]> git.ipfire.org Git - thirdparty/sqlalchemy/sqlalchemy.git/log
thirdparty/sqlalchemy/sqlalchemy.git
17 years agoAdded missing argument check on CheckConstraint
Jason Kirtland [Tue, 6 May 2008 14:40:24 +0000 (14:40 +0000)] 
Added missing argument check on CheckConstraint

17 years ago- _Label adds itself to the proxy collection so that it works in correspoinding colum...
Mike Bayer [Tue, 6 May 2008 00:55:49 +0000 (00:55 +0000)] 
- _Label adds itself to the proxy collection so that it works in correspoinding column.  fixes some eager load with column_property bugs.
- this partially fixes some issues in [ticket:1022] but leaving the "unlabeled" fix for 0.5 for now

17 years ago- added "after_begin()" hook to Session
Mike Bayer [Tue, 6 May 2008 00:47:36 +0000 (00:47 +0000)] 
- added "after_begin()" hook to Session
- Session.rollback() will rollback on a prepared session

17 years agoTidy.
Jason Kirtland [Tue, 6 May 2008 00:46:00 +0000 (00:46 +0000)] 
Tidy.

17 years agoFix typo
Lele Gaifax [Mon, 5 May 2008 23:08:11 +0000 (23:08 +0000)] 
Fix typo

17 years agoAdjusted inplace-binops on set-based collections and association proxies to
Jason Kirtland [Mon, 5 May 2008 21:33:29 +0000 (21:33 +0000)] 
Adjusted inplace-binops on set-based collections and association proxies to
more closely follow builtin (2.4+) set semantics.  Formerly any set duck-type
was accepted, now only types or subtypes of set, frozenset or the collection
type itself are accepted.

17 years agofailing case
Mike Bayer [Mon, 5 May 2008 17:50:19 +0000 (17:50 +0000)] 
failing case

17 years agoFixed duplicate append event emission on repeated instrumented set.add() operations.
Jason Kirtland [Mon, 5 May 2008 17:08:00 +0000 (17:08 +0000)] 
Fixed duplicate append event emission on repeated instrumented set.add() operations.

17 years agoUpdate for r4643
Jason Kirtland [Mon, 5 May 2008 16:53:39 +0000 (16:53 +0000)] 
Update for r4643

17 years agoRenamed rollback_returned to reset_on_return. Future, dialect-aware pools can do...
Jason Kirtland [Mon, 5 May 2008 16:50:58 +0000 (16:50 +0000)] 
Renamed rollback_returned to reset_on_return.  Future, dialect-aware pools can do better than rollback for this function.

17 years ago - same as [ticket:1019] but repaired the non-labeled use case
Mike Bayer [Mon, 5 May 2008 16:46:24 +0000 (16:46 +0000)] 
- same as [ticket:1019] but repaired the non-labeled use case
      [ticket:1022]

17 years ago - added "rollback_returned" option to Pool which will
Mike Bayer [Mon, 5 May 2008 15:52:09 +0000 (15:52 +0000)] 
- added "rollback_returned" option to Pool which will
      disable the rollback() issued when connections are
      returned.  This flag is only safe to use with a database
      which does not support transactions (i.e. MySQL/MyISAM).

17 years ago - Column.copy() respects the value of "autoincrement",
Mike Bayer [Mon, 5 May 2008 14:59:07 +0000 (14:59 +0000)] 
  - Column.copy() respects the value of "autoincrement",
      fixes usage with Migrate [ticket:1021]

17 years ago- fixes to the "exists" function involving inheritance (any(), has(),
Mike Bayer [Fri, 2 May 2008 19:23:58 +0000 (19:23 +0000)] 
- fixes to the "exists" function involving inheritance (any(), has(),
~contains()); the full target join will be rendered into the
EXISTS clause for relations that link to subclasses.

