]> granicus.if.org Git - python/log
python
24 years agoSupport for the in-place operations introduced by augmented assignment. Only
Thomas Wouters [Thu, 24 Aug 2000 20:08:19 +0000 (20:08 +0000)]
Support for the in-place operations introduced by augmented assignment. Only
the list object supports this currently, but other candidates are
gladly accepted (like arraymodule and such.)

24 years agoRough and incomplete documentation on augmented assignment, which follows
Thomas Wouters [Thu, 24 Aug 2000 20:06:04 +0000 (20:06 +0000)]
Rough and incomplete documentation on augmented assignment, which follows
shortly. Markup also needs checking.

24 years agoRene Liebscher <rliebscher@users.sourceforge.net>:
Fred Drake [Thu, 24 Aug 2000 18:11:21 +0000 (18:11 +0000)]
Rene Liebscher <rliebscher@users.sourceforge.net>:
Install the scripts in the BeOS directory, allowing use from the distutils
on BeOS systems.

The target location has been adjusted to match that used for the AIX
helper scripts.

This closes SourceForge patch #101207.

24 years agoAddresses the other half of Bug #112634 -- the documentation suggested
Guido van Rossum [Thu, 24 Aug 2000 16:20:32 +0000 (16:20 +0000)]
Addresses the other half of Bug #112634 -- the documentation suggested
that you can set self.version *after* calling the base class __init__.
In fact it must be done *before*.

(Fred, maybe the version class variable should be documented now?)

24 years agoPromote the server version from a local variable to a class variable,
Guido van Rossum [Thu, 24 Aug 2000 16:18:04 +0000 (16:18 +0000)]
Promote the server version from a local variable to a class variable,
so that a subclass can override it.

This partly addresses Bug #112634 -- but the documentation is still
wrong, since it suggests that you can set self.version *after* calling
the base class __init__.  In fact it must be done *before*.

I'll fix that too.

24 years agoWhitespace cleanup; now passes the regression test (the last checkin made
Fred Drake [Thu, 24 Aug 2000 14:40:35 +0000 (14:40 +0000)]
Whitespace cleanup; now passes the regression test (the last checkin made
it fail on a TabError (inconsistent tab/space usage)).

Removed a comment about including a test since there is a regression test
for this module.

24 years agoUpdated test suite: test repr() and str() of cookies, and test metadata
Andrew M. Kuchling [Thu, 24 Aug 2000 11:56:19 +0000 (11:56 +0000)]
Updated test suite: test repr() and str() of cookies, and test metadata
    fields with quoted values (as in Path="/acme")

24 years agoUpdated version of Cookie.py (rev. 2.29) from timo
Andrew M. Kuchling [Thu, 24 Aug 2000 11:52:33 +0000 (11:52 +0000)]
Updated version of Cookie.py (rev. 2.29) from timo

24 years agoMove references to RFCs to a "See also" section for consistency with
Fred Drake [Thu, 24 Aug 2000 04:58:25 +0000 (04:58 +0000)]
Move references to RFCs to a "See also" section for consistency with
other sections of the library reference.

24 years agoMoshe Zadka <moshez@math.huji.ac.il>:
Fred Drake [Thu, 24 Aug 2000 01:06:40 +0000 (01:06 +0000)]
Moshe Zadka <moshez@math.huji.ac.il>:
Documentation updates for urlretrieve() and URLopener.retrieve(), to
reflect Randall Hopper's patch for those functions.

This closes SourceForge patch #100837.

24 years agoRandall Hopper <aa8vb@yahoo.com>>:
Fred Drake [Thu, 24 Aug 2000 01:01:26 +0000 (01:01 +0000)]
Randall Hopper <aa8vb@yahoo.com>>:
Make it easier to use HTTP POST with urlretrieve().

24 years agoMake sure Thomas Wouters has enough to do. ;-)
Fred Drake [Thu, 24 Aug 2000 00:50:27 +0000 (00:50 +0000)]
Make sure Thomas Wouters has enough to do.  ;-)
(Some of this is in-progress; this also serves as a reminded to me to
check this out once it is written.)

24 years agoCharles G. Waldman <cgw@fnal.gov>:
Fred Drake [Thu, 24 Aug 2000 00:37:50 +0000 (00:37 +0000)]
Charles G. Waldman <cgw@fnal.gov>:
Update the dis module documentation to reflect the EXTENDED_ARG opcode.

