Commit Graph

2231 Commits

Author SHA1 Message Date
Martin v. Löwis 387c547fd3 Revert parts of patch #453627, documenting the resulting test failures
instead.
2001-09-06 08:16:17 +00:00
Tim Peters 6e13a562ae Enable large file support on Win32 systems.
Curious:  the MS docs say stati64 etc are supported even on Win95, but
Win95 doesn't support a filesystem that allows partitions > 2 Gb.

test_largefile:  This was opening its test file in text mode.  I have no
idea how that worked under Win64, but it sure needs binary mode on Win98.
BTW, on Win98 test_largefile runs quickly (under a second).
2001-09-06 00:32:15 +00:00
Tim Peters a40c793d06 Rework the way we try to check for libm overflow, given that C99 no longer
requires that errno ever get set, and it looks like glibc is already
playing that game.  New rules:

+ Never use HUGE_VAL.  Use the new Py_HUGE_VAL instead.

+ Never believe errno.  If overflow is the only thing you're interested in,
  use the new Py_OVERFLOWED(x) macro.  If you're interested in any libm
  errors, use the new Py_SET_ERANGE_IF_OVERFLOW(x) macro, which attempts
  to set errno the way C89 said it worked.

Unfortunately, none of these are reliable, but they work on Windows and I
*expect* under glibc too.
2001-09-05 22:36:56 +00:00
Martin v. Löwis 7c82a3e0fc Patch #449815: Set filesystemencoding based on CODESET. 2001-09-05 17:09:48 +00:00
Martin v. Löwis 655c9557f6 Patch #453627: Define the following macros when compiling on a UnixWare 7.x system:
SCO_ATAN2_BUG, SCO_ACCEPT_BUG, and STRICT_SYSV_CURSES.
Work aroudn a bug in the SCO UnixWare atan2() implementation.
2001-09-05 14:45:54 +00:00
Tim Peters e5ca6c71cd loghelper(): Try to nudge the compiler into doing mults in an order that
minimizes roundoff error.
2001-09-05 04:33:11 +00:00
Tim Peters 785261684e Return reasonable results for math.log(long) and math.log10(long) (we were
getting Infs, NaNs, or nonsense in 2.1 and before; in yesterday's CVS we
were getting OverflowError; but these functions always make good sense
for positive arguments, no matter how large).
2001-09-05 00:53:45 +00:00
Tim Peters 63c9453929 Mechanical fiddling to make this easier to work with in my editor.
Repaired the ldexp docstring (said the name of the func was "ldexp_doc").
2001-09-04 23:17:42 +00:00
Andrew M. Kuchling 4a596e3bee [Bug #457654] bkgd() used a hard-coded A_NORMAL attribute, when it should
have used the attribute argument provided as a parameter
2001-09-04 19:34:32 +00:00
Guido van Rossum 1832de4bc0 PEP 238 documented -Qwarn as warning only for classic int or long
division, and this makes sense.  Add -Qwarnall to warn for all
classic divisions, as required by the fixdiv.py tool.
2001-09-04 03:51:09 +00:00
Guido van Rossum 61c345fa37 Rename the -D option to -Q, to avoid a Jython option name conflict. 2001-09-04 03:26:15 +00:00
Guido van Rossum 97741a3041 Oops. The -W option takes args, not -X. 2001-08-31 18:17:13 +00:00
Guido van Rossum 393661d15f Add warning mode for classic division, almost exactly as specified in
PEP 238.  Changes:

- add a new flag variable Py_DivisionWarningFlag, declared in
  pydebug.h, defined in object.c, set in main.c, and used in
  {int,long,float,complex}object.c.  When this flag is set, the
  classic division operator issues a DeprecationWarning message.

- add a new API PyRun_SimpleStringFlags() to match
  PyRun_SimpleString().  The main() function calls this so that
  commands run with -c can also benefit from -Dnew.

- While I was at it, I changed the usage message in main() somewhat:
  alphabetized the options, split it in *four* parts to fit in under
  512 bytes (not that I still believe this is necessary -- doc strings
  elsewhere are much longer), and perhaps most visibly, don't display
  the full list of options on each command line error.  Instead, the
  full list is only displayed when -h is used, and otherwise a brief
  reminder of -h is displayed.  When -h is used, write to stdout so
  that you can do `python -h | more'.