17 years ago- The collection instrumentation sweep now skips over descriptors that raise Attribut...
Jason Kirtland [Fri, 2 May 2008 18:47:05 +0000 (18:47 +0000)] 
- The collection instrumentation sweep now skips over descriptors that raise AttributeError.

17 years ago- fixed reentrant mapper compile hang when
Mike Bayer [Fri, 2 May 2008 18:23:16 +0000 (18:23 +0000)] 
- fixed reentrant mapper compile hang when
a declared attribute is used within ForeignKey,
ie. ForeignKey(MyOtherClass.someattribute)

17 years ago- Backported attribute sweep removal (instrumentation) and r4493 from 0.5
Jason Kirtland [Fri, 2 May 2008 17:46:44 +0000 (17:46 +0000)] 
- Backported attribute sweep removal (instrumentation) and r4493 from 0.5

17 years agoone-off workaround for mssql + odbc options, user patch
Rick Morrison [Fri, 2 May 2008 17:26:38 +0000 (17:26 +0000)] 
one-off workaround for mssql + odbc options, user patch

17 years ago- an unfortunate naming conflict
Mike Bayer [Fri, 2 May 2008 01:15:26 +0000 (01:15 +0000)] 
- an unfortunate naming conflict
- needed sql import on and()

17 years ago- factored out the logic used by Join to create its join condition
Mike Bayer [Fri, 2 May 2008 01:02:23 +0000 (01:02 +0000)] 
- factored out the logic used by Join to create its join condition
- With declarative, joined table inheritance mappers use a slightly relaxed
function to create the "inherit condition" to the parent
table, so that other foreign keys to not-yet-declared
Table objects don't trigger an error.

17 years agoadded some reference tests for the any() situation
Mike Bayer [Thu, 1 May 2008 15:17:07 +0000 (15:17 +0000)] 
added some reference tests for the any() situation

17 years ago- Fix 2.3 regression from 4598
Jason Kirtland [Wed, 30 Apr 2008 02:52:14 +0000 (02:52 +0000)] 
- Fix 2.3 regression from 4598

17 years ago- added a feature to eager loading whereby subqueries set
Mike Bayer [Wed, 30 Apr 2008 01:16:05 +0000 (01:16 +0000)] 
- added a feature to eager loading whereby subqueries set
as column_property() with explicit label names (which is not
necessary, btw) will have the label anonymized when
the instance is part of the eager join, to prevent
conflicts with a subquery or column of the same name
on the parent object.  [ticket:1019]

17 years agoAnd a copy.copy() test for the proxy cache.
Jason Kirtland [Tue, 29 Apr 2008 18:28:36 +0000 (18:28 +0000)] 
And a copy.copy() test for the proxy cache.

17 years ago- Refresh the cached proxy if the cache was built for a different instance.
Jason Kirtland [Tue, 29 Apr 2008 18:22:23 +0000 (18:22 +0000)] 
- Refresh the cached proxy if the cache was built for a different instance.

17 years agoadded multi-level concrete inheritance test (testing with_polymorphic mapper argument)
Gaëtan de Menten [Tue, 29 Apr 2008 13:35:59 +0000 (13:35 +0000)] 
added multi-level concrete inheritance test (testing with_polymorphic mapper argument)

17 years agomore declarative doc updates
Mike Bayer [Mon, 28 Apr 2008 00:04:05 +0000 (00:04 +0000)] 
more declarative doc updates

17 years agofix docs for declarative
Mike Bayer [Sun, 27 Apr 2008 23:37:02 +0000 (23:37 +0000)] 
fix docs for declarative

17 years agoSavepoints are supported under Firebird
Lele Gaifax [Sun, 27 Apr 2008 08:54:36 +0000 (08:54 +0000)] 
Savepoints are supported under Firebird

17 years agofix order by for MySQL environment
Mike Bayer [Sat, 26 Apr 2008 20:18:31 +0000 (20:18 +0000)] 
fix order by for MySQL environment

