Fred Drake [Mon, 21 Feb 2000 17:59:48 +0000 (17:59 +0000)]
Remove comment that Guido agree's doesn't make sense:
PyEval_EvalCode() is *not* a "backward compatible interface", it's the
one to use!
Fred Drake [Mon, 21 Feb 2000 17:20:16 +0000 (17:20 +0000)]
Script to help locate markup that LaTeX2HTML has a problem with.
Guido van Rossum [Mon, 21 Feb 2000 16:50:31 +0000 (16:50 +0000)]
Mark pointed out a buglet in his patch: i < _sys_nerr isn't strong
enough, it could be negative. Add i > 0 test. (Not i >= 0; zero isn't
a valid error number.)
Jack Jansen [Mon, 21 Feb 2000 11:07:37 +0000 (11:07 +0000)]
Added FreeMem, MaxBlock and CompactMem calls. The values returned by these are lower bounds in the Python case (as malloc doesn't return memory to the heap) but they can be used to decide when to give low-memory warnings.
Greg Stein [Sat, 19 Feb 2000 13:36:23 +0000 (13:36 +0000)]
add TODO section
Andrew M. Kuchling [Fri, 18 Feb 2000 19:16:45 +0000 (19:16 +0000)]
Patch from Vladimir Marangozov <marangoz@python.inrialpes.fr>
The same problem (mixed mallocs) exists for the pcre stack.
The buffers md->... are allocated via PyMem_RESIZE in grow_stack(),
while in free_stack() they are released with free() instead of
PyMem_DEL().
Andrew M. Kuchling [Fri, 18 Feb 2000 18:30:01 +0000 (18:30 +0000)]
Patch from Vladimir Marangozov <marangoz@python.inrialpes.fr>:
The buffers self->regex and self->regex_extra are allocated in
pcre_compile() and pcre_study() via pcre_malloc, but are released
via free() instead of pcre_free.
Greg Stein [Fri, 18 Feb 2000 13:04:10 +0000 (13:04 +0000)]
reduce fsimp coupling: provide fs_imp param to ImportManager constructor,
add clsFilesystemImporter class attribute, alter handling of suffix list
convert suffix importers to funcs rather than instances
remove backwards compat code: Importer.install and 2-tuple get_code()
result values
Greg Stein [Fri, 18 Feb 2000 12:03:40 +0000 (12:03 +0000)]
shift demo importers to importers.py (destined for Demo/ ?)
switch to isinstance() rather than direct type comparisons
removing chaining concept
update ImportManager.install() to take an optional namespace to install
itself in. this will be useful for setting up rexec environments.
minor comment nits
Greg Ward [Fri, 18 Feb 2000 00:36:20 +0000 (00:36 +0000)]
Changed all references to command methods 'set_default_options()' and
'set_final_options()' to 'initialize_options()' and 'finalize_options()'.
Greg Ward [Fri, 18 Feb 2000 00:35:22 +0000 (00:35 +0000)]
Renamed 'set_default_options()' to 'initialize_options()', and
'set_final_options()' to 'finalize_options()'.
Greg Ward [Fri, 18 Feb 2000 00:26:23 +0000 (00:26 +0000)]
Changed references to the command class 'options' attribute to 'user_options'.
Related docstring changes.
Unrelated comment changes.
Greg Ward [Fri, 18 Feb 2000 00:25:39 +0000 (00:25 +0000)]
Renamed all 'options' class attributes to 'user_options'.
Greg Ward [Fri, 18 Feb 2000 00:14:21 +0000 (00:14 +0000)]
Command classes are now named identically to their commands, so reflect this
in 'find_command_class()' method.
Greg Ward [Fri, 18 Feb 2000 00:13:53 +0000 (00:13 +0000)]
Renamed all command classes so they're exactly the same as the name of the
command itself: no more of this "FooBar class for foo_bar command"
silliness.
Greg Ward [Fri, 18 Feb 2000 00:11:52 +0000 (00:11 +0000)]
Changed 'dist' to 'sdist'.
Greg Ward [Thu, 17 Feb 2000 23:56:15 +0000 (23:56 +0000)]
The 'sdist' command to create a source distribution. This is derived from the
old 'dist' command, but the code for dealing with manifests is completely
redone -- and renaming the command to 'sdist' is more symmetric with the
soon-to-exist 'bdist' command.