24 years agoCharles G. Waldman <cgq@fnal.gov>:
Fred Drake [Thu, 24 Aug 2000 00:35:38 +0000 (00:35 +0000)]
Charles G. Waldman <cgq@fnal.gov>:
Update the test suite for the changes introduced by the EXTENDED_ARG
opcode.

This closes the regression test changes of SourceForge patch #100893.

24 years agoCharles G. Waldman <cgw@fnal.gov>:
Fred Drake [Thu, 24 Aug 2000 00:32:09 +0000 (00:32 +0000)]
Charles G. Waldman <cgw@fnal.gov>:
Add the EXTENDED_ARG opcode to the virtual machine, allowing 32-bit
arguments to opcodes instead of being forced to stick to the 16-bit
limit.  This is especially useful for machine-generated code, which
can be too long for the SET_LINENO parameter to fit into 16 bits.

This closes the implementation portion of SourceForge patch #100893.

24 years agoAddendum to previous change: now that 'f' is not unconditionally
Thomas Wouters [Wed, 23 Aug 2000 23:31:34 +0000 (23:31 +0000)]
Addendum to previous change: now that 'f' is not unconditionally
initialized in the 'if (..)', do so manually.

24 years agoAdd extra check on whether 'tp_as_number' is still non-NULL after coercion,
Thomas Wouters [Wed, 23 Aug 2000 23:16:10 +0000 (23:16 +0000)]
Add extra check on whether 'tp_as_number' is still non-NULL after coercion,
in the PyNumber_* functions. Also, remove unnecessary tests from
PyNumber_Multiply: after BINOP(), neither argument can be an instance.

24 years agoThis patch partly (some stuff went in already) ports Python to Monterey.
Trent Mick [Wed, 23 Aug 2000 21:33:05 +0000 (21:33 +0000)]
This patch partly (some stuff went in already) ports Python to Monterey.

- Fix bug in thread_pthread.h::PyThread_get_thread_ident() where
  sizeof(pthread) < sizeof(long).
- Add 'configure' for:
- SIZEOF_PTHREAD is pthread_t can be included via <pthread.h>
- setting Monterey system name
- appropriate CC,LINKCC,LDSHARED,OPT, and CCSHARED for Monterey
- Add section in README for Monterey build

24 years agoRemove the dependency information for version.o; this is not part of
Fred Drake [Wed, 23 Aug 2000 21:16:10 +0000 (21:16 +0000)]
Remove the dependency information for version.o; this is not part of
the sources/build process any more.

24 years agostyle nits
Jeremy Hylton [Wed, 23 Aug 2000 21:11:05 +0000 (21:11 +0000)]
style nits

24 years agoWhen skipping a test, do not include a spurious space between the exception
Fred Drake [Wed, 23 Aug 2000 20:34:40 +0000 (20:34 +0000)]
When skipping a test, do not include a spurious space between the exception
name and the ":" that separates it from the value.  (Minor cleanup.)

24 years agoadd docstring explaining makefile limitation
Jeremy Hylton [Wed, 23 Aug 2000 20:34:17 +0000 (20:34 +0000)]
add docstring explaining makefile limitation
fix support for passing http version to connect in HTTP

24 years agoSort the list of files processed before running the test on each.
Fred Drake [Wed, 23 Aug 2000 20:21:31 +0000 (20:21 +0000)]
Sort the list of files processed before running the test on each.

24 years agotest case output for dospath regr test
Skip Montanaro [Wed, 23 Aug 2000 20:07:21 +0000 (20:07 +0000)]
test case output for dospath regr test

24 years agoUpdated some URLs; removed mention of copyright (we'll have to add
Guido van Rossum [Wed, 23 Aug 2000 19:01:43 +0000 (19:01 +0000)]
Updated some URLs; removed mention of copyright (we'll have to add
something in later after that discussion is over); remove explanation
of 2.0 version number.

24 years agoSimplified inclusions and avoid prototypes copied in from elsewhere.
Fred Drake [Wed, 23 Aug 2000 18:17:42 +0000 (18:17 +0000)]
Simplified inclusions and avoid prototypes copied in from elsewhere.
This also avoids a warning in anal mode.