17 years ago- improved behavior of text() expressions when used as
Mike Bayer [Sat, 26 Apr 2008 16:34:14 +0000 (16:34 +0000)] 
- improved behavior of text() expressions when used as
FROM clauses, such as select().select_from(text("sometext"))
[ticket:1014]
- removed _TextFromClause; _TextClause just adds necessary FromClause descriptors
at the class level

17 years ago- refined mapper._save_obj() which was unnecessarily calling
Mike Bayer [Sat, 26 Apr 2008 16:13:49 +0000 (16:13 +0000)] 
- refined mapper._save_obj() which was unnecessarily calling
__ne__() on scalar values during flush [ticket:1015]

17 years agoExpanded --noncomparable to cover all comparision ops
Jason Kirtland [Fri, 25 Apr 2008 20:44:02 +0000 (20:44 +0000)] 
Expanded --noncomparable to cover all comparision ops

17 years agotypo
Gaëtan de Menten [Fri, 25 Apr 2008 12:31:47 +0000 (12:31 +0000)] 
typo

17 years agoUpdate docstring [ticket:873]
Jason Kirtland [Tue, 22 Apr 2008 20:04:51 +0000 (20:04 +0000)] 
Update docstring [ticket:873]

17 years agoExplicit test of .autoflush(False) to avoid issues with save_on_init=True [ticket...
Jason Kirtland [Tue, 22 Apr 2008 19:57:13 +0000 (19:57 +0000)] 
Explicit test of .autoflush(False) to avoid issues with save_on_init=True [ticket:869]

17 years agoflush(objects=[]) is a no-op [ticket:928]
Jason Kirtland [Tue, 22 Apr 2008 19:29:56 +0000 (19:29 +0000)] 
flush(objects=[]) is a no-op [ticket:928]

17 years ago - fixed Class.collection==None for m2m relationships
Mike Bayer [Tue, 22 Apr 2008 15:05:10 +0000 (15:05 +0000)] 
- fixed Class.collection==None for m2m relationships
      [ticket:4213]

17 years ago- restored usage of append_result() extension method for primary
Mike Bayer [Fri, 18 Apr 2008 14:49:21 +0000 (14:49 +0000)] 
- restored usage of append_result() extension method for primary
query rows, when the extension is present and only a single-
entity result is being returned.

17 years agoAdded 'odbc_options' keyword to the MSSQL dialect. Allows a partial ODBC connection...
Rick Morrison [Thu, 17 Apr 2008 19:07:12 +0000 (19:07 +0000)] 
Added 'odbc_options' keyword to the MSSQL dialect. Allows a partial ODBC connection string to be passed through to the connection string generator.

17 years ago- Support for COLLATE: collate(expr, col) and expr.collate(col)
Jason Kirtland [Wed, 16 Apr 2008 00:53:21 +0000 (00:53 +0000)] 
- Support for COLLATE: collate(expr, col) and expr.collate(col)

17 years ago- simplified __create_lazy_clause to make better usage of the new local/remote pairs...
Mike Bayer [Mon, 14 Apr 2008 18:23:59 +0000 (18:23 +0000)] 
- simplified __create_lazy_clause to make better usage of the new local/remote pairs collection
- corrected the direction of local/remote pairs for manytoone
- added new tests which demonstrate lazyloading working when the bind param is embedded inside of a SQL function,
when _local_remote_pairs argument is used; fixes the viewonly version of [ticket:610]
- removed needless kwargs check from visitors.traverse

17 years agoadded info about _local_remote_pairs to error message
Mike Bayer [Mon, 14 Apr 2008 15:54:23 +0000 (15:54 +0000)] 
added info about _local_remote_pairs to error message

17 years ago- added experimental relation() flag to help with primaryjoins
Mike Bayer [Mon, 14 Apr 2008 15:49:39 +0000 (15:49 +0000)] 
- added experimental relation() flag to help with primaryjoins
across functions, etc., _local_remote_pairs=[tuples].
This complements a complex primaryjoin condition allowing
you to provide the individual column pairs which comprise
the relation's local and remote sides.

