Commit Graph

4001 Commits

Author SHA1 Message Date
Jeremy Hylton 29906eef3a Preliminary support for future nested scopes
compile.h: #define NESTED_SCOPES_DEFAULT 0 for Python 2.1
           __future__ feature name: "nested_scopes"

symtable.h: Add st_nested_scopes slot.  Define flags to track exec and
    import star.

Lib/test/test_scope.py: requires nested scopes

compile.c: Fiddle with error messages.

    Reverse the sense of ste_optimized flag on
    PySymtableEntryObjects.  If it is true, there is an optimization
    conflict.

    Modify get_ref_type to respect st_nested_scopes flags.

    Refactor symtable_load_symbols() into several smaller functions,
    which use struct symbol_info to share variables.  In new function
    symtable_update_flags(), raise an error or warning for import * or
    bare exec that conflicts with nested scopes.  Also, modify handle
    for free variables to respect st_nested_scopes flag.

    In symtable_init() assign st_nested_scopes flag to
    NESTED_SCOPES_DEFAULT (defined in compile.h).

    Add preliminary and often incorrect implementation of
    symtable_check_future().

    Add symtable_lookup() helper for future use.
2001-02-27 04:23:34 +00:00
Fred Drake 3c823aa4b6 Make sure ConfigParser uses .optionxform() consistently; this affects
.has_option(), .remove_option(), and .set().

This closes SF tracker #232913.
2001-02-26 21:55:34 +00:00
Tim Peters ffc215a279 Add __future__.py to std library, + dull test to verify that assignments
therein are of the proper form.
2001-02-26 21:14:49 +00:00
Barry Warsaw c1e100f215 Additional tests for current, PEP described semantics:
- func.__dict__ is None until the first attribute is assigned

- del func.__dict__ is equivalent to func.__dict__ = None

- disallowing assignment to function attribute through unbound method
  (it was always illegal to assign through bound method).

- verifying that setting attribute explicitly on underlying function
  via meth.im_func is okay.
2001-02-26 18:07:26 +00:00
Fred Drake 6fd08baddc Do not hide a failure to create a temporary file; if it fails the work
will not have been done, and applications need to know that.  Also, do
not print a message about it; the exception is the right thing.

This closes SF bug #133717.
2001-02-23 20:04:54 +00:00
Barry Warsaw 22e4182d60 Describe -s a little more generically. 2001-02-23 18:31:40 +00:00
Martin v. Löwis 7edbd4ffb4 Patch #103885: Add dynamic registration and lookup of DOM implementations. 2001-02-22 14:05:50 +00:00
Jeremy Hylton f5d3ea00b9 Fix previous checkin, hopelessly broken as it was; reported by Detlef Lannert. 2001-02-22 13:24:27 +00:00
Jeremy Hylton 3a9a96c778 replace exec with simple assignments 2001-02-21 16:33:24 +00:00
Jack Jansen 2d0589be67 The code to write timestamps couldn't handle negative times (and time
on the Mac is negativevalues > 0x80000000). Fixed.
2001-02-21 10:39:35 +00:00
Tim Peters 0009c4ea59 Whitespace normalization. 2001-02-21 07:29:48 +00:00
Andrew M. Kuchling 9a0f98e0a1 Add test case from bug #124981: zlib decompress of sync-flushed data
fails
2001-02-21 02:17:01 +00:00
Andrew M. Kuchling bc8f72cccc Patch #103854: raises an exception if a non-Attr node is passed to
NamedNodeMap.setNamedItem().  Martin, should I sync the PyXML tree, too,
 or do you want to do it?  (I don't know if you're wrapping the 0.6.4
 release right now.)
2001-02-21 01:30:26 +00:00
Andrew M. Kuchling 8b94b1c74a Added test for patch #103473: test an unquoted cookie value containing '=' 2001-02-21 01:17:54 +00:00
Andrew M. Kuchling c05abb3bda Patch #103473 from dougfort: Some sites (amazon.com for one) drop
cookies that contain '=' as part of the value. This patch modifies
Cookie.py to allow '=' as a legal character, and to make the key
search nongreedy so it stops at the first '='.
2001-02-20 22:11:24 +00:00
Tim Peters 754ba589b7 Improve accuracy. In the .tex file, note the new "% BUG:" comments: an
extra backslash is getting displayed in the generated HTML.
2001-02-20 11:24:35 +00:00
Jeremy Hylton 121b6eb018 SF patch #103749: implicit tuple + default arg 2001-02-19 23:53:42 +00:00
Fred Drake f1da6287fc Fix a few small typos in the docstrings.
get_close_matches():  Do not use %-interpolation for strings when
    concatenation is more efficient.
2001-02-19 19:30:05 +00:00
Guido van Rossum b68c245662 SF Patch # 103839 byt dougfort: Allow ';' in attributes
sgmllib does not recognize HTML attributes containing the semicolon
';' character. This may be in accordance with the HTML spec, but there
are sites that use it (excite.com) and the browsers I regularly use
(IE5, Netscape, Opera) all handle it. Doug Fort Downright Software LLC
2001-02-19 18:39:09 +00:00
Jeremy Hylton f828e2d737 Add simple section for assert, including assert w/ lambdas 2001-02-19 15:54:52 +00:00
Jeremy Hylton 4779399e9f Add test for syntax error on "x = 1 + 1".
Move check_syntax() function into test_support.
2001-02-19 15:35:26 +00:00
Jeremy Hylton c348cd7518 fix long line 2001-02-19 15:34:10 +00:00
Fred Drake b6a4425392 DOMException._get_code():
New method; this is the "alternate" access to the exception code.
    (Useful for Python DOM implementations that support the accessor
    method approach to retrieving attribute values.)
2001-02-19 14:57:02 +00:00
Thomas Heller e09f63949d Enhancements to the bdist_wininst command:
--bitmap command line option allows to use a different bitmap file instead
of the build-in python powered logo.
--title lets you specify the text to display on the background.