Greg Ward [Thu, 17 Feb 2000 23:54:55 +0000 (23:54 +0000)]
The 'dist' command is dead -- long live the 'sdist' command!
Jeremy Hylton [Thu, 17 Feb 2000 22:58:54 +0000 (22:58 +0000)]
changes to _lookupName
- removed now (happily) unused second arg
- need to verify results of [].index are correct; for building consts,
need to have same value and same type, e.g. 2 not the same as 2L
Jeremy Hylton [Thu, 17 Feb 2000 22:09:35 +0000 (22:09 +0000)]
the previous quick hack to fix def foo((x,y)) failed on some cases
(big surprise). new solution is a little less hackish.
Code gen adds a TupleArg instance in the argument slot. The tuple arg
includes a copy of the names that it is responsble for binding. The
PyAssembler uses this information to calculate the correct argcount.
all fix this wacky case: del (a, ((b,), c)), d
which is the same as: del a, b, c, d
(Can't wait for Guido to tell me why.)
solution uses findOp which walks a tree to find out whether it
contains OP_ASSIGN or OP_DELETE or ...
Jeremy Hylton [Thu, 17 Feb 2000 22:06:20 +0000 (22:06 +0000)]
add varargs and kwargs flags to Lambda nodes
Jeremy Hylton [Thu, 17 Feb 2000 17:56:29 +0000 (17:56 +0000)]
satisfy the tabnanny
fix broken references to filename var in generateXXX methods
Fred Drake [Thu, 17 Feb 2000 17:45:52 +0000 (17:45 +0000)]
Update the description of int() to include the radix parameter;
omission noted on c.l.py by Aahz Maruch.
Swapped the order of the descriptions of int() and intern() so that
int() comes first (the functions are in alphabetic order).
Fred Drake [Thu, 17 Feb 2000 17:30:40 +0000 (17:30 +0000)]
Typo in a comment: "wheter" --> "whether"
Guido van Rossum [Thu, 17 Feb 2000 17:12:39 +0000 (17:12 +0000)]
Patches by Piers Lauder.
Reasons for patches:
1st patch (15,21):
version change
2nd patch (66,72):
This is a patch I found in a Zope product release (quite by accident!).
It relaxes the conditions for matching a literal. I've looked over the
logic, and tested it, and it seems sensible.
3rd patch (117,123):
It appears the quoting matcher was too general, and that the IMAP4
protocol requires characters like ':' in commands to be unquoted.
(This is the patch already sent to Guido.)
4th patch (699,705):
Spelling correction in comment.
5th patch (753,761):
Another patch from the Zope product. It seems that some IMAP4 servers
produce unexpected responses in the middle of valid command/response
sequences. This patch ignores the unexpected responses in this
situation. (How I wish users would send me bug reports with examples!).
last 2 patches: (1015,1028) (1038,1044):
Minor improvements to test code.
Guido van Rossum [Thu, 17 Feb 2000 16:14:16 +0000 (16:14 +0000)]
Add primitive customization of window size and font.
A few alternative selections can be made by changing "if 0" to "if 1".
Guido van Rossum [Thu, 17 Feb 2000 15:19:15 +0000 (15:19 +0000)]
Patch by Mark Hammond:
* Changes to a recent patch by Chris Tismer to errors.c. Chris' patch
always used FormatMessage() to get the error message passing the error code
from errno - but errno and FormatMessage use a different numbering scheme.
The main reason the patch looked OK was that ENOFILE==ERROR_FILE_NOT_FOUND -
but that is about the only shared error code :-). The MS CRT docs tell you
to use _sys_errlist()/_sys_nerr. My patch does also this, and adds a very
similar function specifically for win32 error codes.
Guido van Rossum [Thu, 17 Feb 2000 15:18:10 +0000 (15:18 +0000)]
Changes by Mark Hammond related to the new WindowsError exception.
Guido van Rossum [Thu, 17 Feb 2000 15:17:18 +0000 (15:17 +0000)]
Declarations related to new WindowsError exception.
Guido van Rossum [Thu, 17 Feb 2000 15:12:01 +0000 (15:12 +0000)]
Added WindowsError, for Mark Hammond's extensions.