17 years agoPass connection to get_default_schema_name
Jason Kirtland [Sun, 13 Apr 2008 16:21:17 +0000 (16:21 +0000)] 
Pass connection to get_default_schema_name

17 years agoFirebird 2 has a SUBSTRING() builtin, expose it thru a function
Lele Gaifax [Fri, 11 Apr 2008 22:14:34 +0000 (22:14 +0000)] 
Firebird 2 has a SUBSTRING() builtin, expose it thru a function

17 years ago- re-established viewonly relation() configurations that
Mike Bayer [Fri, 11 Apr 2008 15:58:18 +0000 (15:58 +0000)] 
- re-established viewonly relation() configurations that
join across multiple tables.

17 years agoAdd a new 'odbc_autotranslate' engine/dburi kwd parm to the MSSQL pyodbc dialect...
Rick Morrison [Tue, 8 Apr 2008 19:09:33 +0000 (19:09 +0000)] 
Add a new 'odbc_autotranslate' engine/dburi kwd parm to the MSSQL pyodbc dialect; string kwd contents will be passed through to ODBC connection string.
[ticket:1005]

17 years agoremove monetdb typo
Matt Harrison [Mon, 7 Apr 2008 22:48:39 +0000 (22:48 +0000)] 
remove monetdb typo

17 years agorefactor of default_paramstyle, use paramstyle argument on Dialect to change
Matt Harrison [Mon, 7 Apr 2008 22:42:28 +0000 (22:42 +0000)] 
refactor of default_paramstyle, use paramstyle argument on Dialect to change

17 years ago- Avoid cProfile on 2.4 (available via lsprof?)
Jason Kirtland [Mon, 7 Apr 2008 22:17:42 +0000 (22:17 +0000)] 
- Avoid cProfile on 2.4 (available via lsprof?)

17 years agoremove unneeded compile assertion test, doesn't work on MySQL
Mike Bayer [Mon, 7 Apr 2008 21:53:13 +0000 (21:53 +0000)] 
remove unneeded compile assertion test, doesn't work on MySQL

17 years ago*headslap* those mutators cant mutate the collections except for never-generated...
Mike Bayer [Mon, 7 Apr 2008 20:44:39 +0000 (20:44 +0000)] 
*headslap* those mutators cant mutate the collections except for never-generated selectables; its not worth it

17 years ago- removed ancient assertion that mapped selectables require
Mike Bayer [Mon, 7 Apr 2008 19:49:41 +0000 (19:49 +0000)] 
- removed ancient assertion that mapped selectables require
"alias names" - the mapper creates its own alias now if
none is present.  Though in this case you need to use
the class, not the mapped selectable, as the source of
column attributes - so a warning is still issued.

17 years agosome fk fixes for PG
Mike Bayer [Mon, 7 Apr 2008 01:15:37 +0000 (01:15 +0000)] 
some fk fixes for PG

17 years ago- merged -r4458:4466 of query_columns branch
Mike Bayer [Mon, 7 Apr 2008 01:12:44 +0000 (01:12 +0000)] 
- merged -r4458:4466 of query_columns branch
- this branch changes query.values() to immediately return an iterator, adds a new "aliased" construct which will be the primary method to get at aliased columns when using values()
- tentative ORM versions of _join and _outerjoin are not yet public, would like to integrate with Query better (work continues in the branch)
- lots of fixes to expressions regarding cloning and correlation.  Some apparent ORM bug-workarounds removed.
- to fix a recursion issue with anonymous identifiers, bind parameters generated against columns now just use the name of the column instead of the tablename_columnname label (plus the unique integer counter).  this way expensive recursive schemes aren't needed for the anon identifier logic.   This, as usual, impacted a ton of compiler unit tests which needed a search-n-replace for the new bind names.

