:05 UTC |
| 5d64421 | Fred Drake | 07 October 2000, 12:31:50 UTC | Fix a couple of places where the descriptions of *_GET_SIZE() macros said they were similar to *_GetSize(); should be similar to *_Size(). Error noted by William Park <parkw@better.net>. | 07 October 2000, 12:31:50 UTC |
| 613f7c4 | Fred Drake | 07 October 2000, 12:25:17 UTC | Donn Cave <donn@oz.net>: Generated files for BeOS R5. | 07 October 2000, 12:25:17 UTC |
| a2fda0d | Martin v. Löwis | 07 October 2000, 12:10:28 UTC | Record bugs found when comparing the module with DOM Core Level 2. | 07 October 2000, 12:10:28 UTC |
| 8bf9e3b | Mark Hammond | 07 October 2000, 11:10:50 UTC | Prevent possible buffer overflow exploits under Windows. As per (the very quick) patch Patch #101801. | 07 October 2000, 11:10:50 UTC |
| 025468d | Fredrik Lundh | 07 October 2000, 10:16:19 UTC | SRE didn't handle character category followed by hyphen inside a character class. Fix provided by Andrew Kuchling. Closes bug #116251. | 07 October 2000, 10:16:19 UTC |
| 53f3d4a | Marc-André Lemburg | 07 October 2000, 08:54:09 UTC | [ Bug #116174 ] using %% in cstrings sometimes fails with unicode paramsFix for the bug reported in Bug #116174: "%% %s" % u"abc" failed due to the way string formatting delegated work to the Unicode formatting function. | 07 October 2000, 08:54:09 UTC |
| b96d802 | Marc-André Lemburg | 07 October 2000, 08:52:45 UTC | Updated test with a case which checks for the bug reported in | 07 October 2000, 08:52:45 UTC |
| de49583 | Tim Peters | 07 October 2000, 05:09:39 UTC | Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py). tokenize.py has always used naive regexps for matching string literals, and that appears to trigger the sre recursion limit on Skip's platform (he has very long single-line string literals). Replaced all of tokenize.py's string regexps with the "unrolled" forms used in IDLE, where they're known to handle even absurd (multi-megabyte!) string literals without trouble. See Friedl's book for explanation (at heart, the naive regexps create a backtracking choice point for each character in the literal, while the unrolled forms create none). | 07 October 2000, 05:09:39 UTC |
| 70d87d7 | Tim Peters | 07 October 2000, 04:04:07 UTC | Prep Windows installer for 2.0c1: title and build number. | 07 October 2000, 04:04:07 UTC |
| f2fba87 | Tim Peters | 06 October 2000, 23:09:00 UTC | Fix for next iteration of SF bug 115690 (Unicode headaches in IDLE). The parsing functions in support of auto-indent weren't expecting Unicode strings, but text.get() can now return them (although it remains muddy as to exactly when or why that can happen). Fixed that with a Big Hammer. | 06 October 2000, 23:09:00 UTC |
| 0a84a33 | Martin v. Löwis | 06 October 2000, 22:42:55 UTC | Add .toxml test case, as proposed by Alex Martelli in bug report #116244. | 06 October 2000, 22:42:55 UTC |
| 2c8a89c | Martin v. Löwis | 06 October 2000, 22:36:03 UTC | minidom: access attribute value before printing it correct order of constructor args in createAttributeNS pulldom: use symbolic names for uri and localnames correct usage of createAttribute and setAttributeNode signatures. | 06 October 2000, 22:36:03 UTC |
| 80670bc | Martin v. Löwis | 06 October 2000, 21:13:23 UTC | Add a test case for reporting the file name, and for reporting an error for incomplete input. | 06 October 2000, 21:13:23 UTC |
| 31b485f | Martin v. Löwis | 06 October 2000, 21:12:12 UTC | In an incremental parser, close the parser after feeding everything. | 06 October 2000, 21:12:12 UTC |