The editbox in the first screen now longer is
selected (highlighted), it had the WS_TABSTOP flag.

This is the patch
http://sourceforge.net/patch/?func=detailpatch&patch_id=103687&group_id=5470
with two changes:
1. No messagebox displayed when the compilation to .pyc or .pyo files
failes, this will only confuse the user (and it will fail under certain
cases, where sys.path contains garbage).
2. A debugging print statement was removed from bdist_wininst.py.
2001-02-19 09:20:30 +00:00
Marc-André Lemburg 636b90638a This patch makes the default compiler determination more flexible
and also takes the sys.platform name into account. This helps on
platforms where there are multiple possible compiler backends (the
one with which Python itself was compiled is preferred over others
in this case).

The patch uses this new technique to enable using cygwin compiler
per default for cygwin compiled Pythons.

Written by Marc-Andre Lemburg. Copyright assigned to Guido van Rossum.
2001-02-19 09:20:04 +00:00
Fredrik Lundh c0c7ee3a65 detect attempts to repeat anchors (fixes bug #130748) 2001-02-18 21:04:48 +00:00
Tim Peters 8ac3627b91 sre_{parse, compile} no longer define __all__. 2001-02-18 14:44:42 +00:00
Fredrik Lundh f2989b22ff - restored 1.5.2 compatibility (sorry, eric)
- removed __all__ cruft from internal modules (sorry, skip)
- don't assume ASCII for string escapes (sorry, per)
2001-02-18 12:05:16 +00:00
Skip Montanaro 78349072f7 removed __all__ from several modules 2001-02-18 03:30:53 +00:00
Skip Montanaro 1ca2ed35e0 removed __all__ - should probably rename makedict to _makedict unless it is
to be exported
2001-02-18 03:13:08 +00:00
Skip Montanaro 23bafc6fcd add module-level constants to __all__ 2001-02-18 03:10:09 +00:00
Andrew M. Kuchling 27eba5e888 Split the rpath argument into multiple paths, turning it into a list.
This partially fixes bug #128930.
2001-02-17 04:48:41 +00:00
Neil Schemenauer 1a02086885 Linking just got simpiler on AIX and BeOS (closes SF patch #103679). 2001-02-16 03:31:13 +00:00
Tim Peters 0e6d213177 Whitespace normalization. 2001-02-15 23:56:39 +00:00
Skip Montanaro 0de65807e6 bunch more __all__ lists
also modified check_all function to suppress all warnings since they aren't
relevant to what this test is doing (allows quiet checking of regsub, for
instance)
2001-02-15 22:15:14 +00:00
Skip Montanaro c3e11d6569 provide simple recovery/escape from apparent redirect recursion. If the
number of entries into http_error_302 exceeds the value set for the maxtries
attribute (which defaults to 10), the recursion is exited by calling
the http_error_500 method (or if that is not defined, http_error_default).
2001-02-15 16:56:36 +00:00
Guido van Rossum 4ac83474a3 Provide a default for the blocksize arg of storbinary().
SF patch #103517 by mfx.
2001-02-15 13:50:36 +00:00
Andrew M. Kuchling 8e6d44eedc Patch #103748 from Toby Dickenson: fix typo in test_zlib that turns one
test case into a no-op because ''.join('hello world') == 'hello world'
2001-02-14 17:46:20 +00:00
Fred Drake cc1f951b4c Test section name using some strange characters, including a backslash
(SF bug #132288).
2001-02-14 15:30:31 +00:00
Fred Drake d4df94b56d Be much more permissive in what we accept in section names; there has been
at least one addition to the set of accepted characters for every release
since this module was first added; this should take care of the problem
in a more substantial way.

