Fred Drake [Thu, 21 Sep 2000 05:25:30 +0000 (05:25 +0000)]
Fixed a number of small problems reported by Detlef Lannert
<lannert@uni-duesseldorf.de>.
Greg Ward [Thu, 21 Sep 2000 01:23:35 +0000 (01:23 +0000)]
Corran Webster: fix 'change_root()' to handle Mac OS paths.
Guido van Rossum [Wed, 20 Sep 2000 20:31:38 +0000 (20:31 +0000)]
On Unix, use O_EXCL when creating the .pyc/.pyo files, to avoid a race condition
Guido van Rossum [Wed, 20 Sep 2000 20:24:21 +0000 (20:24 +0000)]
Add a definition of rl_library_version. Reported by jpettit@sourceforge.
Fred Drake [Wed, 20 Sep 2000 05:49:09 +0000 (05:49 +0000)]
.use_latex(): Make this a little smarter so that it only runs pdflatex
once if all the temporary files are available from building
a DVI file. This can avoid two runs of pdflatex.
Fred Drake [Wed, 20 Sep 2000 04:48:20 +0000 (04:48 +0000)]
Clarify that the softspace attribute is used by print for state management,
not as a user-controlled parameter.
Fred Drake [Wed, 20 Sep 2000 02:54:58 +0000 (02:54 +0000)]
Add entries for the xml.sax documentation.
Fred Drake [Wed, 20 Sep 2000 02:52:20 +0000 (02:52 +0000)]
First cut at documentation for the xml.sax package (not including any
sub-modules).
Jeremy Hylton [Wed, 20 Sep 2000 02:47:28 +0000 (02:47 +0000)]
change 2-space indent to 4-space indent
Jeremy Hylton [Wed, 20 Sep 2000 02:35:19 +0000 (02:35 +0000)]
distutils setup script
Guido van Rossum [Wed, 20 Sep 2000 00:17:39 +0000 (00:17 +0000)]
Fix typo (newtabwith).
Greg Ward [Tue, 19 Sep 2000 23:56:43 +0000 (23:56 +0000)]
*Very* belated application of Thomas Heller's patch to handle
resource files. The gist of the patch is to treat ".rc" and ".mc"
files as source files; ".mc" files are compiled to ".rc" and then
".res", and ".rc" files are compiled to ".res". Wish I knew what
all these things stood for...
Tim Peters [Tue, 19 Sep 2000 23:46:56 +0000 (23:46 +0000)]
Treat trailing colon in os.path.join("a:", "b") same way for DOS as in
recent changes to ntpath.py and posixmodule.c. Thanks to Guido for
pointing out the inconsistency!
Jack Jansen [Tue, 19 Sep 2000 22:42:38 +0000 (22:42 +0000)]
Added GetArgs dialog to EasyDialogs, a very nifty (if I may say so:-) way to create a unix-style sys.argv.
Marc-André Lemburg [Tue, 19 Sep 2000 21:04:18 +0000 (21:04 +0000)]
This patch adds a new Python C API called PyString_AsStringAndSize()
which implements the automatic conversion from Unicode to a string
object using the default encoding.
The new API is then put to use to have eval() and exec accept
Unicode objects as code parameter. This closes bugs #110924
and #113890.
As side-effect, the traditional C APIs PyString_Size() and
PyString_AsString() will also accept Unicode objects as
parameters.
Guido van Rossum [Tue, 19 Sep 2000 20:51:17 +0000 (20:51 +0000)]
Temporary fix for Bug #114821.
The cause was that the replace code necessarily used a PCRE internal
function to to template expansion.
The fix changes the code to use an SRE internal if SRE is used, and a
PCRE internal if SRE is used; in a way that should work with 1.5.2.
The solution can be sped up tremendously under the assumption that the
choice between sre and pre is not changed during the execution of the
program; especially replace-all will be slow.
But I'll leave that to someone else.
Tim Peters [Tue, 19 Sep 2000 20:39:32 +0000 (20:39 +0000)]
Part of SF patch 101481: on Windows, os.path.join("a:", "b") should yield
"a:b", not "a:/b". Similar change was made to posixmodule.c earlier.
