Tim Peters [Sun, 21 Mar 2004 23:38:41 +0000 (23:38 +0000)]
SF bug 847019 datetime.datetime initialization needs more strict checking
It's possible to create insane datetime objects by using the constructor
"backdoor" inserted for fast unpickling. Doing extensive range checking
would eliminate the backdoor's purpose (speed), but at least a little
checking can stop honest mistakes.
Armin Rigo [Sun, 21 Mar 2004 20:27:49 +0000 (20:27 +0000)]
This is the fastest I could get on Intel GCC. I kept the memset() in to clear
the newly created tuples, but tuples added in the freelist are now cleared in
tupledealloc already (which is very cheap, because we are already
Py_XDECREF'ing all elements anyway).
Python should have a standard Py_ZAP macro like ZAP in pystate.c.
[Part of patch #909005] Repeating exception changed from 'raise socket.error, why' to just raise. Make use of connect_ex() raise socket.error with 2-tuple instead of just error code
Tim Peters [Sun, 21 Mar 2004 16:59:09 +0000 (16:59 +0000)]
recursive_isinstance(), recursive_issubclass(): New code here returned
NULL in case of error, but the functions are declared to return int.
MSVC 6 properly complains about that. Return -1 on error instead.
Brett Cannon [Sun, 21 Mar 2004 14:06:49 +0000 (14:06 +0000)]
Back out last patch that removed an entry from sys.path if it was not an
existent path. Pointed out by jvr that entries could be non-file items for
custom importers.
Tim Peters [Sat, 20 Mar 2004 21:51:12 +0000 (21:51 +0000)]
SF patch 508730 CGIHTTPServer execfile should save cwd
UNTESTED!!!
This simple two-line patch has been sitting on SF for more than 2 years.
I'm guessing it's because nobody knows how to test it -- I sure don't.
It doesn't look like you can get to this part of the code on Unixish
or Windows systems, so the "how to test it?" puzzle has more than one
part. OTOH, if this is dead code, it doesn't matter either if I just
broke it <wink>.
Thomas Wouters [Sat, 20 Mar 2004 20:29:50 +0000 (20:29 +0000)]
test_email: comment out two fail-test cases that no longer fail with the new
parser -- for now. Failure behaviour of the new parser(s) will change in any
case, so this will be revisited later anyway.
Tim Peters [Sat, 20 Mar 2004 20:11:29 +0000 (20:11 +0000)]
Speed HMAC.copy() by installing a secret backdoor argument to
HMAC.__init__(). Adapted from SF patch 895445 "hmac.HMAC.copy() speedup"
by Trevor Perrin, who reported that this approach increased throughput
of his hmac-intensive app by 30%.
Armin Rigo [Sat, 20 Mar 2004 20:03:17 +0000 (20:03 +0000)]
A 2% speed improvement with gcc on low-endian machines. My guess is that this
new pattern for NEXTARG() is detected and optimized as a single (*short)
loading.
Guido van Rossum [Sat, 20 Mar 2004 19:11:58 +0000 (19:11 +0000)]
GCC was complaining that 'value' in dictiter_iternextvalue() wasn't
necessarily always set before used. Between Tim, Armin & me we
couldn't prove GCC wrong, so we decided to fix the algorithm. This
version is Armin's.
Thomas Wouters [Sat, 20 Mar 2004 17:31:29 +0000 (17:31 +0000)]
Merge in Anthony's new parser code, from the anthony-parser-branch:
> ----------------------------
> revision 1.20.4.4
> date: 2003/06/12 09:14:17; author: anthonybaxter; state: Exp; lines: +13 -6
> preamble is None when missing, not ''.
> Handle a couple of bogus formatted messages - now parses my main testsuite.
> Handle message/external-body.
> ----------------------------
> revision 1.20.4.3
> date: 2003/06/12 07:16:40; author: anthonybaxter; state: Exp; lines: +6 -4
> epilogue-processing is now the same as the old parser - the newline at the
> end of the line with the --endboundary-- is included as part of the epilogue.
> Note that any whitespace after the boundary is _not_ part of the epilogue.
> ----------------------------
> revision 1.20.4.2
> date: 2003/06/12 06:39:09; author: anthonybaxter; state: Exp; lines: +6 -4
> message/delivery-status fixed.
