Commit Graph

84 Commits

Author SHA1 Message Date
Mike Bayer 8423dbcf62 - shave about a millisecond off of moderately complex save casades. 2010-12-06 13:29:13 -05:00
Mike Bayer 175f1510d5 callcounts in 2.4 2010-12-05 19:29:47 -05:00
Mike Bayer 15297eb4fb callcounts... 2010-12-05 17:47:09 -05:00
Mike Bayer 1562860fea up this variance a bit, differences between 2.6.3/2.6.4 not worth it 2010-12-05 17:40:51 -05:00
Mike Bayer 1c16f373d9 callcounts for 2.4, 2.5, 2.6 2010-12-05 16:39:15 -05:00
Mike Bayer dd2b208d86 - fix callcount for 2.6 2010-12-05 16:09:53 -05:00
Mike Bayer f1e54a69fb - merge default tip 2010-12-05 14:56:26 -05:00
Mike Bayer 49145a6940 - more inlining. nominal execution on sqlite down to 36 calls, from 51 in 0.6. 2010-12-05 12:09:06 -05:00
Mike Bayer 58b2939433 - sqlalchemy.test and nose plugin moves back to being entirely
outside of "sqlalchemy" and under "test/".

Rationale:

- coverage plugin works without issue, without need for an awkward
additional package install
- command line for "nosetests" isn't polluted with SQLAlchemy options

[ticket:1949]
2010-11-28 14:19:44 -05:00
Mike Bayer 5c0d645353 - the "type_map" dictionary in sqlalchemy.types is now private,
i.e. is named "_type_map". [ticket:1870]
2010-11-28 13:07:00 -05:00
Mike Bayer 17fdeb77ad - fix test related to [ticket:1980] 2010-11-28 12:55:59 -05:00
Gaëtan de Menten 20418096d5 fixed a small potential memory leak in UnicodeResultProcessor (for some weird
reason, it didn't actually leak in my tests) by providing a dealloc method to
the type, and added a test to ensure it stays that way. Closes #1981.
2010-11-27 22:27:42 +01:00
Mike Bayer 61c76f92df - repair remaining tests 2010-11-20 11:35:16 -05:00
Taavi Burns 42a4445bf3 Remove unused import. 2010-11-20 08:32:00 -05:00
Taavi Burns 426501cba7 DecimalResultProcessor memory testing is likely to catch something only when cextensions are enabled. [ticket:1978] 2010-11-20 08:30:07 -05:00
Taavi Burns df1d3cb9d5 Shorter tests for [ticket:1978] 2010-11-19 21:27:30 -05:00
Taavi Burns f19d7e5b97 Fix memory leaks in the cprocessors DecimalResultProcessor, including tests. [ticket:1978] 2010-11-19 17:44:39 -05:00
Mike Bayer e1402efb19 - move sqlalchemy.test to test.lib 2010-11-15 19:37:50 -05:00
Mike Bayer 6d630990dd merge tip 2010-11-13 15:53:32 -05:00
Mike Bayer af4285e6ad - move inline "import" statements to use new "util.importlater()" construct. cuts
down on clutter, timeit says there's a teeny performance gain, at least where
the access is compared against attr.subattr.  these aren't super-critical
calls anyway
- slight inlining in _class_to_mapper
2010-11-13 13:19:36 -05:00
Mike Bayer a41c50ad63 - Implemented sequence check capability for the C
version of RowProxy, as well as 2.7 style
"collections.Sequence" registration for RowProxy.
[ticket:1871]
2010-11-09 11:40:23 -05:00
Mike Bayer 25c08f6def merge tip 2010-10-02 11:00:22 -04:00
Mike Bayer 755aca7f5f - reworked the internals of mapper.cascade_iterator() to
cut down method calls by about 9% in some circumstances.
[ticket:1932]
2010-10-01 14:23:01 -04:00
Mike Bayer 33b70a1f02 merge from tip 2010-09-03 17:57:20 -04:00
Mike Bayer 9914074390 reduce callcounts 2010-09-03 15:29:32 -04:00
Mike Bayer 9cbfe389f7 - pretty much all tests passing, maybe some callcounts are off
- test suite adjusted to use engine/pool events and not listeners
- deprecation warnings
2010-07-25 13:08:39 -04:00
Mike Bayer a6b62cc3fe Python-tidy test/engine and test/aaa_profiling, 80% auto + 20% manual intervention 2010-07-11 13:15:51 -04:00
Mike Bayer 7e72edc9d3 - updated obsolete docs regarding sequences, RETURNING for pg
- 78 chars for postgresql/base.py, engine/default.py
2010-07-06 17:39:20 -04:00
Mike Bayer 43dbfccf4d up the variance on these as the call counts are very low and
things wiggle around even between 2.6.4/2.6.5
2010-07-06 17:09:42 -04:00
Mike Bayer a9f2ed7b46 - latest distribute
- added caveats to unittest README encountered with Py2.7 + current nose 0.11.3
- call counts for py2.7.   all tests pass for sqlite + mysql-python + psycopg2
2010-07-05 19:54:46 -04:00
Mike Bayer c5ed8dfd8d version_info is the tuple 2010-06-15 14:05:02 -04:00
Mike Bayer 1769b2d8de - hypothesize that newer pysqlites are also growing a warning log on non-unicdoe strings 2010-06-14 20:15:27 -04:00
Mike Bayer 03e0f776be - The warning emitted by the Unicode and String types
with convert_unicode=True no longer embeds the actual
value passed.   This so that the Python warning
registry does not continue to grow in size, the warning
is emitted once as per the warning filter settings,
and large string values don't pollute the output.
[ticket:1822]
2010-06-13 16:09:13 -04:00
Mike Bayer 464f295854 fix this test for oracle 2010-05-31 14:46:21 -04:00
Mike Bayer cfe9fadc61 change the weakkeydict to be just an LRU cache. Add tests
for the "many combinations of UPDATE keys" issue.
2010-05-31 11:56:08 -04:00
Mike Bayer 9bab63b693 - Pool classes will reuse the same "pool_logging_name" setting
after a dispose() occurs.