Guido van Rossum [Wed, 16 Feb 2000 21:13:37 +0000 (21:13 +0000)]
Added docs for new crc32() function. By Jim Ahlstrom.
(Fred, please check.)
Guido van Rossum [Wed, 16 Feb 2000 21:13:06 +0000 (21:13 +0000)]
Added test for new crc32() function.
Guido van Rossum [Wed, 16 Feb 2000 21:11:52 +0000 (21:11 +0000)]
Patch by Jim Ahlstrom to add crc32, a useful checksum function
(e.g. used for ZIP files).
The patch includes code that says:
+ Copyright (C) 1986 Gary S. Brown. You may use this program, or
+ code or tables extracted from it, as desired without restriction.
My interpretation (and Jim's) is that Gary S Brown has no claims under
copyright, patent or other rights or interests. Lawyers might disagree.
Guido van Rossum [Wed, 16 Feb 2000 01:22:35 +0000 (01:22 +0000)]
The 0.5 release happened on 2/15, not on 2/14. :-)
Jeremy Hylton [Wed, 16 Feb 2000 00:55:44 +0000 (00:55 +0000)]
support for arglists with implicit tuple unpacks
- added a number of support methods to generate code just before the
body
- hack protocol for communicating number of args to PyAssembler
fix TryExcept generation for case where exception handler has no body
fix visitAssAttr
add comment about incomplete visitAssName
stop using the ExampleASTVisitor
change script invocation to accept a list of .py files (e.g. Lib/*.py)
Jeremy Hylton [Wed, 16 Feb 2000 00:51:37 +0000 (00:51 +0000)]
more robust assignment of lineno for keyword args
get the lineno from the name of the keyword arg
example of case that didn't work--
def foo(x, y, a = None,
b = None):
Jeremy Hylton [Wed, 16 Feb 2000 00:50:29 +0000 (00:50 +0000)]
fix argcount generation for arg lists containing tuple unpacks
this is sort of a hack
Jeremy Hylton [Wed, 16 Feb 2000 00:49:47 +0000 (00:49 +0000)]
add flatten helper function
Jeremy Hylton [Tue, 15 Feb 2000 23:45:26 +0000 (23:45 +0000)]
finish first impl of code generator
add support for nodes TryExcept, TryFinally, Sliceobj
fix visitSubscript to properly handle x[a,b,c]
Jeremy Hylton [Tue, 15 Feb 2000 23:43:19 +0000 (23:43 +0000)]
fix creation of Ellipsis node
Jeremy Hylton [Tue, 15 Feb 2000 21:59:50 +0000 (21:59 +0000)]
add a little debugging support when new.code raises SystemError
Jeremy Hylton [Tue, 15 Feb 2000 21:57:14 +0000 (21:57 +0000)]
no real change -- testing syncmail
Jeremy Hylton [Tue, 15 Feb 2000 21:30:48 +0000 (21:30 +0000)]
add line numbers to nodes in the except clause (when possible)
Jeremy Hylton [Tue, 15 Feb 2000 21:30:00 +0000 (21:30 +0000)]
tidy up tryexcept and tryfinally nodes
Guido van Rossum [Tue, 15 Feb 2000 19:11:26 +0000 (19:11 +0000)]
A bit restructured.
Guido van Rossum [Tue, 15 Feb 2000 18:30:58 +0000 (18:30 +0000)]
Oops, somehow the initial checkin was botched. :-(
Guido van Rossum [Tue, 15 Feb 2000 18:20:28 +0000 (18:20 +0000)]
Added some clarifications.
Guido van Rossum [Tue, 15 Feb 2000 18:20:01 +0000 (18:20 +0000)]
Temporarily add a copy here for easy distribution.
Guido van Rossum [Tue, 15 Feb 2000 18:11:21 +0000 (18:11 +0000)]
More changes.
Guido van Rossum [Tue, 15 Feb 2000 18:08:19 +0000 (18:08 +0000)]
Notice status back and stack viewer.
Guido van Rossum [Tue, 15 Feb 2000 18:05:15 +0000 (18:05 +0000)]
Support for Moshe's status bar.
Guido van Rossum [Tue, 15 Feb 2000 18:04:52 +0000 (18:04 +0000)]
Status bar code -- by Moshe Zadka.