Tim Peters [Tue, 19 Sep 2000 20:31:24 +0000 (20:31 +0000)]
test_userlist.py:
Added new test for new __contains__ method.
Extensive editing to get rid of asserts.
Tim Peters [Tue, 19 Sep 2000 20:29:03 +0000 (20:29 +0000)]
SF patch 101391: implemented UserList.__contains__.
Paul Prescod [Tue, 19 Sep 2000 19:33:02 +0000 (19:33 +0000)]
Fix test errors.
Fred Drake [Tue, 19 Sep 2000 18:21:25 +0000 (18:21 +0000)]
Moshe Zadka <mzadka@geocities.com>:
Document the __contains__() method.
(Patch slightly modified by FLD.)
This closes SourceForge patch #101387.
Fred Drake [Tue, 19 Sep 2000 18:10:05 +0000 (18:10 +0000)]
Many small clarifications, including many suggested by email.
Martin v. Löwis [Tue, 19 Sep 2000 16:35:39 +0000 (16:35 +0000)]
New test cases for the StringIO module
Martin v. Löwis [Tue, 19 Sep 2000 16:22:10 +0000 (16:22 +0000)]
Don't print specific Node instances unless running verbosely.
Closes Bug #114775.
Fred Drake [Tue, 19 Sep 2000 15:36:19 +0000 (15:36 +0000)]
make_nav_sectref(): Create more minimal HTML, hanging the class attribute
off an existing anchor tag if available (I think it
always is, but am not completely sure).
Fred Drake [Tue, 19 Sep 2000 15:18:51 +0000 (15:18 +0000)]
Clarify a number of issues about the file-like object API based on
discussion on python-dev.
Barry Warsaw [Tue, 19 Sep 2000 14:42:09 +0000 (14:42 +0000)]
Fixed the error reporting (raise of TestFailed) for the zip() and
zip(None) tests. Found by Finn Bock a while ago.
Guido van Rossum [Tue, 19 Sep 2000 13:35:40 +0000 (13:35 +0000)]
Randall Hopper: Add fpectl functionality patch for FreeBSD.
Martin v. Löwis [Tue, 19 Sep 2000 11:25:58 +0000 (11:25 +0000)]
Support sizehint in _fileobject.readlines, as documented.
Thomas Heller [Tue, 19 Sep 2000 11:10:23 +0000 (11:10 +0000)]
Set the 'nt' installation scheme for the install command even if run
on other systems, so that data, headers, scripts are included in
the installer.
Martin v. Löwis [Tue, 19 Sep 2000 11:07:44 +0000 (11:07 +0000)]
Support sizehint in StringIO.readlines, as documented.
Martin v. Löwis [Tue, 19 Sep 2000 11:06:46 +0000 (11:06 +0000)]
Implement readlines function. Closes Bug #110686.
Tim Peters [Tue, 19 Sep 2000 08:54:13 +0000 (08:54 +0000)]
Obscure marshal fixes:
When reading a short, sign-extend on platforms where shorts are
bigger than 16 bits.
When reading a long, repair the unportable sign extension that was
being done for 64-bit machines (it assumed that signed right shift
sign-extends).
Guido van Rossum [Tue, 19 Sep 2000 04:11:46 +0000 (04:11 +0000)]
Patch #101121, by Ka-Ping Yee: cosmetic cleanup of cgi.py, using my
style conventions. (Ping has checkin privileges but apparently
ignores them at the moment.)
Ping improves a few doc strings and fixes style violations like foo ( bar ).
An addition of my own: rearrange the printing of various items in
test() so that the (long) environment comes at the end. This avoids
having to scroll if you want to see the current directory or command
line arguments.
Guido van Rossum [Tue, 19 Sep 2000 04:01:01 +0000 (04:01 +0000)]
An honest attempt to make this work on Unix, Windows, and even
Macintosh (the latter untested).
This closes Bug #110839.
Guido van Rossum [Tue, 19 Sep 2000 03:04:52 +0000 (03:04 +0000)]
Only supply popen2, popen3 when fork exists.