24 years agofix list comprehension discussion to use \keyword macro instead of simply
Skip Montanaro [Wed, 23 Aug 2000 17:03:34 +0000 (17:03 +0000)]
fix list comprehension discussion to use \keyword macro instead of simply
quoting keywords.

24 years agoupdate commonprefix doc to make sure user is aware it works
Skip Montanaro [Wed, 23 Aug 2000 16:58:32 +0000 (16:58 +0000)]
update commonprefix doc to make sure user is aware it works
character-by-character and that that behavior may result in it returning
prefixes that are not valid paths

24 years agonew test - like test_ntpath but without splitunc test cases
Skip Montanaro [Wed, 23 Aug 2000 16:55:00 +0000 (16:55 +0000)]
new test - like test_ntpath but without splitunc test cases

24 years agoadded commonprefix test cases
Skip Montanaro [Wed, 23 Aug 2000 16:54:27 +0000 (16:54 +0000)]
added commonprefix test cases

24 years agoupdated to correct failing test cases
Skip Montanaro [Wed, 23 Aug 2000 16:51:56 +0000 (16:51 +0000)]
updated to correct failing test cases

24 years agoAdded note stating that the parser module will need updating for most
Fred Drake [Wed, 23 Aug 2000 15:45:28 +0000 (15:45 +0000)]
Added note stating that the parser module will need updating for most
changes to this file.

24 years agovalidate_listmaker(): Revise to match Skip's latest changes to the
Fred Drake [Wed, 23 Aug 2000 15:35:26 +0000 (15:35 +0000)]
validate_listmaker():  Revise to match Skip's latest changes to the
      Grammar file.  This makes the test suite pass once again.

24 years agoRestored commonprefix() semantics.
Jack Jansen [Wed, 23 Aug 2000 09:13:40 +0000 (09:13 +0000)]
Restored commonprefix() semantics.

24 years agoRehabilitate autotest.py.
Tim Peters [Wed, 23 Aug 2000 05:28:45 +0000 (05:28 +0000)]
Rehabilitate autotest.py.
In README:  Write up (Guido's) rules for intra-test imports; warn against
asserts; document test_support.use_large_resources.

24 years agoAdded PyOS_CheckStack call to PyObject_Compare
Jack Jansen [Tue, 22 Aug 2000 21:52:51 +0000 (21:52 +0000)]
Added PyOS_CheckStack call to PyObject_Compare
Lowered the recursion limit on compares to 60 (one recursion depth can
take a whopping 2K of stack space when running test_b1!)

24 years agoAdded include for limits.h
Jack Jansen [Tue, 22 Aug 2000 21:51:22 +0000 (21:51 +0000)]
Added include for limits.h

24 years agoRegenerated to take advantage of support for multiple names for a single property.
Jack Jansen [Tue, 22 Aug 2000 20:35:17 +0000 (20:35 +0000)]
Regenerated to take advantage of support for multiple names for a single property.

24 years agoOnly output the code to alias properties if the names aren't the same.
Jack Jansen [Tue, 22 Aug 2000 20:34:35 +0000 (20:34 +0000)]
Only output the code to alias properties if the names aren't the same.

24 years agoAt Mark Hammond's request, removing registry key
Guido van Rossum [Tue, 22 Aug 2000 14:04:24 +0000 (14:04 +0000)]
At Mark Hammond's request, removing registry key
Software\Python\PythonCore\2.0\Dll\Python20.dll -- it's no longer
needed according to him.

Note: not yet tested!

24 years agorevert semantics of commonprefix to work character-by-character
Skip Montanaro [Tue, 22 Aug 2000 13:01:53 +0000 (13:01 +0000)]
revert semantics of commonprefix to work character-by-character

24 years agoThomas Heller noticed that the wrong registry entry was written for
Guido van Rossum [Tue, 22 Aug 2000 12:10:02 +0000 (12:10 +0000)]
Thomas Heller noticed that the wrong registry entry was written for
the DLL.  Replace
 %_SYSDEST_%\Python20.dll
with
 %_DLLDEST_%\Python20.dll.

24 years agoRegistered modules could only exist in HKEY_LOCAL_MACHINE - now HKEY_CURRENT_USER...
Mark Hammond [Tue, 22 Aug 2000 11:20:21 +0000 (11:20 +0000)]
Registered modules could only exist in HKEY_LOCAL_MACHINE - now HKEY_CURRENT_USER can override.