17 years agorefined "local_remote_pairs" a bit to account for the same columns repeated multiple...
Mike Bayer [Fri, 4 Apr 2008 20:21:03 +0000 (20:21 +0000)] 
refined "local_remote_pairs" a bit to account for the same columns repeated multiple times

17 years ago- Pool listeners may now be specified as a duck-type of PoolListener or a dict of...
Jason Kirtland [Fri, 4 Apr 2008 19:07:30 +0000 (19:07 +0000)] 
- Pool listeners may now be specified as a duck-type of PoolListener or a dict of callables, your choice.

17 years agofactored down exportable_columns/flatten_cols/proxy_column/oid_etc_yada down to a...
Mike Bayer [Fri, 4 Apr 2008 18:41:08 +0000 (18:41 +0000)] 
factored down exportable_columns/flatten_cols/proxy_column/oid_etc_yada down to a single, streamlined "_populate_column_collection" method called for all selectables

17 years agofixed union() bug whereby oid_column would not be available if no oid_column in embed...
Mike Bayer [Fri, 4 Apr 2008 16:06:58 +0000 (16:06 +0000)] 
fixed union() bug whereby oid_column would not be available if no oid_column in embedded selects

17 years agobump
Mike Bayer [Fri, 4 Apr 2008 02:11:56 +0000 (02:11 +0000)] 
bump

17 years agoYep. rel_0_4_5
Jason Kirtland [Fri, 4 Apr 2008 00:58:11 +0000 (00:58 +0000)] 
Yep.

17 years ago- ReST fixes
Mike Bayer [Fri, 4 Apr 2008 00:49:13 +0000 (00:49 +0000)] 
- ReST fixes
- reverted strange jeklike symbol syntax

17 years ago- changed the name to "local/remote pairs"
Mike Bayer [Fri, 4 Apr 2008 00:31:00 +0000 (00:31 +0000)] 
- changed the name to "local/remote pairs"
- added closing ' to symbol str()  (I'm assuming it's supposed to be that way)

17 years ago- merged sync_simplify branch
Mike Bayer [Fri, 4 Apr 2008 00:21:28 +0000 (00:21 +0000)] 
- merged sync_simplify branch
- The methodology behind "primaryjoin"/"secondaryjoin" has
been refactored.  Behavior should be slightly more
intelligent, primarily in terms of error messages which
have been pared down to be more readable.  In a slight
number of scenarios it can better resolve the correct
foreign key than before.
- moved collections unit test from relationships.py to collection.py
- PropertyLoader now has "synchronize_pairs" and "equated_pairs"
collections which allow easy access to the source/destination
parent/child relation between columns (might change names)
- factored out ClauseSynchronizer (finally)
- added many more tests for priamryjoin/secondaryjoin
error checks

17 years ago- microcleanup
Jason Kirtland [Thu, 3 Apr 2008 17:32:22 +0000 (17:32 +0000)] 
- microcleanup

17 years ago- Experimental: prefer cProfile over hotspot for 2.5+
Jason Kirtland [Thu, 3 Apr 2008 17:08:08 +0000 (17:08 +0000)] 
- Experimental: prefer cProfile over hotspot for 2.5+
- The latest skirmish in the battle against zoomark and sanity:
  3rd party code is factored out in the function call count canary tests

17 years agoA couple of usage examples for the case statement
Ants Aasma [Thu, 3 Apr 2008 16:54:26 +0000 (16:54 +0000)] 
A couple of usage examples for the case statement

17 years ago- case() interprets the "THEN" expressions
Mike Bayer [Thu, 3 Apr 2008 16:34:03 +0000 (16:34 +0000)] 
- case() interprets the "THEN" expressions
as values by default, meaning case([(x==y, "foo")]) will
interpret "foo" as a bound value, not a SQL expression.
use text(expr) for literal SQL expressions in this case.
For the criterion itself, these may be literal strings
only if the "value" keyword is present, otherwise SA
will force explicit usage of either text() or literal().