This closes SF bug #132288.
2001-02-14 15:24:17 +00:00
Tim Peters f9bb4969af Miranda newlines: if anything at all was written to stdout, supply a
newline at the end if there isn't one already.  Expected output has no
way to indicate that a trailing newline was not expected, and in the
interpreter shell *Python* supplies the trailing newline before printing
the next prompt.
2001-02-14 06:35:35 +00:00
Tim Peters 60e23f4cfc Change doctest exception example to one whose detail hasn't changed since 1.5.2. 2001-02-14 00:43:21 +00:00
Tim Peters ea4f931cb9 Teach doctest about newer "(most recent call last)" traceback spelling. 2001-02-13 20:54:42 +00:00
Guido van Rossum 467d723bd7 Added a comment explaining why this file must really have #!
/usr/local/bin/python and not #! /usr/bin/env python.
2001-02-13 13:13:33 +00:00
Skip Montanaro 5bba231d1e The bulk of the credit for these changes goes to Bastian Kleineidam
* restores urllib as the file fetcher (closes bug #132000)
* allows checking URLs with empty paths (closes patches #103511 and 103721)
* properly handle user agents with versions (e.g., SpamMeister/1.5)
* added several more tests
2001-02-12 20:58:30 +00:00
Fred Drake 95b96d3941 Added options that use square brackets in their names; this ensures that
GNOME-style internationalized options can be parsed using ConfigParser
(SF bug #131635).

Converted the tests to use test_support.verify() instead of output
comparison to work.
2001-02-12 17:23:20 +00:00
Fred Drake d83bbbfd22 Allow square brackets in the option names; this makes it possible to use
ConfigParser with GNOME-ish config files that use the internationalization
conventions found in GNOME.

This closes SF bug #131635.
2001-02-12 17:18:11 +00:00
Thomas Heller bfae1964c8 This change makes imputil more closely emulate the standard import
mechanism to support self-modifying modules.
2001-02-12 09:17:06 +00:00
Tim Peters 76c066b103 test_pty started failing on Windows, but if and only if test___all__ was
run first.  Indirectly due to Skip adding check_all("pty") to test___all__:
that caused the expected ImportError due to pty.py trying to import the
non-existent FCNTL to get handled by test___all__, leaving a partial
module object for pty in sys.modules, which caused the later import of
pty via test_pty to succeed.  Then test_tpy died with an AttributeError,
due to trying to access attributes of pty that didn't exist.  regrtest
viewed that as a failure rather than the appropriate "test skipped".
Fixed by deleting partial module objects in test___all__ when test___all__
handles an ImportError.
2001-02-12 03:27:31 +00:00
Skip Montanaro c62c81e013 __all__ for several more modules 2001-02-12 02:00:42 +00:00