24 years ago* added doc strings to urlopen and unquote_plus
Skip Montanaro [Tue, 22 Aug 2000 03:00:52 +0000 (03:00 +0000)]
* added doc strings to urlopen and unquote_plus
* fixed type in doc string for quote

24 years agorequire list comprehensions to start with a for clause
Skip Montanaro [Tue, 22 Aug 2000 02:43:07 +0000 (02:43 +0000)]
require list comprehensions to start with a for clause

24 years agoPatch by Toby Dickenson: don't die when an error occurs during string
Guido van Rossum [Tue, 22 Aug 2000 02:04:46 +0000 (02:04 +0000)]
Patch by Toby Dickenson: don't die when an error occurs during string
conversion in an exception, but instead display <unprintable %s
object> where %s is the type name.

24 years agoDon't bother to 'mkpath()' the 'dist_dir' -- that's now taken care of
Greg Ward [Tue, 22 Aug 2000 01:49:41 +0000 (01:49 +0000)]
Don't bother to 'mkpath()' the 'dist_dir' -- that's now taken care of
in archive_util.py.

24 years agoEnsure destination directory exists before trying to create a tarball
Greg Ward [Tue, 22 Aug 2000 01:48:54 +0000 (01:48 +0000)]
Ensure destination directory exists before trying to create a tarball
or ZIP file.

24 years agoNuked unused variable.
Tim Peters [Tue, 22 Aug 2000 01:44:16 +0000 (01:44 +0000)]
Nuked unused variable.

24 years agoDon't reference a module named test.test_support. Always use plain
Guido van Rossum [Mon, 21 Aug 2000 22:59:29 +0000 (22:59 +0000)]
Don't reference a module named test.test_support.  Always use plain
test_support.  Also fixed the expected output.

24 years agoAdd a minimal test suite for the parser module.
Fred Drake [Mon, 21 Aug 2000 22:30:53 +0000 (22:30 +0000)]
Add a minimal test suite for the parser module.

24 years agoUpdate to reflect recent grammar changes (list comprehensions, extended
Fred Drake [Mon, 21 Aug 2000 22:24:43 +0000 (22:24 +0000)]
Update to reflect recent grammar changes (list comprehensions, extended
print statement), and fix up the extended call syntax support.

Minor stylistic cleanups.

24 years agoDenis S. Otkidach <den@analyt.chem.msu.ru>:
Fred Drake [Mon, 21 Aug 2000 21:47:20 +0000 (21:47 +0000)]
Denis S. Otkidach <den@analyt.chem.msu.ru>:
Let UserString.translate() method work with unicode data.

This closes SourceForge patch #101246.

24 years agoPatch from Paul Schreiber <paul@commerceflow.com>:
Fred Drake [Mon, 21 Aug 2000 21:42:42 +0000 (21:42 +0000)]
Patch from Paul Schreiber <paul@commerceflow.com>:

Patch description
-----------------
This addresses four issues:

(1) usernames and passwords in urls with special characters are now
    decoded properly. i.e. http://foo%2C:bar@www.whatever.com/

(2) Basic Auth support has been added to HTTPS, like it was in HTTP.

(3) Version 1.92 sent the POSTed data, but did not deal with errors
    (HTTP responses other than 200) properly. HTTPS now behaves the
    same way HTTP does.

(4) made URL-checking beahve the same way with HTTPS as it does with
    HTTP (changed == to !=).

24 years agoDescribe the new PRINT_ITEM_TO and PRINT_NEWLINE_TO opcodes.
Barry Warsaw [Mon, 21 Aug 2000 17:19:00 +0000 (17:19 +0000)]
Describe the new PRINT_ITEM_TO and PRINT_NEWLINE_TO opcodes.

24 years agoAdd the new PRINT_ITEM_TO and PRINT_NEWLINE_TO opcodes.
Barry Warsaw [Mon, 21 Aug 2000 17:18:40 +0000 (17:18 +0000)]
Add the new PRINT_ITEM_TO and PRINT_NEWLINE_TO opcodes.

24 years agocom_print_stmt(): Guido rightly points out that the stream expression
Barry Warsaw [Mon, 21 Aug 2000 17:07:20 +0000 (17:07 +0000)]
com_print_stmt(): Guido rightly points out that the stream expression
in extended prints should only be evaluated once.  This patch plays
stack games (documented!) to fix this.

