Commit Graph

123 Commits

Author SHA1 Message Date
Guido van Rossum
b05a5c7698 Instead of importing graminit.h whenever one of the three grammar 'root'
symbols is needed, define these in Python.h with a Py_ prefix.
1997-05-07 17:46:13 +00:00
Guido van Rossum
fc49073cd0 Used operators from abstract.h where possible (arithmetic operators,
get/set/del item).  This removes a pile of duplication.  There's no
abstract operator for 'not' but I removed the function call for it
anyway -- it's a little faster in-line.
1997-05-06 15:06:49 +00:00
Guido van Rossum
a027efa5bf Massive changes for separate thread state management.
All per-thread globals are moved into a struct which is manipulated
separately.
1997-05-05 20:56:21 +00:00
Guido van Rossum
b209a116cf Quickly renamed. 1997-04-29 18:18:01 +00:00
Guido van Rossum
c43b685054 Clarify error message for unexpected keyword parameter. 1997-03-10 22:58:23 +00:00
Guido van Rossum
0d85be19e2 *Don't* kill all local variables on function exit. This will be done
by the frameobject dealloc when it is time for the locals to go.  When
there's still a traceback object referencing this stack frame, we
don't want the local variables to disappear yet.

(Hmm...  Shouldn't they be copied to the f_locals dictionary?)
1997-02-14 16:32:14 +00:00
Guido van Rossum
deb0c5e66c Two small changes:
- Use co->... instead of f->f_code->...; save an extra lookup of what
we already have in a local variable).

- Remove test for nlocals > 0 before setting fastlocals to
f->f_localsplus; 0 is a rare case and the assignment is safe even
then.
1997-01-27 23:42:36 +00:00
Guido van Rossum
d0eb429b88 Plug a leak with calling something other than a function or method is
called with keyword arguments -- the keyword and value were leaked.
This affected for instance with a __call__() method.

Bug reported and fix supplied by Jim Fulton.
1997-01-27 21:30:09 +00:00
Guido van Rossum
950361c6ca Patches for (two forms of) optional dynamic execution profiling --
i.e., counting opcode frequencies, or (with DXPAIRS defined) opcode
pair frequencies.  Define DYNAMIC_EXECUTION_PROFILE on the command
line (for this file and for sysmodule.c) to enable.
1997-01-24 13:49:28 +00:00
Guido van Rossum
8c5df06ec7 Change the control flow for error handling in the function prelude to
jump to the "Kill locals" section at the end.  Add #ifdef macintosh
bandaid to make sure we call sigcheck() on the Mac.
1997-01-24 04:19:24 +00:00
Guido van Rossum
a4240132ec Kill all local variables on function return. This closes a gigantic
leak of memory and file descriptors (thanks for Roj for reporting
that!).  Alas, the speed goes down by 5%. :-(
1997-01-21 21:18:36 +00:00
Guido van Rossum
70d44787a3 Only call sigcheck() at the ticker code if we don't have true signals.
This is safe now that both intrcheck() and signalmodule.c schedule a
sigcheck() call via Py_AddPendingCall().

This gives another 7% speedup (never run such a test twice ;-).
1997-01-21 06:15:24 +00:00
Guido van Rossum
1aa14838d2 Cleanup:
- fix bug in Py_MakePendingCalls() with threading
- fix return type of do_raise
- remove build_slice (same as PySlice_New)
- remove code inside #if 0
- remove code inside #ifdef CHECK_STACK
- remove code inside #ifdef SUPPORT_OBSOLETE_ACCESS
- comment about newimp.py should refer to ni.py
1997-01-21 05:34:20 +00:00
Guido van Rossum
768360243a Changes for frame object speedup:
- get fastlocals differently
- call newframeobject() with fewer arguments
- toss getowner(), which was unused anyway
1997-01-20 04:26:20 +00:00
Guido van Rossum
3dfd53b4c8 Add "if (x != NULL) continue;" (or similar for err==0) before the
break to most cases, as suggested by Tim Peters.  This gives another
8-10% speedup.
1997-01-18 02:46:13 +00:00
Guido van Rossum
62f7d15d0b Use the stack size from the code object and the CO_MAXBLOCKS constant
from compile.h.  Remove all eval stack overflow checks.
1997-01-17 21:05:28 +00:00
Guido van Rossum
408027ea46 Rename DEBUG macro to Py_DEBUG 1996-12-30 16:17:54 +00:00
Guido van Rossum
0aa9ee65ab Moved the raise logic out of the main interpreter loop to a separate function.
The raise logic has one additional feature: if you raise <class>,
<value> where <value> is not an instance, it will construct an
instance using <value> as argument.  If <value> is None, <class> is
instantiated without arguments.  If <value> is a tuple, it is used as
the argument list.

This feature is intended to make it easier to upgrade code from using
string exceptions to using class exceptions; without this feature,
you'd have to change every raise statement from ``raise X'' to ``raise
X()'' and from ``raise X, y'' to ``raise X(y)''.  The latter is still
the recommended form (because it has no ambiguities about the number
of arguments), but this change makes the transition less painful.
1996-12-10 18:07:35 +00:00
Guido van Rossum
150b2df682 Change the Don Beaudry hack into the Don B + Jim F hack; now, if *any*
base class is special it gets invoked.

Make gcc -Wall happy.
1996-12-05 23:17:11 +00:00
Guido van Rossum
d266eb460e New permission notice, includes CNRI. 1996-10-25 14:44:06 +00:00
Guido van Rossum
6d43c5de5a Raise TypeError, not KeyError, on unknown keyword argument. 1996-08-19 22:09:16 +00:00
Guido van Rossum
bf51afa049 Don't test here for negative number to float power; that belongs in
floatobject.c.
1996-08-16 20:49:17 +00:00
Guido van Rossum
0dfcf753ad Disable support for access statement 1996-08-12 22:00:53 +00:00
Guido van Rossum
3b9c6677f8 Better error message if stride used on normal sequence object 1996-07-30 18:40:29 +00:00
Guido van Rossum
8861b74445 Changes for slice and ellipses 1996-07-30 16:49:37 +00:00