]>
git.ipfire.org Git - thirdparty/sqlalchemy/sqlalchemy.git/log
Mike Bayer [Sun, 2 Aug 2009 16:32:53 +0000 (16:32 +0000)]
fix log to be util
Philip Jenvey [Sat, 1 Aug 2009 21:24:47 +0000 (21:24 +0000)]
unused import, rearrange
Mike Bayer [Sat, 1 Aug 2009 00:36:00 +0000 (00:36 +0000)]
- Databases which rely upon postfetch of "last inserted id" to get at a
generated sequence value (i.e. MySQL, MS-SQL) now work correctly
when there is a composite primary key where the "autoincrement" column
is not the first primary key column in the table.
Mike Bayer [Fri, 31 Jul 2009 23:07:21 +0000 (23:07 +0000)]
added test for already fixed case of oracle alias name truncation being applied properly, [ticket:1309]
Mike Bayer [Fri, 31 Jul 2009 14:47:27 +0000 (14:47 +0000)]
dont assume "returning" is present
Philip Jenvey [Fri, 31 Jul 2009 05:28:44 +0000 (05:28 +0000)]
move cx_oracle's bind param quoting out of base and its do_release_savepoint
into base
Philip Jenvey [Thu, 30 Jul 2009 00:25:43 +0000 (00:25 +0000)]
avoid unicodedata.normalize when on Java 5
Mike Bayer [Wed, 29 Jul 2009 21:37:10 +0000 (21:37 +0000)]
fix setuptools command
Mike Bayer [Wed, 29 Jul 2009 04:03:51 +0000 (04:03 +0000)]
don't create a bind for pk col + sequence if the backend doesn't support sequences
Philip Jenvey [Wed, 29 Jul 2009 00:47:32 +0000 (00:47 +0000)]
apparently this passes with implicit returning
Philip Jenvey [Wed, 29 Jul 2009 00:42:27 +0000 (00:42 +0000)]
merge r6218 from trunk
Mike Bayer [Tue, 28 Jul 2009 17:13:31 +0000 (17:13 +0000)]
improved reduce_columns() to navigate around a CompositeSelect that buries foreign-key holding columns down
on later selectables, [ticket:1486]
Mike Bayer [Tue, 28 Jul 2009 17:12:19 +0000 (17:12 +0000)]
boiled down "rendered_returning" etc. into just "returning"
Philip Jenvey [Tue, 28 Jul 2009 05:42:32 +0000 (05:42 +0000)]
avoid dumps output comparisons relying on dict ordering on jython
Philip Jenvey [Mon, 27 Jul 2009 03:06:29 +0000 (03:06 +0000)]
server_version_info for cx_Oracle
Mike Bayer [Mon, 27 Jul 2009 02:09:54 +0000 (02:09 +0000)]
- implicit returning support. insert() will use RETURNING to get at primary key values generated via sequence or default if the dialect detects the feature being availble. works for fb+ pg, needs work for oracle, mssql
Philip Jenvey [Mon, 27 Jul 2009 00:43:59 +0000 (00:43 +0000)]
initial oracle+zxjdbc support
Philip Jenvey [Mon, 27 Jul 2009 00:06:56 +0000 (00:06 +0000)]
this is more of a crash
Philip Jenvey [Sun, 26 Jul 2009 23:16:44 +0000 (23:16 +0000)]
remove unneeded custom ExecutionContext
Mike Bayer [Sun, 26 Jul 2009 18:58:54 +0000 (18:58 +0000)]
merged -r6172:6204 of trunk
Mike Bayer [Sun, 26 Jul 2009 18:39:31 +0000 (18:39 +0000)]
preexecute_pk_sequences indicates run the "default" separately to get its value
Philip Jenvey [Sat, 25 Jul 2009 20:20:26 +0000 (20:20 +0000)]
o move mysqldb's DecimalType behavior out of base
o add a BIT result processor for zxjdbc
Philip Jenvey [Sat, 25 Jul 2009 19:35:14 +0000 (19:35 +0000)]
fix the paramstyle check
Philip Jenvey [Sat, 25 Jul 2009 19:27:34 +0000 (19:27 +0000)]
reuse NoneType
Philip Jenvey [Fri, 24 Jul 2009 23:34:25 +0000 (23:34 +0000)]
revert r6184, toggle the jdbc driver's yearIsDateType option instead
Philip Jenvey [Fri, 24 Jul 2009 23:06:07 +0000 (23:06 +0000)]
mark as failing because zxjdbc relies on SELECT LAST_INSERT_ID()
Philip Jenvey [Fri, 24 Jul 2009 04:53:05 +0000 (04:53 +0000)]
disable mysqldb's Time type conversion on zxjdb