24 years agoAdd TestSkipped as another interesting item defined by test_support.
Fred Drake [Mon, 21 Aug 2000 16:55:57 +0000 (16:55 +0000)]
Add TestSkipped as another interesting item defined by test_support.

24 years agoThomas reminds me to bump the MAGIC number for the extended print
Barry Warsaw [Mon, 21 Aug 2000 16:35:06 +0000 (16:35 +0000)]
Thomas reminds me to bump the MAGIC number for the extended print
opcode additions.

24 years agoPEP 214, Extended print Statement, has been accepted by the BDFL.
Barry Warsaw [Mon, 21 Aug 2000 15:46:50 +0000 (15:46 +0000)]
PEP 214, Extended print Statement, has been accepted by the BDFL.

Additional test cases for the extended print form.

24 years agoPEP 214, Extended print Statement, has been accepted by the BDFL.
Barry Warsaw [Mon, 21 Aug 2000 15:45:16 +0000 (15:45 +0000)]
PEP 214, Extended print Statement, has been accepted by the BDFL.

Document the extended print form.  Fred, please double check the
markup!

24 years agoPEP 214, Extended print Statement, has been accepted by the BDFL.
Barry Warsaw [Mon, 21 Aug 2000 15:44:01 +0000 (15:44 +0000)]
PEP 214, Extended print Statement, has been accepted by the BDFL.

eval_code2(): Implement new bytecodes PRINT_ITEM_TO and
PRINT_NEWLINE_TO, as per accepted SF patch #100970.

Also update graminit.c based on related Grammar/Grammar changes.

24 years agoPEP 214, Extended print Statement, has been accepted by the BDFL.
Barry Warsaw [Mon, 21 Aug 2000 15:38:56 +0000 (15:38 +0000)]
PEP 214, Extended print Statement, has been accepted by the BDFL.

com_print_stmt(): Implement recognition of, and byte compilation for,
extended print using new byte codes PRINT_ITEM_TO and
PRINT_NEWLINE_TO.

24 years agoPEP 214, Extended print Statement, has been accepted by the BDFL.
Barry Warsaw [Mon, 21 Aug 2000 15:36:27 +0000 (15:36 +0000)]
PEP 214, Extended print Statement, has been accepted by the BDFL.

This adds the two new opcodes to support this feature: PRINT_ITEM_TO,
PRINT_NEWLINE_TO.

24 years agoPEP 214, Extended print Statement, has been accepted by the BDFL.
Barry Warsaw [Mon, 21 Aug 2000 15:34:33 +0000 (15:34 +0000)]
PEP 214, Extended print Statement, has been accepted by the BDFL.
This change modifies Python's grammar to include the extended print
form.

24 years agoIgnore Grammar/Makefile now that it's a made Makefile.
Thomas Wouters [Mon, 21 Aug 2000 12:25:17 +0000 (12:25 +0000)]
Ignore Grammar/Makefile now that it's a made Makefile.

24 years agoRemove the winreg module from the project. I don't believe any
Tim Peters [Mon, 21 Aug 2000 02:27:22 +0000 (02:27 +0000)]
Remove the winreg module from the project.  I don't believe any
docs changes are needed (only reference to winreg I could find
was in libwinreg.tex, which is documenting _winreg, and merely
mentions that a higher-level winreg module *may* appear someday;
that's still true).

24 years agoAdd alias for old function name -- removing it broke Alexei Gilchrist's cfm
Andrew M. Kuchling [Sun, 20 Aug 2000 23:37:02 +0000 (23:37 +0000)]
Add alias for old function name -- removing it broke Alexei Gilchrist's cfm

24 years agoStrengthen the warning against using SerialCookie and SmartCookie.
Andrew M. Kuchling [Sun, 20 Aug 2000 23:33:50 +0000 (23:33 +0000)]
Strengthen the warning against using SerialCookie and SmartCookie.
    (If they're security holes, should they be documented at all?)
Minor rewrites.

24 years ago- identify() enum names.
Jack Jansen [Sun, 20 Aug 2000 21:59:03 +0000 (21:59 +0000)]
- identify() enum names.
- don't identify() enum values.
- make aetools.Error available in the package module.