Guido van Rossum [Tue, 15 Feb 2000 18:04:09 +0000 (18:04 +0000)]
Adding the old stack viewer implementation back, for the debugger.
Guido van Rossum [Tue, 15 Feb 2000 18:03:40 +0000 (18:03 +0000)]
New stack viewer, uses a tree widget.
(XXX: the debugger doesn't yet use this.)
Guido van Rossum [Tue, 15 Feb 2000 18:03:01 +0000 (18:03 +0000)]
Correct a typo and remove an unqualified except that was hiding the error.
Guido van Rossum [Tue, 15 Feb 2000 18:02:11 +0000 (18:02 +0000)]
Add an XXX comment about the ClassBrowser AIP.
Guido van Rossum [Tue, 15 Feb 2000 17:19:25 +0000 (17:19 +0000)]
Updated change log.
Guido van Rossum [Tue, 15 Feb 2000 17:17:58 +0000 (17:17 +0000)]
News update. Probably incomplete; what else is new?
Guido van Rossum [Tue, 15 Feb 2000 17:16:40 +0000 (17:16 +0000)]
Updated for pending IDLE 0.5 release (still very rough -- just getting
it out in a more convenient format than CVS).
Guido van Rossum [Tue, 15 Feb 2000 17:15:36 +0000 (17:15 +0000)]
Tiny addition.
Guido van Rossum [Tue, 15 Feb 2000 14:51:46 +0000 (14:51 +0000)]
In response to one particular complaint on edu-sig, change some error
messages from "OverflowError: integer pow()" to "OverflowError:
integer exponentiation". (Not that this takes care of the complaint
in general that the error messages could be greatly improved. :-)
Jeremy Hylton [Mon, 14 Feb 2000 23:57:56 +0000 (23:57 +0000)]
edit a doc string
(real intent is to test out rsync install)
Andrew M. Kuchling [Mon, 14 Feb 2000 22:22:04 +0000 (22:22 +0000)]
Make multiplying a sequence by a long integer (5L * 'b') legal
Jeremy Hylton [Mon, 14 Feb 2000 21:54:57 +0000 (21:54 +0000)]
get rid of spurious print
Guido van Rossum [Mon, 14 Feb 2000 21:42:14 +0000 (21:42 +0000)]
Patch by Gerrit Holl:
* In crlf.py and lfcr.py: regsub -> re
Guido van Rossum [Mon, 14 Feb 2000 21:41:50 +0000 (21:41 +0000)]
Patch by Gerrit Holl:
* In logmerge.py: added '-r' flag to show the oldest checkin
first instead of the newest, and getopt.getopt was used
wrong.
Jeremy Hylton [Mon, 14 Feb 2000 21:33:10 +0000 (21:33 +0000)]
looks like everything is working except for try/except (pystone
compiles correctly)
Jeremy Hylton [Mon, 14 Feb 2000 21:32:42 +0000 (21:32 +0000)]
LeftShift & RightShift: fix reprs, change attr names to left and right
(so they are common with other binary ops)
Fred Drake [Mon, 14 Feb 2000 21:30:52 +0000 (21:30 +0000)]
Fix the question marks next to the expansions of some of the
colorspace name abbreviations, based on email from Gerrit Holl
<gerrit.holl@pobox.com>.
Jeremy Hylton [Mon, 14 Feb 2000 18:34:11 +0000 (18:34 +0000)]
rename several of the generic attribute names for nodes. new node attrs are:
Exec: expr, locals, globals
Dict: items
Assert: test, fail
Jeremy Hylton [Mon, 14 Feb 2000 18:32:46 +0000 (18:32 +0000)]
(), [], and {} should not be represented as constant expressions, they
should be calls to BUILD_ ops for these types with no arguments
Guido van Rossum [Mon, 14 Feb 2000 17:58:25 +0000 (17:58 +0000)]
Patch by Jack Jansen:
If we attempt to import a dynamic module in a newer (or older) version
of Python give an error message tailored to the situation (Python too
new/old).