(This avoids defining non-working versions of these on the Mac.)
Guido van Rossum [Tue, 19 Sep 2000 00:46:46 +0000 (00:46 +0000)]
Make better use of GNU Pth -- patch by Andy Dustman.
I can't test this, so I'm just checking it in with blind faith in Andy.
I've tested that it doesn't broeak a non-Pth build on Linux.
Changes include:
- There's a --with-pth configure option.
- Instead of _GNU_PTH, we test for HAVE_PTH.
- Better signal handling.
- (The config.h.in file is regenerated in a slightly different order.)
Tim Peters [Tue, 19 Sep 2000 00:38:35 +0000 (00:38 +0000)]
Updated Windows build info.
Jeremy Hylton [Mon, 18 Sep 2000 22:50:38 +0000 (22:50 +0000)]
Do not close socket when a Content-Length is 0. This make the
interface consistent: The client is responsible for closing the
socket, regardless of the amount of data received.
Restore suport for set_debuglevel call.
Paul Prescod [Mon, 18 Sep 2000 20:44:15 +0000 (20:44 +0000)]
Test output that goes with updated test_minidom.
Fred Drake [Mon, 18 Sep 2000 19:38:11 +0000 (19:38 +0000)]
Fix serious typo!
Add the new constants to the module docstring.
Paul Prescod [Mon, 18 Sep 2000 19:07:26 +0000 (19:07 +0000)]
Change assertions to confirmations so that optimization doesn't disable
checks.
Fred Drake [Mon, 18 Sep 2000 17:56:58 +0000 (17:56 +0000)]
Fix up the cleanup of the temporary DB so it works for BSD DB's
compatibility layer as well as "classic" ndbm.
Fred Drake [Mon, 18 Sep 2000 17:40:22 +0000 (17:40 +0000)]
Reduce the number of imports needed.
Make the code conform better to the Python style guide.
Fred Drake [Mon, 18 Sep 2000 16:48:13 +0000 (16:48 +0000)]
Document the "printable" and "punctuation" constants added to the string
module.
Fred Drake [Mon, 18 Sep 2000 16:46:17 +0000 (16:46 +0000)]
Richard Mortier <rmm1002@users.sourceforge.net>:
Add the constants "printable" and "punctuation" to the string module.
Fred Drake [Mon, 18 Sep 2000 16:31:08 +0000 (16:31 +0000)]
Remove two unnecessary imports.
Update the module docstring to reflect the actual list of modules in the
xml.sax package.
Make the code better conform to the Python style guide.
Fred Drake [Mon, 18 Sep 2000 16:22:30 +0000 (16:22 +0000)]
Once we're done with the sys module, remove it from the namespace so
this module is "import *" safe.
Marc-André Lemburg [Mon, 18 Sep 2000 16:22:27 +0000 (16:22 +0000)]
Deferred the attribute name object type checking to the underlying
PyObject_Set/GetAttr() calls.
This patch fixes bug #113829.
Fred Drake [Mon, 18 Sep 2000 16:21:11 +0000 (16:21 +0000)]
Add missing \declaremodule for zipfile.
Marc-André Lemburg [Mon, 18 Sep 2000 16:20:57 +0000 (16:20 +0000)]
PyObject_SetAttr() and PyObject_GetAttr() now also accept Unicode
objects for the attribute name. Unicode objects are converted to
a string using the default encoding before trying the lookup.
Note that previously it was allowed to pass arbitrary objects as
attribute name in case the tp_getattro/setattro slots were defined.
This patch fixes this by applying an explicit string check first:
all uses of these slots expect string objects and do not check
for the type resulting in a core dump. The tp_getattro/setattro
are still useful as optimization for lookups using interned
string objects though.
This patch fixes bug #113829.
Tim Peters [Mon, 18 Sep 2000 15:34:57 +0000 (15:34 +0000)]
Repaired some glitches in the MD5 and SHA docs; copied the descriptions of
the MD5 methods into the SHA docs (substituting "sha" for "md5", of course,
and changing the stuff that depended on digest size accordingly).
Fred, don't trust me!