24 years agoUpdated the applescript documentation for te new framework, and replaced the Eudora...
Jack Jansen [Sun, 20 Aug 2000 21:57:38 +0000 (21:57 +0000)]
Updated the applescript documentation for te new framework, and replaced the Eudora example with one that uses Disk Copy (which everyone running MacOS 8 or
higher should have).

24 years agoAdded a package for the other browser too.
Jack Jansen [Sun, 20 Aug 2000 20:23:58 +0000 (20:23 +0000)]
Added a package for the other browser too.

24 years agoMoved to unsupported: it never lived up to its expectancies, and for the Mac all...
Jack Jansen [Sun, 20 Aug 2000 20:21:38 +0000 (20:21 +0000)]
Moved to unsupported: it never lived up to its expectancies, and for the Mac all the functionality is available in the IDE.

24 years agoMoved to unsupported: it doesn't work anymore and it has outlived its usefulness...
Jack Jansen [Sun, 20 Aug 2000 20:10:39 +0000 (20:10 +0000)]
Moved to unsupported: it doesn't work anymore and it has outlived its usefulness so I don't want to fix it.

24 years agoUpdated for new AppleScript structure and moved to Lib (it's far too useful to lurk...
Jack Jansen [Sun, 20 Aug 2000 20:06:51 +0000 (20:06 +0000)]
Updated for new AppleScript structure and moved to Lib (it's far too useful to lurk in the source folder).

24 years agoFixed again to work with the old scripting interface (so we can test the new one...
Jack Jansen [Sun, 20 Aug 2000 20:02:39 +0000 (20:02 +0000)]
Fixed again to work with the old scripting interface (so we can test the new one actually still works).

24 years agoUpdated for the new applescript interface (which actually makes it so ridiculously...
Jack Jansen [Sun, 20 Aug 2000 19:56:13 +0000 (19:56 +0000)]
Updated for the new applescript interface (which actually makes it so ridiculously simple that its main reason for existence is backward compatability).

24 years agoEnums we cannot find are set to None, and enumsubst understands this (no substitution...
Jack Jansen [Sun, 20 Aug 2000 19:42:52 +0000 (19:42 +0000)]
Enums we cannot find are set to None, and enumsubst understands this (no substitution done). This is need for what I think are bugs in the Finder aete resources (some events use unknown enums).

24 years agoFixed event inheritance code (which didn't work at all).
Jack Jansen [Sun, 20 Aug 2000 19:30:56 +0000 (19:30 +0000)]
Fixed event inheritance code (which didn't work at all).
Do two passes over the suites so we can figure out dependencies for enums and such. The manual method was getting too cumbersome for Finder suites.

24 years agoFinder suite.
Jack Jansen [Sun, 20 Aug 2000 19:29:03 +0000 (19:29 +0000)]
Finder suite.

24 years agoRegenerated with fixes to gensuitemodule.
Jack Jansen [Sun, 20 Aug 2000 19:28:27 +0000 (19:28 +0000)]
Regenerated with fixes to gensuitemodule.

24 years agoFix the bug Sjoerd Mullender discovered, where find_from_args() wasn't
Thomas Wouters [Sun, 20 Aug 2000 14:01:53 +0000 (14:01 +0000)]
Fix the bug Sjoerd Mullender discovered, where find_from_args() wasn't
trying hard enough to find out what the arguments to an import were. There
is no test-case for this bug, yet, but this is what it looked like:

from encodings import cp1006, cp1026
ImportError: cannot import name cp1026

'__import__' was called with only the first name in the 'arguments' list.

24 years agoChanged the popen2.py _test function to use the "more" cmd when
Tim Peters [Sun, 20 Aug 2000 05:57:36 +0000 (05:57 +0000)]
Changed the popen2.py _test function to use the "more" cmd when
os.name == "nt".  This makes test_popen2 pass under Win98SE.
HOWEVER, the Win98 "more" invents a leading newline out
of thin air, and I'm not sure that the other Windows flavors
of "more" also do that.
So, somebody please try under other Windows flavors!

24 years agoDavid Goodger's new getopt test module (thanks, David!).
Tim Peters [Sun, 20 Aug 2000 04:18:40 +0000 (04:18 +0000)]
David Goodger's new getopt test module (thanks, David!).
https://sourceforge.net/patch/?func=detailpatch&patch_id=101110&group_id=5470
Accepted as-is, except for purging an "import *".