Philip Jenvey [Fri, 24 Jul 2009 04:51:55 +0000 (04:51 +0000)]
accommodate zxjdbc's datetime result
Philip Jenvey [Fri, 24 Jul 2009 04:12:10 +0000 (04:12 +0000)]
grab lastrowid on a different cursor so we don't munge the original's state
Philip Jenvey [Fri, 24 Jul 2009 01:46:12 +0000 (01:46 +0000)]
adjust select to not assume set order
Philip Jenvey [Thu, 23 Jul 2009 05:11:19 +0000 (05:11 +0000)]
close cursors
Philip Jenvey [Thu, 23 Jul 2009 04:58:06 +0000 (04:58 +0000)]
these are unnecessary
Philip Jenvey [Thu, 23 Jul 2009 04:49:25 +0000 (04:49 +0000)]
make Table doesn't exist detection portable
Philip Jenvey [Thu, 23 Jul 2009 04:48:25 +0000 (04:48 +0000)]
mysql+zxjdbc guards/allowances
Philip Jenvey [Thu, 23 Jul 2009 04:47:39 +0000 (04:47 +0000)]
o close connections to avoid maxing out the pool on Jython
o make test_charset_caching/test_num/test_expression more portable
Philip Jenvey [Thu, 23 Jul 2009 01:08:17 +0000 (01:08 +0000)]
add/strengthen guards for concurrent GC
Mike Bayer [Wed, 22 Jul 2009 20:41:58 +0000 (20:41 +0000)]
merged r6174 of trunk
Mike Bayer [Wed, 22 Jul 2009 20:33:40 +0000 (20:33 +0000)]
merged -r6134:6172 of trunk
Philip Jenvey [Wed, 22 Jul 2009 02:35:15 +0000 (02:35 +0000)]
apply the unique __hash__ workaround to all ClauseElements, and only apply it on Jython
fixes #1475
Philip Jenvey [Wed, 22 Jul 2009 02:17:26 +0000 (02:17 +0000)]
defer the heavier plugin setup until begin() when the plugin is definitely
enabled
Philip Jenvey [Wed, 22 Jul 2009 00:43:42 +0000 (00:43 +0000)]
revert the unicode check from r6162, zxjdbc just needs the correct
description_encoding
fixes #1479
Philip Jenvey [Tue, 21 Jul 2009 23:22:20 +0000 (23:22 +0000)]
really close connections
Philip Jenvey [Tue, 21 Jul 2009 21:07:04 +0000 (21:07 +0000)]
close connections
Philip Jenvey [Tue, 21 Jul 2009 05:58:20 +0000 (05:58 +0000)]
add fails_if for lack of supports_sane_rowcount support
Philip Jenvey [Tue, 21 Jul 2009 05:26:36 +0000 (05:26 +0000)]
expunge the session again so we're not recycling anything, for Jython
Philip Jenvey [Tue, 21 Jul 2009 05:26:26 +0000 (05:26 +0000)]
refactor to handle lazy loaded columns raising UnboundExecutionErrors. fixes
orm.test_merge test_no_load_with_backrefs on Jython, CPython avoided this due
to the luck of dict ordering
Philip Jenvey [Tue, 21 Jul 2009 03:10:53 +0000 (03:10 +0000)]
use supports_unicode_statements on Jython trunk and handle unicode cursor
description names from zxJDBC
Philip Jenvey [Tue, 21 Jul 2009 01:14:56 +0000 (01:14 +0000)]
don't rely on unicode repr() output
Philip Jenvey [Tue, 21 Jul 2009 01:04:26 +0000 (01:04 +0000)]
detect disconnections on closed cursors
Philip Jenvey [Tue, 21 Jul 2009 00:47:03 +0000 (00:47 +0000)]
ensure _BindParamClause hash codes are unique on Jython
fixes #1475
Philip Jenvey [Mon, 20 Jul 2009 05:02:32 +0000 (05:02 +0000)]
o we don't need to escape % in text on zxjdbc
o don't assume ProgrammingError on zxjdbc, hopefully zxjdbc won't use plain
Error so much one day
Philip Jenvey [Sun, 19 Jul 2009 21:12:12 +0000 (21:12 +0000)]
o don't assume exact insertion order in test_polymorph2.testjointo
o zxjdbc fails like pg8000 on test_query.test_values_with_boolean_selectspn
Mike Bayer [Sun, 19 Jul 2009 20:13:17 +0000 (20:13 +0000)]
removed disabled method
Mike Bayer [Sun, 19 Jul 2009 20:12:24 +0000 (20:12 +0000)]
one more neg to positive change
Mike Bayer [Sun, 19 Jul 2009 20:06:07 +0000 (20:06 +0000)]
- the "dont_load=True" flag on Session.merge() is deprecated and is now
"load=False".
