c4d429fc2a
fix the timeout test once and for all
2010-03-13 11:52:17 -05:00
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
0744264be8
further refinement
2010-03-12 23:45:53 +00:00
d816c09f66
merging master...
2010-03-12 23:23:44 +00:00
6f00938055
consolidate unicode/CLOB handling
2010-03-12 23:22:20 +00:00
b9a1e94a73
adjust to allow mock DBAPIs
2010-03-12 16:49:32 -05:00
6efad7dfc6
oursql doesn't like the cursor being reused without fetch being finished
2010-03-12 16:19:41 -05:00
b9d275fdc1
merge from oracle WITH_UNICODE branch
2010-03-12 21:07:03 +00:00
022e8124ca
- Added preliminary support for Oracle's WITH_UNICODE
...
mode. At the very least this establishes initial
support for cx_Oracle with Python 3.
[ticket:1670]
2010-03-12 21:05:53 +00:00
d1983c22b4
these pass on +zxjdbc, seemingly because it lacks a sane rowcount
2010-03-12 11:07:53 -08:00
1b16954864
- introduce an optimizing type _NativeUnicodeMixin to oracle plus supporting
...
changes to Enum/SchemaType to re-support adaptation of string types.
This approach can be adapted by "conditional" unicode returning dialects
(i.e. pyodbc and possibly mxodbc) to remove the overhead
of isinstance(value, unicode) calls when the dialect returned type is
of dbapi.UNICODE, dbapi.NVARCHAR, etc.
2010-03-12 18:30:38 +00:00
b4f1fbc708
think I came across a syntactical weirdism there
2010-03-11 22:38:15 -05:00
6e14027620
refine that a bit
2010-03-11 22:12:49 -05:00
e66ca9edd5
adding windows-appropriate levels of pain to unicode handling
2010-03-11 22:10:25 -05:00
c486f6292f
merge tds changes
2010-03-11 21:55:09 -05:00
6c2e82ff8c
add a third state to converts_unicode_strings - "conditional". at the moment
...
this will have us do a check. i.e. for MSSQL where NVARCHAR is unicode and VARCHAR is not.
2010-03-11 21:52:54 -05:00
7c8942d4eb
Added a non-freetds specific exclusion for pyodbc
2010-03-11 21:23:25 -05:00
875487b8b7
formatting
2010-03-11 18:46:21 -05:00
3264433782
- query.scalar() now raises an exception if more than one
...
row is returned. All other behavior remains the same.
[ticket:1735]
2010-03-11 18:38:51 -05:00
409c8ed703
- Calling query.order_by() or query.distinct() before calling
...
query.select_from(), query.with_polymorphic(), or
query.from_statement() raises an exception now instead of
silently dropping those criterion. [ticket:1736]
2010-03-11 18:04:57 -05:00
30327525de
this resolves the returning failures from #1733
2010-03-11 21:36:03 +00:00
b763015642
re-enable these tests
2010-03-11 21:34:13 +00:00
ec0fafc7b0
added better typing for integer expressions, since integer is implementing _DateAffinity
2010-03-11 16:18:37 -05:00
3254699f85
force an oracle test to pass. this is as usual an extremely fringy test for a feature nobody should be using normally.
2010-03-11 18:52:02 +00:00
ee8eea4101
these fail for unknown reasons. Seems to be potentially VARCHAR used with RETURNING, or out params, not sure.
2010-03-11 18:39:15 +00:00
47f7b6f68e
fix this test for oracle so all types return datetime.date
2010-03-11 18:34:27 +00:00
9a3a909ed2
this test requires a length for the varchar in mysql/oracle
2010-03-11 18:21:44 +00:00
67f5f4831b
- fix the sqlalchemy.test.schema.Column function to work with copies
...
- add sequences to new associationproxy tests
- test/ext passes 100% on oracle here
2010-03-11 18:01:04 +00:00
757bf74ec0
merged default tip
2010-03-11 17:43:55 +00:00
e45a14e6fc
add autoincrement flags to decls
2010-03-11 17:43:34 +00:00
19de4da70f
- the copy() method of Column now copies over uninitialized
...
"on table attach" events. Helps with the new declarative
"mixin" capability.
2010-03-11 12:42:06 -05:00
0712f4aa0a
- mark tests unsupported by sqlite unsupported by oracle as well
2010-03-11 17:22:49 +00:00
86df449c2b
- fixed internal error which would occur if calling has()
...
or similar complex expression on a single-table inheritance
relation(). [ticket:1731]
2010-03-11 12:07:08 -05:00
ea4916001d
- Oracle 'DATE' now does not perform any result processing,
...
as the DATE type in Oracle stores full date+time objects,
that's what you'll get. Note that the generic types.Date
type *will* still call value.date() on incoming values,
however. When reflecting a table, the reflected type
will be 'DATE'.
2010-03-11 10:27:18 -05:00
9520c878b6
- emphasized query.join() in ORM tutorial as per [ticket:1708]
...
- cleaned up tutorial w.r.t. eagerload, added a section for contains_eager as this function
is equally important
- added better linkages in sqlalchemy.orm reference documentation, updated antiquated
docs for contains_eager(), got aliased()/AliasedClass documented as well as Sphinx will
allow us
2010-03-10 18:29:32 -05:00
53832b9fb2
fixed up docs for execution_options() across all three locations.
2010-03-09 18:45:30 -05:00
dcd6590252
- moved the logic within _BindParamClause which decides about coercing the bind's type
...
based on the compared type into AbstractType. The new method is called _coerce_compared_value()
and receives the operator and the raw python value to be coerced. TypeDecorator
overrides this to provude the "old" 0.5 behavior of coercing the other side of the expression
unconditonally.
- added docs to TypeDecorator attempting to explain this though they are a little verbose.
- added caveats to Interval, which in "non-native" mode can't really handle proper expression behavior. the "typing" would have to move into the compiler, such that on a DB like SQLite we're actually wrapping left/right side into its epoch functions, etc., its a big job. current infrastructure, as powerful as it is, still isn't at that level.
2010-03-09 17:46:00 -05:00
087c16c132
meaningless change
2010-03-09 16:47:45 -05:00
e9b059f782
base tests pass
2010-03-09 14:32:38 -05:00
969c04860f
almost there, some tests failing still.
2010-03-09 13:41:43 -05:00
62458eb8ad
merge latest tip. practicing on mercurial.
2010-03-08 23:54:30 -05:00
c3fd9a5536
still thinking about stuff here
2010-03-08 21:15:16 -05:00
b9cc0ef326
working on getting operators/left hand type awareness into the "bind" coercion. this system has to be figured out somehow
2010-03-08 18:27:35 -05:00
649d1f11c3
fix last commit
2010-03-08 13:04:32 -08:00
d9c7d9e0b0
small cleanup
2010-03-08 12:22:33 -08:00
35254ca25a
Remove isnew argument from row_processors. It wasn't used anyway in most cases.
2010-03-08 17:02:56 +01:00
5e188cce57
distutils choked on the trailing slash on the lib directory in the manifest.
2010-03-07 20:43:11 -06:00
4bac9453e2
the mercurial extension seems to not work in py3k and also I don't
...
like its usage during build/install. going with the manual approach
for the moment.
2010-03-07 20:56:32 -05:00
3ef21da0ac
use setuptools_hg to auto-discover the same way we did with SVN.
...
Thanks *much* to Jannis Leidel for the suggestion/creation !
2010-03-07 19:29:39 -05:00
2b83b38002
- plug a minor ORM speed hit in Events (**kwargs).
...
- added comment explaining some strange code
2010-03-07 20:45:39 +01:00
Mike Bayer