Thomas Wouters [Sat, 12 Aug 2000 20:32:46 +0000 (20:32 +0000)]
Bring Tools/compiler almost up to date. Specifically:
- fix tab space issues (SF patch #101167 by Neil Schemenauer)
- fix co_flags for classes to include CO_NEWLOCALS (SF patch #101145 by Neil)
- fix for merger of UNPACK_LIST and UNPACK_TUPLE into UNPACK_SEQUENCE,
(SF patch #101168 by, well, Neil :)
- Adjust bytecode MAGIC to current bytecode.
Trent Mick [Sat, 12 Aug 2000 19:37:27 +0000 (19:37 +0000)]
Add the current Win64 compiler to the list of those that need the
huge switch statement broken up. This will probably not be necessary when
the Win64 compiler matures.
Guido van Rossum [Sat, 12 Aug 2000 14:45:50 +0000 (14:45 +0000)]
Whoever added the Makefile rule to auto-build the Grammar didn't have
his build directory in a different place than his source directory. I
do, and it is supposed to be supported. The naive patch caused an
endless recursion in the Make process. This should take care of that.
Trent Mick [Fri, 11 Aug 2000 23:20:32 +0000 (23:20 +0000)]
These files all used to be tagged as binary in CVS. I recently changed this so
that these files are treated as normal text files (which they are). However,
the files also had to be changed to be stored in CVS internally with UNIX line
terminators (they had DOS line terminators internally before this commit).
Thomas Wouters [Fri, 11 Aug 2000 22:26:35 +0000 (22:26 +0000)]
'make' in Grammar too, so graminit.h and graminit.c get re-build when
necessary. Do Grammar after Parser because Grammar needs Parser, and not the
other way 'round. This patch doesn't bother with dependencies because it's
tricky to get right (for instance for the modules that want graminit.h,
like cPickle) and other dependencies are broken to begin with.
Thomas Wouters [Fri, 11 Aug 2000 22:15:52 +0000 (22:15 +0000)]
Merge UNPACK_LIST and UNPACK_TUPLE into a single UNPACK_SEQUENCE, since they
did the same anyway.
I'm not sure what to do with Tools/compiler/compiler/* -- that isn't part of
distutils, is it ? Should it try to be compatible with old bytecode version ?
Fred Drake [Fri, 11 Aug 2000 19:55:06 +0000 (19:55 +0000)]
getopt(): revise description of long_options parameter slightly so it will
be less confusing; add a paragraph separation so that comments about
the options and long_options parameters don't have references that
are easily misinterpreted.
Adjust the interactive examples to not need the string module.
Add an example showing how the module is commonly used in a script.
Trent Mick [Fri, 11 Aug 2000 19:02:59 +0000 (19:02 +0000)]
Add largefile support for Linux64 and WIn64. Add test_largefile and some minor
change to regrtest.py to allow optional running of test_largefile ('cause it's
slow on Win64).
This closes patches:
http://sourceforge.net/patch/index.php?func=detailpatch&patch_id=100510&group_id=5470
and
http://sourceforge.net/patch/index.php?func=detailpatch&patch_id=100511&group_id=5470
Fred Drake [Thu, 10 Aug 2000 21:41:08 +0000 (21:41 +0000)]
Clean up compiler warning exposed by GCC's -Wall option: make sure
Python.h is included before standard headers since we set _GNU_SOURCE
there. This ensures that strdup() is prototyped.
Fredrik Lundh [Wed, 9 Aug 2000 18:51:01 +0000 (18:51 +0000)]
-- use explicit conversion instead of u-string literal in the test
code, to make the new Tkinter.py run under 1.5.2
-- changed Text.yview argument name (for consistency with xview)
Removing UTF-16 aware Unicode comparison code. This kind of compare
function (together with other locale aware ones) should into a new collation
support module. See python-dev for a discussion of this removal.
Note: This patch should also be applied to the 1.6 branch.
Fredrik Lundh [Mon, 7 Aug 2000 20:59:04 +0000 (20:59 +0000)]
-- reset marks if repeat_one tail doesn't match
(this should fix Sjoerd's xmllib problem)
-- added skip field to INFO header
-- changed compiler to generate charset INFO header
-- changed trace messages to support post-mortem analysis
When returning an error from jcompile() (which is passed through by
PyNode_Compile()), make sure that an exception is actually set --
otherwise someone stomped on our error. [2.0 checkin of this fix.]
Greg Ward [Mon, 7 Aug 2000 00:48:04 +0000 (00:48 +0000)]
Fix so the 'install_libbase' directory -- where .pth files are installed --
participates in the "--root" hack, ie. it also has a new root directory
hacked on at the very last minute (essential if the .pth file is to be
included in an RPM or other smart installer!).
Greg Ward [Mon, 7 Aug 2000 00:45:51 +0000 (00:45 +0000)]
Fix so we clear or reinitialize various data structures before populating
(allows the same FancyGetopt object to be used multiple times with different
option tables).
Thomas Wouters [Sat, 5 Aug 2000 23:28:51 +0000 (23:28 +0000)]
Cleanup configure.in. Specifically:
- Don't call both AC_CHECK_FUNCS and AC_REPLACE_FUNC for 'hypot', as the
latter already does everything the former does (because it's implemented as
a call to the former.)
- Don't call AC_CHECK_FUNC() without any 'action' clauses or with an action
clause that just defines HAVE_<function>. Instead, call AC_CHECK_FUNCS,
which defines 'HAVE_<function>' of itself, possibly with aditional 'action'
clauses.
No checks are removed by this patch, only moved around, and some duplicates
are removed.
Jack Jansen [Sat, 5 Aug 2000 21:29:58 +0000 (21:29 +0000)]
Changed H specifier to mean "bitfield", i.e. any value from
-32768..65535 is acceptable. Added B specifier (with values from
-128..255). No L added (which would have completed the set) because l
already accepts any value (and the letter L is taken for quadwords).
Greg Ward [Sat, 5 Aug 2000 01:25:24 +0000 (01:25 +0000)]
Drop the 'extend()' function -- old 1.5.1 compatibility hack that
wasn't actually used anywhere.
Drop the "from xxx_util import*" backwards compability hacks.
Greg Ward [Sat, 5 Aug 2000 00:58:14 +0000 (00:58 +0000)]
Pragmas that instruct the linker to link against python20.lib (or
python20_d.lib) only active on MSVC++; different library formats needed
for different compilers, and it's handled by the Distutils anyways.
Thomas Wouters [Fri, 4 Aug 2000 13:34:43 +0000 (13:34 +0000)]
Raise 'TestSkipped' (from the test_support) module rather than 'ImportError'
to signify a test that should be marked as 'skipped' rather than 'failed'.
Also 'document' it, in README.
Moshe Zadka [Fri, 4 Aug 2000 12:59:40 +0000 (12:59 +0000)]
In case the user isn't allowed to access /dev/dsp or /dev/dsp isn't there
at all (my computer doesn't have a Sound Blaster), this doesn't mean
there's a bug in linuxaudiodev. The only error the test suite skips
is currently ImportError -- so that's what we raise. If you see a problem
with this patch, say so and I'll retract. If you think raising an ImportError
sucks, you're right -- but I ain't gonna buy a SB and I sure ain't gonna
let the test-suite fail on my machine.
Greg Ward [Fri, 4 Aug 2000 01:30:03 +0000 (01:30 +0000)]
Rewrote 'find_library_file()' much more cleanly (and consistently with
MSVCCompiler's version, to aid in factoring common code out of the two
classes when the time comes).