Notes:

- I don't want to use the -W option to control whether the classic
  division warning is issued or not, because the machinery to decide
  whether to display the warning or not is very expensive (it involves
  calling into the warnings.py module).  You can use -Werror to turn
  the warnings into exceptions though.

- The -Dnew option doesn't select future division for all of the
  program -- only for the __main__ module.  I don't know if I'll ever
  change this -- it would require changes to the .pyc file magic
  number to do it right, and a more global notion of compiler flags.

- You can usefully combine -Dwarn and -Dnew: this gives the __main__
  module new division, and warns about classic division everywhere
  else.
2001-08-31 17:40:15 +00:00
Sjoerd Mullender 89dfe9e292 Removed unreachable return to silence SGI compiler. 2001-08-30 14:37:07 +00:00
Sjoerd Mullender 6f848c175f Removed an unreachable break statement to silence SGI compiler. 2001-08-30 14:15:38 +00:00
Neil Schemenauer 45c9ae5c97 gcmodule is now always compiled 2001-08-30 00:13:38 +00:00
Neil Schemenauer 7d0e19e1f4 gcmodule is now enabled here 2001-08-30 00:12:32 +00:00
Neil Schemenauer 43411b5683 Make more things internal to this file. Remove
visit_finalizer_reachable since it's the same as visit_reachable.
Rename visit_reachable to visit_move.  Objects can now have the GC type
flag set, reachable by tp_traverse and not be in a GC linked list.  This
should make the collector more robust and easier to use by extension
module writers.  Add memory management functions for container objects
(new, del, resize).
2001-08-30 00:05:51 +00:00
Neil Schemenauer e83c00efd0 Use new GC API. 2001-08-29 23:54:21 +00:00
Neil Schemenauer 01b66a80c4 Remove bogus PyGC_HEAD_SIZE. 2001-08-29 23:50:42 +00:00
Tim Peters 79248aa1e4 SF bug [#456252] Python should never stomp on [u]intptr_t.
pyport.h:  typedef a new Py_intptr_t type.
    DELICATE ASSUMPTION:  That HAVE_UINTPTR_T implies intptr_t is
    available as well as uintptr_t.  If that turns out not to be
    true, things must get uglier (C99 wants both, so I think it's
    an assumption we're *likely* to get away with).
thread_nt.h, PyThread_start_new_thread:  MS _beginthread is documented
    as returning unsigned long; no idea why uintptr_t was being used.
Others:  Always use Py_[u]intptr_t, never [u]intptr_t directly.
2001-08-29 21:37:10 +00:00
Tim Peters 12778e314b load_int: The fallback to long ints was coded in such a way that it
couldn't succeed.  Fixed.
2001-08-28 22:08:34 +00:00
Tim Peters 402d5985d8 SF patch [ #455137 ] Makes popen work with COMMAND.COM on WNT, from
Brian Quinlan.
2001-08-27 06:37:48 +00:00
Tim Peters 89675078cb Back out trying to use the C values for CO_xxx.
__future__.py reverted to 1.9.
newmodule.c reverted to 2.32.
2001-08-24 06:29:12 +00:00
Barry Warsaw 60f018846d Merge changes from r22a2-branch back into trunk. Also, change patch
level to 2.2a2+
2001-08-22 19:24:42 +00:00
Skip Montanaro 1a10aac87c make the gettmarg error message more correct by making it more vague ;-)
see SF bug 434143, part of which this addresses
2001-08-22 12:39:16 +00:00
Barry Warsaw cfb1675736 SSL_dealloc(): Apply the change suggested in SF bug #425370 which
changes the order of the free calls to be the reverse of the alloc
    calls.  Closes that bug.
2001-08-20 22:26:24 +00:00
Jeremy Hylton 1ba6bada67 It will always be a string, because it is created just before this call. 2001-08-20 19:06:36 +00:00
Guido van Rossum a92d16aaec SF patch #452239 by Gordon McMillan, to fix SF bug #451547.
This patch attempts to do to cPickle what Guido did
   for pickle.py v 1.50. That is: save_global tries
   importing the module, and fetching the name from the
   module. If that fails, or the returned object is not
   the same one we started with, it raises a
   PicklingError. (All this so pickling a lambda will
   fail at save time, rather than load time).
2001-08-18 21:22:07 +00:00
Tim Peters aa32070f4d Expose the CO_xxx flags via the "new" module (re-solving a problem "the
right way").  Fiddle __future__.py to use them.