Jeremy Hylton [Mon, 14 Feb 2000 14:14:29 +0000 (14:14 +0000)]
split compile.py into two files
add StackDepthFinder (and remove push/pop from CodeGen)
add several nodes, including Ellipsis, Bit&|^, Exec
Jeremy Hylton [Sat, 12 Feb 2000 00:12:38 +0000 (00:12 +0000)]
change MODULE_NAMESPACE/FUNCTION_NAMESPACE stuff to have a single flag
named OPTIMIZED, which matches compile.c and makes more sense for
classes
revamp call signature for PythonVMCode.__init__; doesn't really matter
'cuz this code is going to be refactored out of existence
add generateClassCode and modify Func & Lambda generation
add support for nodes Classdef, Keyword,
fix CallFunc to generate right op arg when calling w/ keywords
add ugly hack to properly compute offsets when the same stack ref is
used multiple times
Jack Jansen [Fri, 11 Feb 2000 23:17:14 +0000 (23:17 +0000)]
And added a note about needing two "cvs update"s.
Jack Jansen [Fri, 11 Feb 2000 23:14:46 +0000 (23:14 +0000)]
Explained that you have to checkout the Mac portion of Python in a separate
folder and then move it to the Python hierarchy.
Jeremy Hylton [Fri, 11 Feb 2000 20:27:07 +0000 (20:27 +0000)]
add support for Lambda nodes
change resolution of local name ops (LOAD_FAST). i think it makes
sense now. if it is an argument or a local var name that it used, it
must be in varnames. if it is a local var name that is used, it must
also be in names
Greg Ward [Fri, 11 Feb 2000 02:52:39 +0000 (02:52 +0000)]
Ditched '_find_SET()', since it was a no-value-added wrapper around
'get_msvc_paths()'.
Renamed '_do_SET()' to 'set_path_env_var()', tweaked docstring, and
cosmetically tweaked code.
Stylistic changes to MSVCCompiler constructor (variable renaming
and type consistency).
Greg Ward [Fri, 11 Feb 2000 02:47:15 +0000 (02:47 +0000)]
Latest patch from Thomas Heller/Robin Becker:
* tweak my docstrings
* fix None returns to empty list
* reshuffle responsibilities between '_find_exe()', '_find_SET()', and
the MSVCCompiler constructor -- now the constructor worries about
fetching the version list and determining the most recent one
* added "/W3" compile option
Also, I added/tweaked some docstrings.
Jeremy Hylton [Thu, 10 Feb 2000 20:55:50 +0000 (20:55 +0000)]
add loop handling via
Loop object to handle StackRegs
loops stack attr on CodeGenreeator to hold the current loop object
add support for nodes While, Break, Continue
Jeremy Hylton [Thu, 10 Feb 2000 20:54:27 +0000 (20:54 +0000)]
add an __len__ to Set and Stack
Barry Warsaw [Thu, 10 Feb 2000 20:26:45 +0000 (20:26 +0000)]
Document the API changes to the nntplib module (exceptions become
classes, NNTP.__init__() grows a new optional argument to set reader
mode on the server).
Barry Warsaw [Thu, 10 Feb 2000 20:25:53 +0000 (20:25 +0000)]
Added new exception classes:
NNTPError - derived from Exception, it's the base class for all
other exceptions in this module
NNTPReplyError - what used to be error_reply
NNTPTemporaryError - what used to be error_temp
NNTPPermanentError - what used to be error_perm
NNTPProtocolError - what used to be error_proto
NNTPDataError - what used to be error_data
All the old names are retained for backwards compatibility; they point
to the class that replaces them. Also, any code in this module that
raises an exception, now does so with the exception class.
NNTP.__init__(): Added a new optional argument `readermode', which is
a flag that defaults to false. When set to true, the "mode reader"
command is sent to the NNTP server before user authentication. Reader
mode is sometimes necessary if you are connecting to an NNTP server on
the local machine and intend to call reader-specific comamnds, such as
`group'. If you get unexpected NNTPPermanentErrors, you might need to
set readermode. Patch provided by Thomas Wouters (who include the
standard disclaimer on is patches@python.org submission), and inspired
by Jim Tittsler.
Fred Drake [Thu, 10 Feb 2000 17:23:44 +0000 (17:23 +0000)]
Enable -t when compiling Python library modules in libinstall (.pyc
version only).
Enable -tt for the regression test.
Very similar to Skip's patch.
