Commit Graph

34272 Commits

Author SHA1 Message Date
Guido van Rossum 6374bb5a69 Shut up compiler warnings. 2007-06-13 16:28:25 +00:00
Guido van Rossum fa331631d4 Discard unused function. 2007-06-13 16:27:47 +00:00
Guido van Rossum ff0ffb3e4a Make test_tmpfile() pass. (And hence test_os.py as a whole passes.)
tmpfile() now is a binary file.
2007-06-13 01:55:50 +00:00
Guido van Rossum 88e860c005 Fix bz2_test.py by removing the tests for universal newline mode.
If you want text support, wrap a TextIOWrapper around it.
Remove references to universal newlines from the BZ2File docstring.
2007-06-13 01:46:31 +00:00
Guido van Rossum 755114a22a Make it compile with GCC 2.96. 2007-06-13 01:04:27 +00:00
Guido van Rossum f09ca140ac Checkpoint: half-fixed the bz2 module. 'U' is no longer supported. 2007-06-13 00:03:05 +00:00
Guido van Rossum da5b8f2d28 Rip out the file object's implementation.
Fixed test_import.py while I was at it.

However, there's still a problem in import.c -- get_file() can leak a
FILE struct (not a file descriptor though).  I'm not sure how to fix
this; closing the FILE* closes the file descriptor, and that's the
wrong thing to do when there's still a Python file object keeping the
file descriptor open.  I also would rather not mess with dup(), as it
won't port to Windows.
2007-06-12 23:30:11 +00:00
Walter Dörwald 2d5c219fe0 unicode is named str now => fix type check. 2007-06-12 18:07:38 +00:00
Walter Dörwald 2c849f2f20 Fix test_shlex: Use io.StringIO. 2007-06-12 17:43:43 +00:00
Walter Dörwald 26e0f51280 Mention name of left operand, if "foo in unicode_string"
fails.
2007-06-12 16:51:31 +00:00
Walter Dörwald 42748a8d6d Rip out all codecs that can't work in a unicode/bytes world:
base64, uu, zlib, rot_13, hex, quopri, bz2, string_escape.

However codecs.escape_encode() and codecs.escape_decode()
still exist, as they are used for pickling str8 objects
(so those two functions can go, when the str8 type is removed).
2007-06-12 16:40:17 +00:00
Walter Dörwald 6e8fcae38f Make module docstrings unicode objects. 2007-06-12 15:23:50 +00:00
Martin v. Löwis 427dbff8f0 Revert 55876. Use PyUnicode_AsEncodedString instead. 2007-06-12 05:53:00 +00:00
Guido van Rossum 543c09e700 Duh. Delete the outdated comment too. 2007-06-12 00:41:35 +00:00
Guido van Rossum 616c8c1e80 Seems this test is just fine. It fails because __phello__.spam
doesn't exist, but that seems a separate issue.
2007-06-12 00:41:02 +00:00
Guido van Rossum 5397039504 Minimal changes to make the "freeze" tool work again.
There are other issues left, but these were basics (e.g. keys().sort()).
2007-06-12 00:28:30 +00:00
Guido van Rossum 0aa35f8709 Update the frozen bytecode for __hello__. 2007-06-12 00:25:52 +00:00
Guido van Rossum d59cde817d Some quick fixes of code that was sorting dict.keys() etc. 2007-06-12 00:25:08 +00:00
Walter Dörwald 1a0270f64a Remove unused code. 2007-06-11 21:48:58 +00:00
Walter Dörwald 5de48bdd19 Simplify various spots where: str() is called on something
that already is a string or the existence of the str class
is checked or a check is done for str twice. These all stem
from the initial unicode->str replacement.
2007-06-11 21:38:39 +00:00
Walter Dörwald 80bfb725af Add versionadded notes to PyUnicode_FromString(),
PyUnicode_FromFormat() and PyUnicode_FromFormatV().
2007-06-11 16:44:48 +00:00
Walter Dörwald c0aa45fca1 Document PyUnicode_FromFormat(). 2007-06-11 16:43:18 +00:00
Walter Dörwald d7fb7644da Add a format specifier %V to PyUnicode_FromFormat(), that works similar to %U,
but requires an additional char * that will be used if the unicode object is
NULL.

Use %V in descrobject.c and classobject.c.
2007-06-11 16:36:59 +00:00
Walter Dörwald 3ef72bb0a9 Fix PyErr_Format() call (missing exception class).
Remove unused variable.
2007-06-11 16:12:10 +00:00
Walter Dörwald e42109d79c Use PyErr_Format() directly instead of
PyOS_snprintf()+PyErr_SetString().
2007-06-11 16:08:41 +00:00
Walter Dörwald 4ddbdad8e7 Simplify error formatting. Fix error message in
check_unoptimized().
2007-06-11 16:06:26 +00:00
Walter Dörwald 5d4ede1811 Simplify ste_repr(). 2007-06-11 16:03:16 +00:00
Walter Dörwald 75163600de Simplify error formatting and type_repr(). 2007-06-11 15:47:13 +00:00
Walter Dörwald d376dd9700 Simplify error formatting. 2007-06-11 15:37:20 +00:00
Walter Dörwald a29d1d7ff5 Simplify error formatting (no default encoding required). 2007-06-11 15:00:18 +00:00
Walter Dörwald 7815c5e91e Check unicode identifier directly instead of converting
it to an 8bit string first.
2007-06-11 14:55:19 +00:00
Walter Dörwald 4dbd01b435 __module__ is a unicode string now:
use PyUnicode_CompareWithASCIIString() instead of
strcmp().