| 5fece7f | Martin v. Löwis | 06 October 2000, 21:11:20 UTC | Don't use a file object as system id; try to propagate the file name to the InputSource. | 06 October 2000, 21:11:20 UTC |
| 2aa93ef | Martin v. Löwis | 06 October 2000, 21:10:34 UTC | Don't report a final chunk for an external entity parser. | 06 October 2000, 21:10:34 UTC |
| ee1dc15 | Martin v. Löwis | 06 October 2000, 21:08:59 UTC | Move translation from expat.error to SAXParseException into feed, so that callers of feed will get a SAXException. In close, feed the last chunk first before calling endDocument, so that the parser may report errors before the end of the document. Don't do anything in a nested parser. Don't call endDocument in parse; that will be called in close. Use self._source for finding the SystemID; XML_GetBase will be cleared in case of an error. | 06 October 2000, 21:08:59 UTC |
| c2bac87 | Fred Drake | 06 October 2000, 21:07:14 UTC | Include more information from the docstrings. | 06 October 2000, 21:07:14 UTC |
| 3a844a2 | Fred Drake | 06 October 2000, 21:00:40 UTC | Another name. | 06 October 2000, 21:00:40 UTC |
| db810ac | Fred Drake | 06 October 2000, 20:42:33 UTC | Donn Cave <donn@oz.net>: Fix large file support for BeOS. This closes SourceForge patch #101773. Refer to the patch discussion for information on possible alternate fixes. | 06 October 2000, 20:42:33 UTC |
| 24c532a | Fred Drake | 06 October 2000, 20:28:46 UTC | Fix a couple of typos in docstrings. | 06 October 2000, 20:28:46 UTC |
| 5197444 | Fred Drake | 06 October 2000, 20:16:50 UTC | Fix really bad typo, noted by Neil Schemenauer <nas@arctrix.com>. | 06 October 2000, 20:16:50 UTC |
| c6243e4 | Fred Drake | 06 October 2000, 20:04:48 UTC | Add notes on the requirements for subclasses. This closes SourceForge bug #115928. | 06 October 2000, 20:04:48 UTC |
| 2e06c20 | Fred Drake | 06 October 2000, 20:01:23 UTC | Revise the versioning information to say that this was revised in 1.6, not added then, and note what the change was (ncurses, change to a package). | 06 October 2000, 20:01:23 UTC |
| 062ea2e | Fred Drake | 06 October 2000, 19:59:22 UTC | Made a number of revisions suggested by Fredrik Lundh. Revised the first paragraph so it doesn't sound like it was written when 7-bit strings were assumed; note that Unicode strings can be used. | 06 October 2000, 19:59:22 UTC |
| e2b7c4d | Jeremy Hylton | 06 October 2000, 19:39:55 UTC | test_linuxaudio: read the header from the .au file and do a sanity check pass only the data to the audio device call flush() so that program does not exit until playback is complete call all the other methods to verify that they work minimally call setparameters with a bunch of bugs arguments linuxaudiodev.c: use explicit O_WRONLY and O_RDONLY instead of 1 and 0 add a string name to each of the entries in audio_types[] add AFMT_A_LAW to the list of known formats add x_mode attribute to lad object, stores imode from open call test ioctl return value as == -1, not < 0 in read() method, resize string before return add getptr() method, that calls does ioctl on GETIPTR or GETOPTR depending on x_mode in setparameters() method, do better error checking and raise ValueErrors; also use ioctl calls recommended by Open Sound System Programmer's Guido (www.opensound.com) use PyModule_AddXXX to define names in module | 06 October 2000, 19:39:55 UTC |
| d88d0a1 | Fred Drake | 06 October 2000, 19:39:47 UTC | It turns out that Guido does not like or encourage the use of the term "disciplines" for the __*__() methods, so they should be referred to as "methods" or "special methods", as appropriate in context. | 06 October 2000, 19:39:47 UTC |
