Thomas Heller [Sun, 18 Jun 2006 21:27:04 +0000 (21:27 +0000)]
Fix typos.
Fix doctest example.
Mention in the tutorial that 'errcheck' is explained in the ref manual.
Use better wording in some places.
Remoce code examples that shouldn't be in the tutorial.
Remove some XXX notices.
Neal Norwitz [Sun, 18 Jun 2006 19:37:40 +0000 (19:37 +0000)]
The hppa ubuntu box sometimes hangs forever in these tests. My guess
is that the wait is failing for some reason. Use WNOHANG, so we won't
wait until the buildbot kills the test suite.
I haven't been able to reproduce the failure, so I'm not sure if
this will help or not. Hopefully, this change will cause the test
to fail, rather than hang. That will be better since we will get
the rest of the test results. It may also help us debug the real problem.
Neal Norwitz [Sun, 18 Jun 2006 19:35:01 +0000 (19:35 +0000)]
Prevent spurious leaks when running regrtest.py -R. There may be more
issues that crop up from time to time, but this change seems to have been
pretty stable (no spurious warnings) for about a week.
Other modules which use threads may require similar use of
threading_setup/threading_cleanup from test_support.
Tim Peters [Thu, 15 Jun 2006 18:04:40 +0000 (18:04 +0000)]
Try to reduce the extreme peak memory and disk-space use
of this test. It probably still requires more disk space
than most buildbots have, and in any case is still so
intrusive that if we don't find another way to test this I'm
taking my buildbot offline permanently ;-)
Neal Norwitz [Thu, 15 Jun 2006 09:57:03 +0000 (09:57 +0000)]
Steal the trick from test_compiler to print out a slow msg.
This will hopefully get the buildbots to pass. Not sure this
test will be feasible or even work. But everything is red now,
so it can't get much worse.
Gregory P. Smith [Thu, 15 Jun 2006 08:52:32 +0000 (08:52 +0000)]
- bsddb: multithreaded DB access using the simple bsddb module interface
now works reliably. It has been updated to use automatic BerkeleyDB
deadlock detection and the bsddb.dbutils.DeadlockWrap wrapper to retry
database calls that would previously deadlock. [SF python bug #775414]
Neal Norwitz [Thu, 15 Jun 2006 08:16:44 +0000 (08:16 +0000)]
Re-revert this change. Install the version check and don't run the test
until Gerhard has time to fully debug the issue. This affects versions
before 3.2.1 (possibly only versions earlier than 3.1.3).
Neal Norwitz [Thu, 15 Jun 2006 04:54:29 +0000 (04:54 +0000)]
Speculative checkin (requires approval of Gerhard Haering)
This backs out the test changes in 46962 which prevented crashes
by not running the tests via a version check. All the version checks
added in that rev were removed from the tests.
Code was added to the error handler in connection.c that seems
to work with older versions of sqlite including 3.1.3.
Gerhard Häring [Wed, 14 Jun 2006 22:28:37 +0000 (22:28 +0000)]
- Added version checks in C code to make sure we don't trigger bugs in older
SQLite versions.
- Added version checks in test suite so that we don't execute tests that we
know will fail with older (buggy) SQLite versions.
Now, all tests should run against all SQLite versions from 3.0.8 until 3.3.6
(latest one now). The sqlite3 module can be built against all these SQLite
versions and the sqlite3 module does its best to not trigger bugs in SQLite,
but using SQLite 3.3.3 or later is recommended.
Tim Peters [Wed, 14 Jun 2006 04:09:25 +0000 (04:09 +0000)]
SequenceMatcher.get_matching_blocks(): This now guarantees that
adjacent triples in the result list describe non-adjacent matching
blocks. That's _nice_ to have, and Guido said he wanted it.
Not a bugfix candidate: Guido or not ;-), this changes visible
endcase semantics (note that some tests had to change), and
nothing about this was documented before. Since it was working
as designed, and behavior was consistent with the docs, it wasn't
"a bug".
Gerhard Häring [Tue, 13 Jun 2006 22:24:47 +0000 (22:24 +0000)]
Merged changes from external pysqlite 2.3.0 release. Documentation updates will
follow in a few hours at the latest. Then we should be ready for beta1.
Brett Cannon [Tue, 13 Jun 2006 21:46:41 +0000 (21:46 +0000)]
If a classic class defined a __coerce__() method that just returned its two
arguments in reverse, the interpreter would infinitely recourse trying to get a
coercion that worked. So put in a recursion check after a coercion is made and
the next call to attempt to use the coerced values.
Andrew MacIntyre [Tue, 13 Jun 2006 15:04:24 +0000 (15:04 +0000)]
Patch #1454481: Make thread stack size runtime tunable.
Heavily revised, comprising revisions:
46640 - original trunk revision (backed out in r46655)
46647 - markup fix (backed out in r46655)
46692:46918 merged from branch aimacintyre-sf1454481
branch tested on buildbots (Windows buildbots had problems
not related to these changes).
Ronald Oussoren [Tue, 13 Jun 2006 11:19:56 +0000 (11:19 +0000)]
Linecache contains support for PEP302 loaders, but fails to deal with loaders
that return None to indicate that the module is valid but no source is
available. This patch fixes that.
Tim Peters [Tue, 13 Jun 2006 03:30:07 +0000 (03:30 +0000)]
get_matching_blocks(): rewrote code & comments so they match; added
more comments about why it's this way at all; and removed what looked
like needless expense (sorting (i, j, k) triples directly should give
exactly the same order as sorting (i, (i, j, k)) pairs).
Fix the CRT argument error handling for VisualStudio .NET 2005. Install a CRT error handler and disable the assertion for debug builds. This causes CRT to set errno to EINVAL.
This update fixes crash cases in the test suite where the default CRT error handler would cause process exit.