Marc-André Lemburg [Mon, 18 Sep 2000 11:06:00 +0000 (11:06 +0000)]
sys.setdefaultencoding() should only be called in case the standard
default encoding ("ascii") is changed. This safes quite a few cycles
during startup since the first call to .setdefaultencoding() will
initialize the codec registry and the encodings package.
See python-dev for a discussion (Subject: "[Python-Dev] [comp.lang.python] sys.setdefaultencoding (2.0b1)").
Jeremy Hylton [Mon, 18 Sep 2000 01:46:01 +0000 (01:46 +0000)]
satisfy the tabnanny
Greg Ward [Mon, 18 Sep 2000 00:41:10 +0000 (00:41 +0000)]
Catch up to recent changes in TextFile (spotted by Bastian Kleineidam).
Fred Drake [Sun, 17 Sep 2000 16:10:25 +0000 (16:10 +0000)]
Added documentation for new functions.
Deprecated sequenceIncludes().
Based on patch by Denis S. Otkidach <ods@users.sourceforge.net>,
this closes SourceForge patch #101390.
Fred Drake [Sun, 17 Sep 2000 16:09:27 +0000 (16:09 +0000)]
Add invert() and __invert__() as aliases for inv()/__inv__().
Add contains() as alias for __contains__().
Make PyArg_ParseTuple() formats include the function name.
Based on patch by Denis S. Otkidach <ods@users.sourceforge.net>,
this closes SourceForge patch #101390.
Tim Peters [Sun, 17 Sep 2000 14:40:17 +0000 (14:40 +0000)]
Fix for SF bug 110688: Instance deallocation neglected to account for
that Py_INCREF boosts global _Py_RefTotal when Py_REF_DEBUG is defined
but Py_TRACE_REFS isn't.
There are, IMO, way too many preprocessor gimmicks in use for refcount
debugging (at least 3 distinct true/false symbols, but not all 8 combos
are supported by the code, etc etc), and no coherent documentation of
this stuff -- 'twas too painful to track this one down.
Greg Ward [Sun, 17 Sep 2000 00:54:58 +0000 (00:54 +0000)]
Fixed to respect 'define_macros' and 'undef_macros' on Extension object.
Greg Ward [Sun, 17 Sep 2000 00:53:02 +0000 (00:53 +0000)]
Added 'expand_makefile_vars()' to (duh) expand make-style variables
in a string (gives you something to do with the dictionary returned
by 'parse_makefile()').
Pulled the regexes in 'parse_makefile()' out -- they're now globals,
as 'expand_makefile_vars()' needs (two of) them.
Cosmetic tweaks to 'parse_makefile()'.
Greg Ward [Sun, 17 Sep 2000 00:45:18 +0000 (00:45 +0000)]
Added 'read_setup_file()' to read old-style Setup files. Could make life
easier for people porting Makefile.pre.in-based extensions to Distutils.
Also loosened argument-checking in Extension constructor to make life
easier for 'read_setup_file()'.
Tim Peters [Sat, 16 Sep 2000 22:31:29 +0000 (22:31 +0000)]
arraymodule: Fix SF bug 113960.
reverse() didn't work at all due to bad arg check.
Fixed that.
Added Brad Chapman to ACKS file, as the proud new owner of two
implicitly copyrighted lines of Python source code <wink>.
Repaired buffer_info's total lack of arg-checking.
Replaced memmove by memcpy in reverse() guts, as memmove is
often slower and the memory areas are guaranteed disjoint.
Replaced poke-and-hope unchecked decl of tmp buffer size by
assert-checked larger tmp buffer.
Got rid of inconsistent spaces before open paren in docstrings.
Added reverse() sanity tests to test_array.py.
Barry Warsaw [Sat, 16 Sep 2000 22:09:51 +0000 (22:09 +0000)]
Allow this script to act like a module by only calling main() if
__name__ == '__main__'. Closes SF bug #110844.
Fred Drake [Sat, 16 Sep 2000 21:31:33 +0000 (21:31 +0000)]
Make the <body> tag match those of the generated HTML; rely on the style
sheet instead of encoding anything there.
