summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Make sure that a forked child process exits even if it raises an exception.Guido van Rossum2007-04-091-4/+6
|
* Fix a place where it's pretty clear that a binary open mode was intended.Guido van Rossum2007-04-091-1/+1
| | | | (The new I/O library revealed the problem.)
* Cleanup.Guido van Rossum2007-04-081-61/+124
| | | | | | | | | Add closed attribute. Support int argument to open() -- wrapping a file descriptor. For b/w compat, support readline(n). Support readlines() and readlines(n). Flush on __del__. Added some XXX comments.
* Make it possible to instantiate a _FileIO() with an integer file descriptorGuido van Rossum2007-04-081-18/+50
| | | | | instead of a filename. Add a 'closed' attribute.
* Somehow this contained a bogus reference to "exceptions" which causedGuido van Rossum2007-04-071-2/+0
| | | | "import exceptions" to behave bizarrely.
* A class used as an exception should inherit from Exception.Guido van Rossum2007-04-071-1/+1
| | | | (Should be backported to 2.6, really.)
* Add some backwards compatibility stuff.Guido van Rossum2007-04-071-4/+29
| | | | | This now appears to work when io.open is substituted for the real open in fileinput.py -- at least the latter's unit tests pass.
* Checkpoint.Guido van Rossum2007-04-062-77/+111
| | | | | Some cleanup of test_io.py and io.py. Added seeking to buffered reader and writer, but no tests yet.
* Get rid of duplicate definition of BufferedIOBase.Guido van Rossum2007-04-061-12/+6
|
* Added a working Text I/O layer, by Mark Russell.Guido van Rossum2007-04-062-47/+366
| | | | This is essentially a checkpoint.
* make the markup match that of the trunk to avoid insane spread ofFred Drake2007-03-291-1/+1
| | | | mystery italics
* Fix refcounting bug reported by Amaury Forgeot d'Arc.Guido van Rossum2007-03-291-1/+2
|
* Make readonly members defined in C throw an AttributeError on modification. ↵Collin Winter2007-03-285-9/+9
| | | | This brings them into sync with Python-level attributes. Fixes bug #1687163.
* More fixes related to removal of exception slicing.Guido van Rossum2007-03-281-9/+9
|
* Fix errors in bsddb3 tests due to removal of exception slicing.Guido van Rossum2007-03-282-11/+11
| | | | | | (There was also a segfault but it disappeared when the tests stopped erroring out; I presume the segfault is a pre-existing problem somewhere in a destructor.)
* Moving xreload to the sandbox for now.Guido van Rossum2007-03-282-293/+0
|
* By default, != returns the opposite of ==, unless the latter returnsGuido van Rossum2007-03-273-1/+25
| | | | | | NotImplemented. (Is this worth backporting to 2.6? It seems so useful...!)
* Looks like this file wasn't checked in.Guido van Rossum2007-03-271-168/+170
|
* Fix two bugs related to the remove of exception subscripting.Collin Winter2007-03-272-2/+2
|
* Address some XXXs from Neal Norwitz.Collin Winter2007-03-231-8/+19
|
* When removing indexing/slicing on exceptions some places were changedBrett Cannon2007-03-212-2/+2
| | | | | | | inappropriately from ``e[0]`` to ``e.message`` instead of ``e.args[0]``. The reason it needs to be the last option is the dichotomy of 'message' and 'args': 'message' can be the empty string but args[0] can have a value if more than one argument was passed.
* Fix refleak in compiler.Guido van Rossum2007-03-211-1/+4
| | | | (A symbol table entry was leaked every time a class was compiled.)
* Remove isCallable() and sequenceIncludes() from the operator module.Collin Winter2007-03-214-21/+4
|
* Little fixes:Neal Norwitz2007-03-212-8/+17
| | | | | | | | * make some module variables static to prevent name pollution * Add some comments to clarify what's going on and some XXXs to address * Add a space after "for" before ( * exc_value and tb can be NULL in some cases * Get working on Windows (I think)
* Patch #1680961: remove sys.exitfunc and replace it with a private C API. ↵Collin Winter2007-03-2113-204/+360
| | | | Also, reimplement atexit in C so it can take advantage of this private API.