Jeremy's pyassem.py may also want to use them (by-hand duplication of
magic numbers is brittle), but leaving that to his judgment.

Beef up __future__'s test to verify the exported feature names appear
correct.
2001-08-18 20:18:49 +00:00
Skip Montanaro 95618b5bc9 added warnings about security risk of using tmpnam and tempnam 2001-08-18 18:52:10 +00:00
Martin v. Löwis 339d0f720e Patch #445762: Support --disable-unicode
- Do not compile unicodeobject, unicodectype, and unicodedata if Unicode is disabled
- check for Py_USING_UNICODE in all places that use Unicode functions
- disables unicode literals, and the builtin functions
- add the types.StringTypes list
- remove Unicode literals from most tests.
2001-08-17 18:39:25 +00:00
Tim Peters 4b046c252e Stop adding 3 to FD_SETSIZE -- it makes no sense. If it turns out it
actually does <wink>, perhaps an Insure run will catch it.
Also removed senseless Windows comment.
2001-08-16 21:59:46 +00:00
Barry Warsaw 214a0b1382 init_sre(): Plug a little leak reported by Insure. 2001-08-16 20:33:48 +00:00
Barry Warsaw 2f7045576d Fixed a couple of minor formatting nits where lines were > 79 columns wide. 2001-08-16 16:55:10 +00:00
Barry Warsaw b44740f741 select_select(): Closing bug #448351 the easy way, i.e. by changing
the "#ifdef MS_WINDOWS" to "#ifdef SELECT_USES_HEAP" and by
    setting SELECT_USES_HEAP when FD_SETSIZE > 1024.

    The indirection seems useful since this subtly changes the path
    that "normal" Windows programs take (where Timmie sez FD_SETSIZE =
    512).  If that's a problem for Windows, he has only one place to
    change.
2001-08-16 16:52:59 +00:00
Fred Drake 7fdc0a1835 Use METH_O where possible (two functions). This does not lead to real
performance changes since the affected functions are not expected to be
used frequently, but reduces the volume of code.
2001-08-16 14:11:30 +00:00
Guido van Rossum 83f56cb2db Change the type names to xxsublist.<name>. 2001-08-16 09:10:42 +00:00
Guido van Rossum 88dcf03068 Add 'state' as a get/set attribute to spamlist. This serves as an
example.
2001-08-15 18:18:58 +00:00
Guido van Rossum e6b90eaea1 Add 'state' as a read-only attribute to spamdict. This serves as an
example.
2001-08-15 18:09:11 +00:00
Martin v. Löwis f95dd0a298 Fix portability problems with glibc 2.0, as reported in #449157. 2001-08-15 17:14:33 +00:00
Fred Drake 13130bc5b1 Use the abstract object interfaces when digging around in module objects
instead of directly manipulating the underlying dictionary.
2001-08-15 16:44:56 +00:00
Jack Jansen e925faff8e With WITH_NEXT_FRAMEWORK defined we now also expect a normal Python
installation.
If there is no LANDMARK we assume we're a bare framework in the
install directory (again WITH_NEXT_FRAMEWORK only).
2001-08-15 01:14:40 +00:00
Martin v. Löwis f65b1a175f Bump size of sprintf buffer. Suggested by Alex Coventry. 2001-08-12 09:28:40 +00:00
Martin v. Löwis 1ba3fd5d51 Autotest for netpacket/packet.h, as it is not available on all Linux versions.
Depend AF_PACKET on HAVE_NETPACKET_PACKET_H. Fixes #449157
2001-08-10 20:29:40 +00:00
Martin v. Löwis c4416d5a2c _DATE_FMT does not always accompany CODESET. 2001-08-10 19:41:45 +00:00
Neil Schemenauer 17e7be60b4 Remove "referents" structure (it's not needed). Check return value
of PyList_Append.
2001-08-10 14:46:47 +00:00
Martin v. Löwis 9b75dca192 Expose nl_langinfo through locale where available. 2001-08-10 13:58:50 +00:00
Fred Drake 428e75fc3e Add wrappers around the rich-comparison operations.
This closes SF patch #428320.

Add wrappers to expose "floor" and "true" division.
This closes SF feature request #449093.
2001-08-09 20:14:34 +00:00
Neil Schemenauer c7c8d8e32d Add get_objects function. This is a low level function (like
get_referents, and is not yet documented in the library manual).
Suggestions for a better name welcome.
2001-08-09 15:58:59 +00:00