24 years agoDisallow "import mod.submod as m", because the result is ambiguous. Does it
Thomas Wouters [Sat, 19 Aug 2000 20:55:02 +0000 (20:55 +0000)]
Disallow "import mod.submod as m", because the result is ambiguous. Does it
load mod.submod as m, or mod as m ? Both can be achieved differently, and
unambiguously. Also attempt to document this restriction (editor
appreciated!)

Note that this is an artificial check during compile, because incorporating
this in the grammar is hard, and then adjusting the compiler to do the right
thing with the right nodes is harder.

24 years agoAdd entry for Cookie module.
Fred Drake [Sat, 19 Aug 2000 16:55:31 +0000 (16:55 +0000)]
Add entry for Cookie module.

24 years agoMake it format.
Fred Drake [Sat, 19 Aug 2000 16:54:57 +0000 (16:54 +0000)]
Make it format.
Adjust markup for hyperlinking to the relevant RFCs.
Give it a little organization.
Minor nits.

24 years agoAdding tests of the "attrs" optional argument, and of the js_output
Moshe Zadka [Sat, 19 Aug 2000 15:57:33 +0000 (15:57 +0000)]
Adding tests of the "attrs" optional argument, and of the js_output
functionality.

24 years agoUse METH_VARARGS constant in example module.
Andrew M. Kuchling [Sat, 19 Aug 2000 15:36:41 +0000 (15:36 +0000)]
Use METH_VARARGS constant in example module.
Fix comment typo

24 years agoTest case for Cookie.py
Andrew M. Kuchling [Sat, 19 Aug 2000 15:21:12 +0000 (15:21 +0000)]
Test case for Cookie.py

24 years agoInitial revision. Markup unchecked.
Moshe Zadka [Sat, 19 Aug 2000 14:11:41 +0000 (14:11 +0000)]
Initial revision. Markup unchecked.

24 years agoAdded Tim O'Malley's Cookie.py module (master version at
Andrew M. Kuchling [Sat, 19 Aug 2000 13:01:19 +0000 (13:01 +0000)]
Added Tim O'Malley's Cookie.py module (master version at
http://www.timo-tasi.org/python/Cookie.py)
This is revision 2.26 according to Tim's RCS history.

24 years agocom_error(): Quiet gcc -Wall warning.
Barry Warsaw [Fri, 18 Aug 2000 19:59:20 +0000 (19:59 +0000)]
com_error(): Quiet gcc -Wall warning.

24 years agoRemove a couple of warnings turned up by "gcc -Wall".
Fred Drake [Fri, 18 Aug 2000 19:53:25 +0000 (19:53 +0000)]
Remove a couple of warnings turned up by "gcc -Wall".

24 years agoPyUnicode_AsUTF8String(): /F picks up what I missed: the local var
Barry Warsaw [Fri, 18 Aug 2000 19:30:40 +0000 (19:30 +0000)]
PyUnicode_AsUTF8String(): /F picks up what I missed: the local var
`str' is no longer necessary.  Gotta turn on -Wall!

24 years agoDo not set a MemoryError exception over another MemoryError exception,
Vladimir Marangozov [Fri, 18 Aug 2000 18:01:06 +0000 (18:01 +0000)]
Do not set a MemoryError exception over another MemoryError exception,
thus preserving the first one that has been raised.

24 years agoBetter conformance to the Python Style Guide: use spaces around operators.
Fred Drake [Fri, 18 Aug 2000 16:09:56 +0000 (16:09 +0000)]
Better conformance to the Python Style Guide: use spaces around operators.

24 years agoWhen a KeyboardInterrupt is caught, just use the "raise" syntax to
Fred Drake [Fri, 18 Aug 2000 16:04:05 +0000 (16:04 +0000)]
When a KeyboardInterrupt is caught, just use the "raise" syntax to
re-raise it instead of re-raising it "manually" the ugly way.

24 years agoRevise to use atexit instead of monkeying with sys.exitfunc directly.
Fred Drake [Fri, 18 Aug 2000 15:50:54 +0000 (15:50 +0000)]
Revise to use atexit instead of monkeying with sys.exitfunc directly.

24 years agoConvert some old-style string exceptions to class exceptions.
Fred Drake [Fri, 18 Aug 2000 14:59:33 +0000 (14:59 +0000)]
Convert some old-style string exceptions to class exceptions.