* Use the local python binary to build the docs, not whatever 'python' is on ↵Collin Winter2007-03-211-1/+1
| | | | the PATH (which is probably Python 2.x, and we need 3.x).
* Run 2to3 over Doc/tools/.Collin Winter2007-03-2115-64/+66
|
* Make test_socket_ssl.py pass by fixing some code that wasGuido van Rossum2007-03-191-4/+4
| | | | | incorrectly assuming that err.message was the Py3k way of writing err[0] in 2.x. The correct spelling is err.args[0].
* Use input() now that it is available.Guido van Rossum2007-03-191-4/+2
|
* "from ... import x" should not be a syntax error... makeGeorg Brandl2007-03-192-3/+8
| | | | import_stmt accept ELLIPSes and DOTs.
* Fix the compiler package w.r.t. the new metaclass syntax.Guido van Rossum2007-03-197-30/+40
| | | | | | (It is still broken w.r.t. the new nonlocal keyword.) Remove a series of debug prints I accidentally left in test_ast.py.
* NEWS item for PEP 3115.Georg Brandl2007-03-181-0/+5
|
* Refactor astgen.py with 2to3.Georg Brandl2007-03-181-38/+38
|
* Make ELLIPSIS a separate token. This makes it a syntax error to write ". . ↵Georg Brandl2007-03-189-95/+111
| | | | ." for Ellipsis.
* Remove the deprecated and useless "pend" argument fromGeorg Brandl2007-03-186-29/+10
| | | | PyFloat_FromString. (fixes bug #1650903)
* Update the silly version number.Guido van Rossum2007-03-181-2/+2
|
* Implement PEP 3115 -- new metaclass syntax and semantics.Guido van Rossum2007-03-1825-236/+603
| | | | | | | The compiler package hasn't been updated yet; test_compiler.py fails. Otherwise all tests seem to be passing now. There are no occurrences of __metaclass__ left in the standard library. Docs have not been updated.
* Bug 1679498: remove unused instance variables _readable, _writable andGuido van Rossum2007-03-181-4/+0
| | | | _seekable.
* Bug 1679498: unset variable 'bs'.Guido van Rossum2007-03-181-0/+2
|
* Add some XXX comments and fix BufferedReader signature.Guido van Rossum2007-03-151-2/+6
|
* Add Mike Verdone (co-author of io.py and PEP 3116).Guido van Rossum2007-03-151-0/+1
| | | | Daniel Stutzbach (already in) is main author of _fileio.c.
* Check the keys of the locals dict -- they need not be a list.Georg Brandl2007-03-121-1/+12
|
* Patch #1591665: implement the __dir__() special function lookup in PyObject_Dir.Georg Brandl2007-03-105-162/+259
|
* Add a missing forward declaration for PyFileIO_Type. The _fileio module now ↵Collin Winter2007-03-081-0/+2
| | | | compiles with --with-pydebug.
* Move test_unittest, test_doctest and test_doctest2 higher up in the test orderCollin Winter2007-03-081-0/+3
|
* Fix another unittest that broke due to dict views (more precisely, dueGuido van Rossum2007-03-081-1/+1
| | | | to the cowboy way they are currently hacked in).
* Fix embarrassing syntax mix.Guido van Rossum2007-03-081-1/+1
|
* Check in Daniel Stutzbach's _fileio.c and test_fileio.pyGuido van Rossum2007-03-085-1/+1002
| | | | | | (see SF#1671314) with small tweaks. The io module now uses this instead of its own implementation of the FileIO class, if it can import _fileio.
* Delete TESTFN after the test.Guido van Rossum2007-03-071-0/+3
|
* Change the specs for readinto() -- it should *not* shorten the buffer toGuido van Rossum2007-03-072-7/+16
| | | | the amount of data read.
generated by cgit v0.12 at 2024-11-28 16:01:14 (GMT)