- Engine gains an "execution_options" argument and
update_execution_options() method, which will apply to
all connections generated by this engine.

- Added more aggressive caching to the mapper's usage of
UPDATE, INSERT, and DELETE expressions.  Assuming the
statement has no per-object SQL expressions attached,
the expression objects are cached by the mapper after
the first create, and their compiled form is stored
persistently in a cache dictionary for the duration of
the related Engine.

- change #3 required change #1 so that we could test
a set of mappers operating over the course of many engines without
memory usage increase.
2010-05-30 20:24:08 -04:00
Mike Bayer ab62d418b2 guessed too soon on that 2010-04-16 14:31:36 -04:00
Mike Bayer 490a01c097 since we are calling traverse_single within ddl.py, call traverse_single at the entrypoint as well so that ddl.py
completely controls traversal
2010-04-16 13:48:45 -04:00
Mike Bayer cf65d1e42e check_reverse was failing a not well covered m2m case. 2010-04-10 20:13:51 -04:00
Mike Bayer 5df7420140 zoomark has additional callcount drops on this side 2010-04-07 14:28:33 -04:00
Mike Bayer e9d01db09b holy callcount batman 2010-04-07 14:23:06 -04:00
Mike Bayer fcf3536d1d add some callcounts with cextensions, [ticket:1723] 2010-03-26 12:47:01 -04:00
Mike Bayer 065fcbd9d2 - The official name for the relation() function is now
relationship(), to eliminate confusion over the relational
algebra term.  relation() however will remain available
in equal capacity for the foreseeable future.  [ticket:1740]
2010-03-17 17:48:29 -04:00
Mike Bayer 5dcc32fd5a - The visit_pool() method of Dialect is removed, and replaced with
on_connect().  This method returns a callable which receives
the raw DBAPI connection after each one is created.   The callable
is assembled into a first_connect/connect pool listener by the
connection strategy if non-None.   Provides a simpler interface
for dialects.
2010-03-15 13:08:31 -04:00
Mike Bayer f20138e2f9 callcount... 2010-03-13 14:54:41 -05:00
Mike Bayer 1f27092a45 move the initial connect out of the profile zone so we don't get a method call hit on that 2010-03-12 19:57:06 -05:00
Mike Bayer 969c04860f almost there, some tests failing still. 2010-03-09 13:41:43 -05:00
Mike Bayer 8c17344d1e preload type affinity entries so that these comparisons don't get shoved into the callcounts 2010-03-01 17:47:58 +00:00
Mike Bayer fbe6772d44 more bumps 2010-03-01 00:25:50 +00:00
Mike Bayer d840042748 some 2.4 bumps probably due to the tuple/list thing 2010-03-01 00:25:02 +00:00