Simplify repr formatting.
2007-06-11 14:03:45 +00:00
Martin v. Löwis 641d5cc6a6 Short-cut lookup of utf-8 codec, to make import work
on OSX.
2007-06-11 04:19:13 +00:00
Martin v. Löwis 9b9905b2f4 Expect unicode in class_name. 2007-06-10 21:13:34 +00:00
Martin v. Löwis 826b9ddbe3 Fix getclassname. Fixes test_descrtut. 2007-06-10 21:09:40 +00:00
Martin v. Löwis 5b222135f8 Make identifiers str (not str8) objects throughout.
This affects the parser, various object implementations,
and all places that put identifiers into C string literals.

In testing, a number of crashes occurred as code would
fail when the recursion limit was reached (such as the
Unicode interning dictionary having key/value pairs where
key is not value). To solve these, I added an overflowed
flag, which allows for 50 more recursions after the
limit was reached and the exception was raised, and
a recursion_critical flag, which indicates that recursion
absolutely must be allowed, i.e. that a certain call
must not cause a stack overflow exception.

There are still some places where both str and str8 are
accepted as identifiers; these should eventually be
removed.
2007-06-10 09:51:05 +00:00
Walter Dörwald 38e43c25ee Adapt to new exception message.
Simplify formatting (use "%r" % x instead of "%s" % repr(x)).
2007-06-09 16:13:23 +00:00
Walter Dörwald 3f1e65cfde Fix Cookie.py: Fix example in the docstring (encoded SerialCookies contain
unicode now). Fix _quote() and Morsel.set() which were using str8.translate().
As cPickle.dumps() returns bytes now value_encode() and value_decode() methods
must encode/decode (however output() might better return a bytes object).
2007-06-08 15:33:46 +00:00
Walter Dörwald 9b775535f8 Rename checks for test_support.have_unicode (we always
have unicode support now) and either drop the tests or
merge them into the existing tests.
2007-06-08 14:30:53 +00:00
Walter Dörwald 0157ebe999 Fix chr() test (1000000 was accepted on a UCS4 build). 2007-06-08 10:38:38 +00:00
Guido van Rossum 2bf7138bb7 Make test_socket work.
Don't exclude test_socket from the tests to run.
2007-06-08 00:07:57 +00:00
Guido van Rossum 7cad4f3d02 Just enoug fixes so that test_mailbox fails instead of loops forever. 2007-06-07 23:58:54 +00:00
Guido van Rossum 48fc58ad31 Accellerate binary readline() a bit. 2007-06-07 23:45:37 +00:00
Guido van Rossum 1325790b93 Merged revisions 55795-55816 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/branches/p3yk

........
  r55797 | neal.norwitz | 2007-06-07 00:00:57 -0700 (Thu, 07 Jun 2007) | 3 lines

  Get rid of some remnants of classic classes.  types.ClassType == type.
  Also get rid of almost all uses of the types module and use the builtin name.
........
  r55798 | neal.norwitz | 2007-06-07 00:12:36 -0700 (Thu, 07 Jun 2007) | 1 line

  Remove a use of types, verify commit hook works
........
  r55809 | guido.van.rossum | 2007-06-07 11:11:29 -0700 (Thu, 07 Jun 2007) | 2 lines

  Fix syntax error introduced by Neal in last checkin.
........
2007-06-07 23:15:56 +00:00
Alexandre Vassalotti 7b955bd125 Fix pdb help command. 2007-06-07 22:37:45 +00:00
Guido van Rossum 46a05a7db5 The bufsize argument to Popen() should accept None meaning the default (0). 2007-06-07 21:56:45 +00:00
Guido van Rossum ad5b9de288 Change normalize_encodings() to avoid using .translate() or depending on
the string type.  It will always return a Unicode string.  The algoritm's
specification is unchanged.
2007-06-07 21:43:46 +00:00
Walter Dörwald c3b6ac796f Fix libc_ver(): libc_ver() was reading sys.executable
in binary mode and comparing the content to strings,
which failed. Now the bytes get decoded into unicode
using latin-1 (the comparison compares ASCII strings
only anyway, and we don't want the decoding to fail).
2007-06-07 19:26:24 +00:00
Walter Dörwald 94093ffcb6 Change most sys attributes that were str8 objects into str objects
(executable, prefix and exec_prefix are still str8).
2007-06-07 18:41:59 +00:00
Guido van Rossum 3d1d712afc Don't lie in error messages from str8.
In some cases we use the more neutral term string; in others we are explicit.
2007-06-07 17:54:36 +00:00