- changed two internal "no_"/"dont_" arguments to positive names
Philip Jenvey [Sun, 19 Jul 2009 07:03:16 +0000 (07:03 +0000)]
assert -> eq_
Philip Jenvey [Sun, 19 Jul 2009 06:36:20 +0000 (06:36 +0000)]
skip sqlite tests when it's not available
Philip Jenvey [Sun, 19 Jul 2009 06:29:43 +0000 (06:29 +0000)]
disable test_prune on jython for now:
http://www.sqlalchemy.org/trac/ticket/1473
Philip Jenvey [Sun, 19 Jul 2009 05:39:39 +0000 (05:39 +0000)]
o handle array.ArrayTypes for Binary on jython
o tag zxjdbc as lacking datetime microsecond support
Mike Bayer [Sun, 19 Jul 2009 04:59:18 +0000 (04:59 +0000)]
generate the RETURNING col lists the same was as visit_select() does (except for oracle).
mssql gets extra label stuff to deal with column adaption (not sure if column adaption should
blow away labels like that...). fixes potential column targeting issues on all platforms
+ fixes mssql failures
Mike Bayer [Sun, 19 Jul 2009 04:06:39 +0000 (04:06 +0000)]
fixed [ticket:1471]
Philip Jenvey [Sun, 19 Jul 2009 02:52:40 +0000 (02:52 +0000)]
skip interval test_roundtrip on zxjdbc for now
Mike Bayer [Sun, 19 Jul 2009 02:20:18 +0000 (02:20 +0000)]
- returning() support is native to insert(), update(), delete(). Implementations
of varying levels of functionality exist for Postgresql, Firebird, MSSQL and
Oracle.
- MSSQL still has a few glitches that need to be resolved via label/column targeting logic.
- its looking like time to take another look at positional column targeting overall.
Philip Jenvey [Sun, 19 Jul 2009 02:05:14 +0000 (02:05 +0000)]
o disable array/pyformat/positional tests on zxjdbc
o zxjdbc execute doesn't like None params
Philip Jenvey [Sun, 19 Jul 2009 02:03:01 +0000 (02:03 +0000)]
this has been already moved to dialect/test_postgresql
Philip Jenvey [Sun, 19 Jul 2009 00:45:06 +0000 (00:45 +0000)]
reuse the simple mock_dbapi
Philip Jenvey [Sun, 19 Jul 2009 00:44:11 +0000 (00:44 +0000)]
don't assume sqlite is always available
Philip Jenvey [Sun, 19 Jul 2009 00:06:45 +0000 (00:06 +0000)]
long from zxjdbc is also a reasonable result
Philip Jenvey [Sun, 19 Jul 2009 00:06:04 +0000 (00:06 +0000)]
temp. workaround for deadlock on Jython
Philip Jenvey [Sat, 18 Jul 2009 23:41:41 +0000 (23:41 +0000)]
zxjdbc doesn't like None params here
Philip Jenvey [Sat, 18 Jul 2009 23:15:55 +0000 (23:15 +0000)]
fix Select's _copy_internals not maintaining _froms' order
Mike Bayer [Sat, 18 Jul 2009 22:01:33 +0000 (22:01 +0000)]
more doc stuff.