Fred Drake [Sat, 16 Sep 2000 21:23:25 +0000 (21:23 +0000)]
Document \cfuncdesc, \ctypedesc, and \cvardesc.
Guido van Rossum [Sat, 16 Sep 2000 21:16:01 +0000 (21:16 +0000)]
Improve the test output a bit.
Fred Drake [Sat, 16 Sep 2000 20:40:44 +0000 (20:40 +0000)]
Push more of the display control into the style sheet.
Greg Ward [Sat, 16 Sep 2000 18:33:36 +0000 (18:33 +0000)]
Changed so lines that are all comment (or just whitespace + comment)
are completely skipped, rather than being treated as blank lines
(and then subject to the 'skip_blanks' flag). This allows us
to process old-style Setup files, which rely on
hello \\
# boo!
there
coming out as "hello there".
Greg Ward [Sat, 16 Sep 2000 18:09:22 +0000 (18:09 +0000)]
Andrew Kuchling:
Fixed precendence bug that meant setting skip_blanks to false didn't work
under some circumstances.
Greg Ward [Sat, 16 Sep 2000 18:06:31 +0000 (18:06 +0000)]
[change from 2000/08/11, propagating now to distutils copy]
Factored the guts of 'warn()' out to 'gen_error()', and added the
'error()' method (trivial thanks to the refactoring).
Greg Ward [Sat, 16 Sep 2000 18:04:55 +0000 (18:04 +0000)]
[change from 2000/04/17, propagating now to distutils copy]
Dropped the 'collapse_ws' option and replaced it with 'collapse_join' --
it's *much* faster (no 're.sub()') and this is the reason I really added
'collapse_ws', ie. to remove leading whitespace from a line being joined
to the previous line.
Guido van Rossum [Sat, 16 Sep 2000 16:37:53 +0000 (16:37 +0000)]
Use PyOS_setsig() instead of directly calling signal() or sigaction().
This fixes the first half of bug #110611: the immediate exit when ^C
is hit when readline and threads are configured.
Also added a new module variable, readline.library_version.
Guido van Rossum [Sat, 16 Sep 2000 16:35:28 +0000 (16:35 +0000)]
Use typedef PyOS_sighandler_t and APIs PyOS_getsig() and
PyOS_setsig(), instead of directly calling signal() or sigaction().
This fixes the second half of bug #110611: the mysterious ignoring of
the first ^C when readline isn't used.
Guido van Rossum [Sat, 16 Sep 2000 16:32:19 +0000 (16:32 +0000)]
Add PyOS_getsig() and PyOS_setsig() -- wrappers around signal() or
sigaction() (if HAVE_SIGACTION is defined).
Guido van Rossum [Sat, 16 Sep 2000 16:31:31 +0000 (16:31 +0000)]
Add typedef PyOS_sighandler_t and prototypes for PyOS_getsig() and
PyOS_setsig().
Guido van Rossum [Sat, 16 Sep 2000 16:30:48 +0000 (16:30 +0000)]
Document new APIs PyOS_getsig() and PyOS_setsig().
(Is there no macro to document a typedef?)
Fred, please check my latex!
Greg Ward [Sat, 16 Sep 2000 16:04:59 +0000 (16:04 +0000)]
Rene Liebscher: if we have to run the same sub-command multiple times
(eg. "bdist_dumb", to generate both ZIP and tar archives in the same
run), tell all but the last run to keep temp files -- this just gets
rid of the need to pseudo-install the same files multiple times.
Greg Ward [Sat, 16 Sep 2000 15:56:32 +0000 (15:56 +0000)]
Renamed --keep-tree to --keep-temp.
Greg Ward [Sat, 16 Sep 2000 15:54:18 +0000 (15:54 +0000)]
Renamed --clean to --no-keep-temp and --noclean to --keep-temp.
Greg Ward [Sat, 16 Sep 2000 15:53:41 +0000 (15:53 +0000)]
Renamed --keep-tree option to --keep-temp.
Greg Ward [Sat, 16 Sep 2000 15:30:47 +0000 (15:30 +0000)]
Ensure sub-commands of "install" are reinitialized too.