| cc773d3 | Fred Drake | 06 October 2000, 19:26:01 UTC | __getslice__(): Make this use the constructor form that gets a sequence as a parameter; this was the only use of the base constructor or surgical alteration of another object's data attribute. This change simplifies the constructor requirements for subclasses. This relates to SourceForge bug #115928. | 06 October 2000, 19:26:01 UTC |
| e60de4d | Jim Fulton | 06 October 2000, 19:24:23 UTC | Added a new "base" type, IOobject for which most of the operations are defined. This will, hopefully clarify some of the logic. Added close test to raise proper error when operations are performed on closed StringIOs. Added a position argument to the truncate method. Added a size argument to readline. Added PyArg_Parse calls for methods that don't take arguments to make sure they don't take arguments. | 06 October 2000, 19:24:23 UTC |
| d1229f5 | Jim Fulton | 06 October 2000, 19:21:32 UTC | Uncommented tests that failed for cStringIO, Added missing clode to make the clode test test a close. ;) | 06 October 2000, 19:21:32 UTC |
| 6d699ca | Tim Peters | 06 October 2000, 18:46:22 UTC | Adding Jeremy's new test_import (SF patch 101709). | 06 October 2000, 18:46:22 UTC |
| 962c9e7 | Martin v. Löwis | 06 October 2000, 17:41:52 UTC | Add SAXReaderNotAvailable, and use it to distinguish between an ImportError, and a missing driver. | 06 October 2000, 17:41:52 UTC |
| 75698a4 | Guido van Rossum | 06 October 2000, 17:38:42 UTC | Removing these scripts. redemo.py lives on in Tools/scripts/. regexdemo.py is obsolete with the regex module. | 06 October 2000, 17:38:42 UTC |
| 70280b8 | Guido van Rossum | 06 October 2000, 17:37:12 UTC | I'm moving redemo.py here from Demo/tkinter/guido, since it is somewhat useful to learn regular expressions, and this way it'll be installed on Windows. This closes bug report 115609. | 06 October 2000, 17:37:12 UTC |
| 2834b97 | Guido van Rossum | 06 October 2000, 16:58:26 UTC | [ Bug #113803 ] [2.0b1 NT4.0] printing non asci char causes idle to abort http://sourceforge.net/bugs/?func=detailbug&bug_id=113803&group_id=5470 Add Unicode support and error handling to AsString(). Both AsString() and Merge() now return NULL and set a proper Python exception condition when an error happens; Merge() and other callers of AsString() check for errors from AsString(). Also fixed cleanup in Merge() and Tkapp_Call() return cleanup code; the fv array was not necessarily completely initialized, causing calls to ckfree() with garbage arguments! (Also reindented some lines that were longer than 80 chars and reformatted some code that used an alien coding standard.) | 06 October 2000, 16:58:26 UTC |
| 4764913 | Fred Drake | 06 October 2000, 16:37:47 UTC | Use the cvsinfo module instead of a module stuff off in my personal collection. | 06 October 2000, 16:37:47 UTC |
| 72fbd82 | Fred Drake | 06 October 2000, 16:36:48 UTC | Support module to help work with checked-out CVS trees. | 06 October 2000, 16:36:48 UTC |
| 60ad79f | Fred Drake | 06 October 2000, 16:17:21 UTC | Donn Cave <donn@oz.net>: New README for BeOS R5. This closes SourceForge patch #101779. | 06 October 2000, 16:17:21 UTC |
| 02776ad | Fred Drake | 06 October 2000, 16:11:21 UTC | Donn Cave <donn@u.washington.edu>: Script to regenerate platform-specific modules of constants. [I moved common paths to variables for easier reading by humans. -- FLD] This closes SourceForge patch #101781. | 06 October 2000, 16:11:21 UTC |