> HeaderParser fixed.
> ----------------------------
> revision 1.20.4.1
> date: 2003/06/12 06:08:56; author: anthonybaxter; state: Exp; lines: +163 -129
> A work-in-progress snapshot of the new parser. A couple of known problems:
>
> - first (blank) line of MIME epilogues is being consumed
> - message/delivery-status isn't quite right
>
> It still needs a lot of cleanup, but right now it parses a whole lot of
> badness that the old parser failed on. I also need to think about adding
> back the old 'strict' flag in some way.
> =============================================================================
Tim Peters [Sat, 20 Mar 2004 17:24:51 +0000 (17:24 +0000)]
Update copyright years, and change copyright.tex to use the same spelling
as license.tex (SF patch 795531 complained about the spelling discrepancy,
although this "repairs" that flaw in a different way than the patch).
Fred Drake [Sat, 20 Mar 2004 08:15:30 +0000 (08:15 +0000)]
commit the portion of PyXML patch #919008 that is relevant to the
standard library:
str() of xml.sax.SAXParseException should not fail if the line and/or
column number returned by the locator are None
(tests added)
Improve deque iteration.
* The default __reversed__ performed badly, so reintroduced a custom
reverse iterator.
* Added length transparency to improve speed with map(), list(), etc.
Make the new dictionary iterators transparent with respect to length.
This gives another 30% speedup for operations such as
map(func, d.iteritems()) or list(d.iteritems()) which can both take
advantage of length information when provided.
* Split into three separate types that share everything except the
code for iternext. Saves run time decision making and allows
each iternext function to be specialized.
* Inlined PyDict_Next(). In addition to saving a function call, this
allows a redundant test to be eliminated and further specialization
of the code for the unique needs of each iterator type.
* Created a reusable result tuple for iteritems(). Saves the malloc
time for tuples when the previous result was not kept by client code
(this is the typical use case for iteritems). If the client code
does keep the reference, then a new tuple is created.
Results in a 20% to 30% speedup depending on the size and sparsity
of the dictionary.
* Factored constant structure references out of the inner loops for
PyDict_Next(), dict_keys(), dict_values(), and dict_items().
Gave measurable speedups to each (the improvement varies depending
on the sparseness of the dictionary being measured).
* Added a freelist scheme styled after that for tuples. Saves around
80% of the calls to malloc and free. About 10% of the time, the
previous dictionary was completely empty; in those cases, the
dictionary initialization with memset() can be skipped.
Gregory P. Smith [Tue, 16 Mar 2004 18:50:26 +0000 (18:50 +0000)]
* supply a more useful error message when append() is called on the
wrong type of database in dbshelve.
* fix a typo in the exception name when checking args
Revert last change. Found an application that was worse off with resize
exact turned on. The tiny space savings wasn't worth the additional time
and code.
list_resize() now has an "exact" option for bypassing the overallocation
scheme in situations that likely won't benefit from it. This further
improves memory utilization from Py2.3 which always over-allocates
except for PyList_New().
Situations expected to benefit from over-allocation:
list.insert(), list.pop(), list.append(), and list.extend()
Situations deemed unlikely to benefit:
list_inplace_repeat, list_ass_slice, list_ass_subscript
The most gray area was for listextend_internal() which only runs
when the argument is a list or a tuple. This could be viewed as
a one-time fixed length addition or it could be viewed as wrapping
a series of appends. I left its over-allocation turned on but
could be convinced otherwise.
SF feature request #686323: Minor array module enhancements
array.extend() now accepts iterable arguments implements as a series
of appends. Besides being a user convenience and matching the behavior
for lists, this the saves memory and cycles that would be used to
create a temporary array object.
Update the array overallocation scheme to match the approach used for
lists. Speeds append() operations and reduces memory requirements
(because of more conservative overallocation).
Paves the way for the feature request for array.extend() to support
arbitrary iterable arguments.
Jack Jansen [Sat, 13 Mar 2004 23:50:48 +0000 (23:50 +0000)]
Two issues spotted by Ronald OUssoren:
- there were no accessor functions for the global per-database fields
- packages and their dependencies were installed in order in stead
of in reverse order.