17 years agosome cleanup, some method privating, some pep8, fixed up _col_aggregate and merged
Mike Bayer [Thu, 3 Apr 2008 16:25:47 +0000 (16:25 +0000)] 
some cleanup, some method privating, some pep8, fixed up _col_aggregate and merged
its functionality with _count()

17 years agoThe case() function now also takes a dictionary as its whens parameter. But beware...
Ants Aasma [Thu, 3 Apr 2008 14:08:22 +0000 (14:08 +0000)] 
The case() function now also takes a dictionary as its whens parameter. But beware that it doesn't escape literals, use the literal construct for that.

17 years ago- Added some convenience descriptors to Query:
Mike Bayer [Thu, 3 Apr 2008 13:12:42 +0000 (13:12 +0000)] 
- Added some convenience descriptors to Query:
query.statement returns the full SELECT construct,
query.whereclause returns just the WHERE part of the
SELECT construct.

17 years agoAdded a new 'max_identifier_length' keyword to the mssql_pyodbc dialect
Rick Morrison [Wed, 2 Apr 2008 23:03:00 +0000 (23:03 +0000)] 
Added a new 'max_identifier_length' keyword to the mssql_pyodbc dialect

17 years agoCascade traversal algorithm converted from recursive to iterative to support deep...
Ants Aasma [Wed, 2 Apr 2008 22:45:43 +0000 (22:45 +0000)] 
Cascade traversal algorithm converted from recursive to iterative to support deep object graphs.

17 years ago- Got PG server side cursors back into shape, added fixed
Mike Bayer [Wed, 2 Apr 2008 22:33:50 +0000 (22:33 +0000)] 
- Got PG server side cursors back into shape, added fixed
unit tests as part of the default test suite.  Added
better uniqueness to the cursor ID [ticket:1001]
- update().values() and insert().values() take keyword
arguments.

17 years ago- Re-tuned call counts for 2.3 through 2.5.
Jason Kirtland [Wed, 2 Apr 2008 18:13:53 +0000 (18:13 +0000)] 
- Re-tuned call counts for 2.3 through 2.5.

17 years ago- Run profiling tests first.
Jason Kirtland [Wed, 2 Apr 2008 17:55:11 +0000 (17:55 +0000)] 
- Run profiling tests first.

17 years agofixed OracleRaw type adaptation [ticket:902]
Mike Bayer [Wed, 2 Apr 2008 17:34:24 +0000 (17:34 +0000)] 
fixed OracleRaw type adaptation [ticket:902]

17 years agosome fixes to the MS-SQL aliasing so that result_map is properly populated
Mike Bayer [Wed, 2 Apr 2008 16:35:06 +0000 (16:35 +0000)] 
some fixes to the MS-SQL aliasing so that result_map is properly populated

17 years agodoh
Mike Bayer [Wed, 2 Apr 2008 16:15:57 +0000 (16:15 +0000)] 
doh

17 years agosome test fixup for oracle
Mike Bayer [Wed, 2 Apr 2008 16:12:07 +0000 (16:12 +0000)] 
some test fixup for oracle

17 years agoreduced 2.4 callcounts...
Mike Bayer [Wed, 2 Apr 2008 15:22:33 +0000 (15:22 +0000)] 
reduced 2.4 callcounts...

17 years agoslight function call reduction
Mike Bayer [Wed, 2 Apr 2008 15:07:06 +0000 (15:07 +0000)] 
slight function call reduction

17 years ago- Assorted flakes.
Jason Kirtland [Wed, 2 Apr 2008 11:51:06 +0000 (11:51 +0000)] 
- Assorted flakes.

17 years ago- Revamped the Connection memoize decorator a bit, moved to engine
Jason Kirtland [Wed, 2 Apr 2008 11:39:26 +0000 (11:39 +0000)] 
- Revamped the Connection memoize decorator a bit, moved to engine
- MySQL character set caching is more aggressive but will invalidate the cache if a SET is issued.
- MySQL connection memos are namespaced: info[('mysql', 'server_variable')]

17 years ago- More 2.4 generator squashing.
Jason Kirtland [Wed, 2 Apr 2008 10:39:06 +0000 (10:39 +0000)] 
- More 2.4 generator squashing.