| b59c460 | Fred Drake | 06 October 2000, 15:57:45 UTC | Donn Cave <donn@u.washington.edu>: Added definition of VERSION so this works as expected. | 06 October 2000, 15:57:45 UTC |
| a6c2eb5 | Fred Drake | 06 October 2000, 15:48:38 UTC | Donn Cave <donn@u.washington.edu>: Do not assume that all platforms using a MetroWorks compiler can use POSIX threads; the assumption breaks on BeOS. This fix only helps for BeOS. This closes SourceForge patch #101772. | 06 October 2000, 15:48:38 UTC |
| 51d9036 | Fred Drake | 06 October 2000, 15:37:06 UTC | Norman Vine <nhv@users.sourceforge.net>: tcp.h is not available on CygWin, so do not try to include it there. This closes SourceForge patch #101802. | 06 October 2000, 15:37:06 UTC |
| 5d63a39 | Fred Drake | 06 October 2000, 15:29:56 UTC | is_zipfile() description: Use the same name for the parameter as the code, in case someone wants to use it as a keyword paramter. ZIP_DEFLATED description: Do not reveal the specific value of the constant, since code should only use the symbolic name. | 06 October 2000, 15:29:56 UTC |
| 50a22a4 | Fred Drake | 06 October 2000, 15:25:01 UTC | Added a name. | 06 October 2000, 15:25:01 UTC |
| 36ed560 | Fred Drake | 06 October 2000, 01:58:48 UTC | Added section on threads problems on Reliant UNIX; this relates to bug #113797. We should be able to resolve this for the next release. Reflowed the comments on Monterey (64-bit AIX) to match the flow of the other platform-specific sections. | 06 October 2000, 01:58:48 UTC |
| 00236f3 | Guido van Rossum | 06 October 2000, 00:38:51 UTC | [ Bug #110677 ] PRIVATE: various minor Tkinter things (PR#388) http://sourceforge.net/bugs/?func=detailbug&group_id=5470&bug_id=110677 Canvas.CanvasItem & Canvas.Group: - bind lacks an optional "add" param - unbind lacks an optional "funcid" param - tkraise/lower should call self.canvas.tag_XXXX (markus.oberhumer@jk.uni-linz.ac.at) Note: I'm *not* fixing "bbox() return value is inconsistent with Canvas.bbox()" -- it might break existing code. | 06 October 2000, 00:38:51 UTC |
| c54d190 | Tim Peters | 06 October 2000, 00:36:09 UTC | SF bug 115831 and Ping's SF patch 101751, 0.0**-2.0 returns inf rather than raise ValueError. Checked in the patch as far as it went, but also changed all of ints, longs and floats to raise ZeroDivisionError instead when raising 0 to a negative number. This is what 754-inspired stds require, as the "true result" is an infinity obtained from finite operands, i.e. it's a singularity. Also changed float pow to not be so timid about using its square-and-multiply algorithm. Note that what math.pow does is unrelated to what builtin pow does, and will still vary by platform. | 06 October 2000, 00:36:09 UTC |
| 4779a0a | Mark Hammond | 05 October 2000, 22:10:33 UTC | Remove some debugging messages - although this code is a complete hack, we dont need to announce it to the world every time they use freeze! | 05 October 2000, 22:10:33 UTC |
| 034c2a9 | Fred Drake | 05 October 2000, 20:42:44 UTC | Add support for "import re" -- it uses pre, but user code does not need to. | 05 October 2000, 20:42:44 UTC |
| 410cb6b | Neil Schemenauer | 05 October 2000, 19:38:24 UTC | The _PyTuple_Resize() last_is_sticky flag must now always be false. | 05 October 2000, 19:38:24 UTC |
| 08b53e6 | Neil Schemenauer | 05 October 2000, 19:36:49 UTC | Simplify _PyTuple_Resize by not using the tuple free list and dropping support for the last_is_sticky flag. A few hard to find bugs may be fixed by this patch since the old code was buggy. | 05 October 2000, 19:36:49 UTC |