Mike Bayer [Sat, 18 Jul 2009 21:53:38 +0000 (21:53 +0000)]
- documentation crap
- we're probably going to want ez_setup.py around for now
- rubber malleted the access dialect to not look for old compiler classes
Mike Bayer [Sat, 18 Jul 2009 21:01:32 +0000 (21:01 +0000)]
merged -r6112:6134 of trunk
Philip Jenvey [Fri, 17 Jul 2009 05:35:02 +0000 (05:35 +0000)]
skip profiling tests on Jython and platforms that lack hotshot/cProfile
Philip Jenvey [Fri, 17 Jul 2009 02:50:39 +0000 (02:50 +0000)]
make our test skips recognizable by nose
Philip Jenvey [Fri, 17 Jul 2009 01:25:12 +0000 (01:25 +0000)]
close all connections
Philip Jenvey [Thu, 16 Jul 2009 23:02:00 +0000 (23:02 +0000)]
skip deadlocking test_two_phase_recover on mysql+zxjdbc, ensure cursors are
closed
Philip Jenvey [Thu, 16 Jul 2009 22:56:19 +0000 (22:56 +0000)]
ensure cursors are closed
Philip Jenvey [Thu, 16 Jul 2009 22:29:26 +0000 (22:29 +0000)]
ensure cursors are closed, various fetchone() -> first()
patch from Demi-god
Philip Jenvey [Thu, 16 Jul 2009 01:54:22 +0000 (01:54 +0000)]
correct the zxJDBC disconnect matching
Philip Jenvey [Thu, 16 Jul 2009 01:15:31 +0000 (01:15 +0000)]
small change
Philip Jenvey [Thu, 16 Jul 2009 01:15:01 +0000 (01:15 +0000)]
disable two phase transaction support on postgresql+zxjdbc, the jdbc driver
can't handle it without some more work
Philip Jenvey [Wed, 15 Jul 2009 20:58:01 +0000 (20:58 +0000)]
use locale neutral ISO date format
patch from Demi-God
Mike Bayer [Wed, 15 Jul 2009 18:48:10 +0000 (18:48 +0000)]
- fix driver imports
- update info on special schema names in unittest README
Mike Bayer [Mon, 13 Jul 2009 22:53:37 +0000 (22:53 +0000)]
beefed up the description of dialects
Mike Bayer [Mon, 13 Jul 2009 00:46:44 +0000 (00:46 +0000)]
renamed "postgres" to "postgresql", "Postgres" to "PostgreSQL". since
that's been the thing's name since 1996 or so. backwards compatibility
with the old name should be more or less complete.
Mike Bayer [Sun, 12 Jul 2009 23:45:01 +0000 (23:45 +0000)]
merged -r6082:6112 of trunk
Lele Gaifax [Sun, 12 Jul 2009 14:08:52 +0000 (14:08 +0000)]
Implements the boolean type for FB
Also, on reflection restore the heuristic needed to find the sequence linked to the PK.
Lele Gaifax [Sun, 12 Jul 2009 13:50:02 +0000 (13:50 +0000)]
Upgrade a few FB specific tests
Lele Gaifax [Sun, 12 Jul 2009 08:22:13 +0000 (08:22 +0000)]
Whitespace normalization
Mike Bayer [Sun, 12 Jul 2009 01:38:33 +0000 (01:38 +0000)]
bah
Mike Bayer [Sun, 12 Jul 2009 01:37:18 +0000 (01:37 +0000)]
FB uses AS for modern dialect
Mike Bayer [Sun, 12 Jul 2009 01:23:03 +0000 (01:23 +0000)]
- firebird support. reflection works fully, overall test success in the 75% range approx
- oracle and firebird now normalize column names to SQLA "lowercase" for result.keys()
allowing consistent identifier name visibility on the client side.
Mike Bayer [Sat, 11 Jul 2009 18:58:10 +0000 (18:58 +0000)]
- oracle fixes
- upon further consideration it seems appropriate that reflection would
coerce all "default" values into text(), since this represents a DDL expression
straight from the DB
- adjustments to mysql to accommodate that
- for some reason zoomark dropped 300 calls in create_tables. investigation pending
Mike Bayer [Wed, 8 Jul 2009 21:10:37 +0000 (21:10 +0000)]
continued strictness/documentation of the type system
Mike Bayer [Wed, 8 Jul 2009 02:17:18 +0000 (02:17 +0000)]
- type renaming extravaganza
- changed the contract a bit on reflection.cache
Mike Bayer [Mon, 6 Jul 2009 00:18:57 +0000 (00:18 +0000)]
- start moving get_default_schema_name to an initialized var
- fix up MSSQL foreign key reflection ala oracle
Mike Bayer [Sun, 5 Jul 2009 23:41:43 +0000 (23:41 +0000)]
use NVARCHAR/NTEXT for unicode...
Mike Bayer [Sun, 5 Jul 2009 23:12:44 +0000 (23:12 +0000)]
- a new type dialect convention starting out in the PG dialect.
- support for MSSQL/PyODBC/FreeTDS on a mac. Win32 on deck..