17 years agocontinue attempting to get proper count for pybot on 2.5, ensure order_by for oracle...
Mike Bayer [Wed, 2 Apr 2008 02:42:29 +0000 (02:42 +0000)] 
continue attempting to get proper count for pybot on 2.5, ensure order_by for oracle query

17 years ago- added verbose activity to profiling.function_call_count
Mike Bayer [Tue, 1 Apr 2008 22:36:40 +0000 (22:36 +0000)] 
- added verbose activity to profiling.function_call_count
- simplified oracle non-ansi join generation, removed hooks from base compiler
- removed join() call from _label generation, fixed repeat label gen

17 years ago- More zzzeek enablement.
Jason Kirtland [Tue, 1 Apr 2008 18:33:03 +0000 (18:33 +0000)] 
- More zzzeek enablement.

17 years ago- Squashed 2.4 generators.
Jason Kirtland [Tue, 1 Apr 2008 18:31:20 +0000 (18:31 +0000)] 
- Squashed 2.4 generators.

17 years agoadded an order by to fix potential mysql test failure
Mike Bayer [Tue, 1 Apr 2008 18:11:13 +0000 (18:11 +0000)] 
added an order by to fix potential mysql test failure

17 years agoseems like the recent itertools add to select()._get_display_froms() adds overhead...
Mike Bayer [Tue, 1 Apr 2008 18:08:42 +0000 (18:08 +0000)] 
seems like the recent itertools add to select()._get_display_froms() adds overhead in 2.4?  not sure why

17 years agofix up some unit tests
Mike Bayer [Tue, 1 Apr 2008 17:46:36 +0000 (17:46 +0000)] 
fix up some unit tests

17 years ago- merge() may actually work now, though we've heard that before...
Mike Bayer [Tue, 1 Apr 2008 17:13:09 +0000 (17:13 +0000)] 
- merge() may actually work now, though we've heard that before...
- merge() uses the priamry key attributes on the object if _instance_key not present.  so merging works for instances that dont have an instnace_key, will still issue UPDATE for existing rows.
- improved collection behavior for merge() - will remove elements from a destination collection that are not in the source.
- fixed naive set-mutation issue in Select._get_display_froms
- simplified fixtures.Base a bit

17 years ago- Tighten up r4399 _set_iterable docs
Jason Kirtland [Tue, 1 Apr 2008 16:49:55 +0000 (16:49 +0000)] 
- Tighten up r4399 _set_iterable docs

17 years ago- Light collections refactor, added public collections.bulk_replace.
Jason Kirtland [Tue, 1 Apr 2008 16:38:23 +0000 (16:38 +0000)] 
- Light collections refactor, added public collections.bulk_replace.
- Collection attribs gain some private load-from-iterable flexiblity.

17 years agoweird, old cruft
Mike Bayer [Tue, 1 Apr 2008 04:51:02 +0000 (04:51 +0000)] 
weird, old cruft

17 years ago- removed redundant get_history() method
Mike Bayer [Tue, 1 Apr 2008 03:16:47 +0000 (03:16 +0000)] 
- removed redundant get_history() method
- the little bit at the bottom of _sort_circular_dependencies is absolutely covered by test/orm/cycles.py !  removing it breaks the test as run on PG.

17 years agoC-u 66 C-x f M-q
Jason Kirtland [Mon, 31 Mar 2008 22:45:34 +0000 (22:45 +0000)] 
C-u 66 C-x f M-q

17 years agoMSSQL adjustments to pyodbc connection string building
Rick Morrison [Mon, 31 Mar 2008 21:54:32 +0000 (21:54 +0000)] 
MSSQL adjustments to pyodbc connection string building