Run "install" the right way, by calling 'run_command()'.
Greg Ward [Sat, 16 Sep 2000 15:27:17 +0000 (15:27 +0000)]
Remove some debugging output from the last change.
Greg Ward [Sat, 16 Sep 2000 15:25:55 +0000 (15:25 +0000)]
In 'reinitialize_subcommand()', pass 'reinit_subcommands' flag on to the
real implementation in Distribution.
Greg Ward [Sat, 16 Sep 2000 15:23:28 +0000 (15:23 +0000)]
Generalized 'reinitialize_command()' so it can optionally reinitialize
the command's sub-commands as well (off by default). This is essential if
we want to be be able to run (eg.) "install" twice in one run, as happens
when generating multiple built distributions in one run.
Greg Ward [Sat, 16 Sep 2000 15:09:17 +0000 (15:09 +0000)]
Added the "sub-command" machinery to formalize the notion of "command
families" -- eg. install and its brood, build and its brood, and so forth.
Specifically: added the 'sub_commands' class attribute (empty list, sub-
classes must override it) and a comment describing it, and the
'get_sub_commands()' method.
Greg Ward [Sat, 16 Sep 2000 15:06:57 +0000 (15:06 +0000)]
Factored the "sub-command" machinery out to Command. Mainly, this
meant removing 'get_sub_commands()', and moving the 'sub_commands' class
attribute to the end and restructuring it to conform to the new regime.
Fred Drake [Sat, 16 Sep 2000 14:16:57 +0000 (14:16 +0000)]
Use a really ugly hack that depends on output processing by LaTeX2HTML.
This replaces the tilde character with the string ";SPMtilde;", which
LaTeX2HTML converts to a tilde on output.
Fred Drake [Sat, 16 Sep 2000 13:46:42 +0000 (13:46 +0000)]
Update URL to unicode.org's FTP site using an HTTP URL to the same
document.
Fred Drake [Sat, 16 Sep 2000 06:18:26 +0000 (06:18 +0000)]
Update *both* uses of the URL to the zlib home page, and use the version
advertised as the canonical form.
Fred Drake [Sat, 16 Sep 2000 05:07:03 +0000 (05:07 +0000)]
Update link to the zlib homepage.
Fred Drake [Sat, 16 Sep 2000 05:02:24 +0000 (05:02 +0000)]
Update the link to Andrew Kuchling's crypto code.
Fred Drake [Sat, 16 Sep 2000 04:59:33 +0000 (04:59 +0000)]
Update link to the FIPS PUB 180-1 (PDF is available, but not PostScript).
Fred Drake [Sat, 16 Sep 2000 04:21:22 +0000 (04:21 +0000)]
Update some stale links.
Tim Peters [Sat, 16 Sep 2000 04:02:48 +0000 (04:02 +0000)]
Repair senseless random.seed docstring (reported on c.l.py).
Tim Peters [Sat, 16 Sep 2000 03:54:24 +0000 (03:54 +0000)]
Fixed legit gripe from c.l.py that math.fmod docs aren't confusing enough.
FRED, please check my monkey-see-monkey-do Tex fiddling!
Greg Ward [Sat, 16 Sep 2000 02:06:45 +0000 (02:06 +0000)]
Tweaked the build temp dir names again.
Greg Ward [Sat, 16 Sep 2000 01:59:06 +0000 (01:59 +0000)]
Typo fix.
Greg Ward [Sat, 16 Sep 2000 01:54:46 +0000 (01:54 +0000)]
Include the Python version in the platform-specific build directories:
with the recent change in 'get_platform()', we now have directory names
like "build/lib-1.5-linux-i586". Idea and original patch by
Rene Liebscher.
Greg Ward [Sat, 16 Sep 2000 01:44:45 +0000 (01:44 +0000)]
Document the directory separatory for include dir and library dir lists.
Jack Jansen [Fri, 15 Sep 2000 22:44:08 +0000 (22:44 +0000)]
This will eventually evolve into a package that creates CodeWarrior projects from templates (for distutils, and possibly the standard extensions too). For now it generates a correct xml file based on sources/output/expfile, but that is about all.