Jeremy Hylton [Thu, 10 Feb 2000 17:20:39 +0000 (17:20 +0000)]
add namespace attr to CodeGenerator, can be either MODULE_NAMESPACE or
FUNCTION_NAMESPACE. initialize in __init__ and reset in
generateFunctionCode.
replace direct issue of STORE_FAST, STORE_GLOBAL, etc. with call to
storeName; same for loadName and deleteName
the new {store,load,delete}Name methods use the namespace attr and the
local variable stack to determine the correct bytecode to issue
Fred Drake [Thu, 10 Feb 2000 17:17:14 +0000 (17:17 +0000)]
Untabify to pass the -tt test.
Fred Drake [Thu, 10 Feb 2000 16:21:11 +0000 (16:21 +0000)]
Detabify.
I ran "expand" instead of using Skip's patch, but it's all the same.
Jack Jansen [Thu, 10 Feb 2000 16:15:53 +0000 (16:15 +0000)]
In AskYesNoCancel() treat the "yes" and "no" buttons the same way as the cancel
button (i.e. an empty label hides the button).
Fred Drake [Thu, 10 Feb 2000 15:31:07 +0000 (15:31 +0000)]
Make this pass the -tt test.
Greg Ward [Thu, 10 Feb 2000 02:52:42 +0000 (02:52 +0000)]
Stylistic changes to the registry-grovelling code: code formatting, changed
function names, dbetter (hopefully) ocstrings, and comments.
Greg Ward [Thu, 10 Feb 2000 02:51:32 +0000 (02:51 +0000)]
Typecheck 'output_dir' argument to compile/link methods.
Greg Ward [Thu, 10 Feb 2000 02:17:06 +0000 (02:17 +0000)]
Path from Thomas Heller: resurrect the .def file kludge while preserving the
/export option mini-kludge.
Greg Ward [Thu, 10 Feb 2000 02:15:52 +0000 (02:15 +0000)]
Patch from Thomas heller:
* don't need to mention python<ver>.lib -- it's done by a pragma
* add debug flags for compile and link, and use them
* fix 'link_shared_library()' to pass everything to 'link_shared_object()'
* change filename when shared object with debug info (ugh)
Jeremy Hylton [Thu, 10 Feb 2000 00:47:08 +0000 (00:47 +0000)]
add ExampleASTVisitor:
* prints out examples of nodes that are handled by visitor. simply a
development convenience
remove NestedCodeGenerator -- it was bogus after all
replace with generateFunctionCode, a method to call to generate code
for a function instead of a top-level module
fix impl of visitDiscard (most pop stack)
emit lineno for pass
handle the following new node types: Import, From, Getattr, Subscript,
Slice, AssAttr, AssTuple, Mod, Not, And, Or, List
LocalNameFinder: remove names declared as globals for locals
PythonVMCode: pass arg names to constructor, force varnames to contain
them all (even if they aren't referenced)
add -q option on command line to disable stdout
Jeremy Hylton [Thu, 10 Feb 2000 00:43:22 +0000 (00:43 +0000)]
add remove method to set
Greg Ward [Wed, 9 Feb 2000 02:20:14 +0000 (02:20 +0000)]
Added 'debug' option, and changed compile/link calls to use it.
Greg Ward [Wed, 9 Feb 2000 02:19:49 +0000 (02:19 +0000)]
Added 'debug' option (just there for 'build_ext' and 'build_lib' commands
to fallback to if the user doesn't set it for those commands.
Greg Ward [Wed, 9 Feb 2000 02:18:39 +0000 (02:18 +0000)]
Added 'debug' flags to compile and link methods, and added dummy code for
someone who knows Windows/MSVC++ to come along and add the right flags.
Comment noting that 'link_static_lib()' signature is inconsistent with
the other compiler classes (uh-oh!)
Greg Ward [Wed, 9 Feb 2000 02:17:00 +0000 (02:17 +0000)]
Added 'debug' flags to compile and link methods, and modified code to add
'-g' flag to compiler/linker command lines when it's true.
Greg Ward [Wed, 9 Feb 2000 02:16:14 +0000 (02:16 +0000)]
Added 'debug' flag to compile and link method signatures.
Doc fix: several paragraphs under 'link_static_lib()' moved to
'link_shared_lib()', where they belong.