]> git.ipfire.org Git - thirdparty/sqlalchemy/sqlalchemy.git/log
thirdparty/sqlalchemy/sqlalchemy.git
11 years ago- changelog
Mike Bayer [Tue, 17 Dec 2013 20:40:19 +0000 (15:40 -0500)] 
- changelog

11 years ago- The :func:`.cast` function, when given a plain literal value,
Mike Bayer [Tue, 17 Dec 2013 20:38:35 +0000 (15:38 -0500)] 
- The :func:`.cast` function, when given a plain literal value,
will now apply the given type to the given literal value on the
bind parameter side according
to the type given to the cast.   This essentially replaces what would
normally be the detected type of the literal value.   This only
takes effect if the auto-detected type of the literal value is either
"nulltype" (e.g. couldn't detect)
or a type that is of the same "affinity" as the cast type.
The net change here is that the :func:`.cast` function includes more
of the functionality already present in the :func:`.type_coerce` function.

11 years agothis test appears to be failing with pg 9.3, not sure how to restore it
Mike Bayer [Tue, 17 Dec 2013 20:37:50 +0000 (15:37 -0500)] 
this test appears to be failing with pg 9.3, not sure how to restore it

11 years ago- using AssertsCompiledSQL so need __dialect__
Mike Bayer [Tue, 17 Dec 2013 20:33:27 +0000 (15:33 -0500)] 
- using AssertsCompiledSQL so need __dialect__

11 years ago- rework JSON expressions to be based off __getitem__ exclusively
Mike Bayer [Tue, 17 Dec 2013 20:13:39 +0000 (15:13 -0500)] 
- rework JSON expressions to be based off __getitem__ exclusively
- add support for "standalone" JSON objects; this involves getting CAST
to upgrade the given type of a bound parameter.  should add a core-only test
for this.
- add tests for "standalone" json round trips both with and without unicode
- add mechanism by which we remove psycopg2's "json" handler in order to get
the effect of using our non-native result handlers

11 years ago- enhance the exclusions system to support database versions within the __only_on__...
Mike Bayer [Tue, 17 Dec 2013 19:24:48 +0000 (14:24 -0500)] 
- enhance the exclusions system to support database versions within the __only_on__ attribute

11 years agoMerge branch 'issue_2581' of github.com:nathan-rice/sqlalchemy into pg_json
Mike Bayer [Tue, 17 Dec 2013 19:03:20 +0000 (14:03 -0500)] 
Merge branch 'issue_2581' of github.com:nathan-rice/sqlalchemy into pg_json

11 years ago- for [ticket:2651], leaving CheckConstraint alone, preferring to keep
Mike Bayer [Tue, 17 Dec 2013 00:32:10 +0000 (19:32 -0500)] 
- for [ticket:2651], leaving CheckConstraint alone, preferring to keep
backwards compatibility.  A note about backslashing escapes is added.
Because the Text() construct now supports bind params better, the example
given in the code raises an exception now, so that should cover us.
The exception itself has been enhanced to include the key name of the
bound param.  We're backporting this to 0.8 but 0.8 doesn't have the
text->bind behavior that raises.

11 years ago- An adjustment to the :func:`.subqueryload` strategy which ensures that
Mike Bayer [Tue, 17 Dec 2013 00:17:41 +0000 (19:17 -0500)] 
- An adjustment to the :func:`.subqueryload` strategy which ensures that
the query runs after the loading process has begun; this is so that
the subqueryload takes precedence over other loaders that may be
hitting the same attribute due to other eager/noload situations
at the wrong time. [ticket:2887]

11 years ago- fix up the FAQ regarding the "foo_id" issue
Mike Bayer [Mon, 16 Dec 2013 23:52:52 +0000 (18:52 -0500)] 
- fix up the FAQ regarding the "foo_id" issue
- add session object states to the glossary

11 years ago- Fixed bug when using joined table inheritance from a table to a
Mike Bayer [Mon, 16 Dec 2013 20:25:48 +0000 (15:25 -0500)] 
- Fixed bug when using joined table inheritance from a table to a
select/alias on the base, where the PK columns were also not same
named; the persistence system would fail to copy primary key values
from the base table to the inherited table upon INSERT.
[ticket:2885]

11 years agowrong method name
Mike Bayer [Mon, 16 Dec 2013 19:35:51 +0000 (14:35 -0500)] 
wrong method name

11 years ago- add "force_nocheck" as a way to turn on unicode=force without even
Mike Bayer [Mon, 16 Dec 2013 02:23:01 +0000 (21:23 -0500)] 
- add "force_nocheck" as a way to turn on unicode=force without even
doing the isinstance() check - currently used only by psycopg2 + native enum + py2k.
- didn't realize psycopg2 had UNICODEARRAY extension all this time; replace _PGArray
with just using UNICODEARRAY instead.
- replace unnecessary/inconsistent __import__ in _isolation_lookup.

11 years agoload_on_pending is different from enable_relationship_loading and should
Mike Bayer [Mon, 16 Dec 2013 01:06:27 +0000 (20:06 -0500)] 
load_on_pending is different from enable_relationship_loading and should
not be superseded.  both have a potential use.

11 years agomake the error message for [ticket:2889] more accurate, as we support
Mike Bayer [Thu, 12 Dec 2013 18:36:25 +0000 (13:36 -0500)] 
make the error message for [ticket:2889] more accurate, as we support
composites to many-to-ones now also

11 years ago- :func:`.composite` will raise an informative error message when the
Mike Bayer [Thu, 12 Dec 2013 18:30:16 +0000 (13:30 -0500)] 
- :func:`.composite` will raise an informative error message when the
columns/attribute (names) passed don't resolve to a Column or mapped
attribute (such as an erroneous tuple); previously raised an unbound
local. [ticket:2889]

11 years ago- Error message when a string arg sent to :func:`.relationship` which
Mike Bayer [Thu, 12 Dec 2013 17:49:57 +0000 (12:49 -0500)] 
- Error message when a string arg sent to :func:`.relationship` which
doesn't resolve to a class or mapper has been corrected to work
the same way as when a non-string arg is received, which indicates
the name of the relationship which had the configurational error.
[ticket:2888]

11 years agomove things that are 90% behavioral improvements to that section. the list of things
Mike Bayer [Thu, 12 Dec 2013 01:19:56 +0000 (20:19 -0500)] 
move things that are 90% behavioral improvements to that section.  the list of things
that can definitely people should be more focused.

11 years ago- implement "literal binds" for the text() clause, [ticket:2882]
Mike Bayer [Thu, 12 Dec 2013 01:00:39 +0000 (20:00 -0500)] 
- implement "literal binds" for the text() clause, [ticket:2882]

11 years ago- The :class:`.ForeignKey` class more aggressively checks the given
Mike Bayer [Thu, 12 Dec 2013 00:48:27 +0000 (19:48 -0500)] 
- The :class:`.ForeignKey` class more aggressively checks the given
column argument.   If not a string, it checks that the object is
at least a :class:`.ColumnClause`, or an object that resolves to one,
and that the ``.table`` attribute, if present, refers to a
:class:`.TableClause` or subclass, and not something like an
:class:`.Alias`.  Otherwise, a :class:`.ArgumentError` is raised.
[ticket:2883]

11 years ago- The :class:`.exc.StatementError` or DBAPI-related subclass
Mike Bayer [Wed, 11 Dec 2013 19:30:18 +0000 (14:30 -0500)] 
- The :class:`.exc.StatementError` or DBAPI-related subclass
now can accomodate additional information about the "reason" for
the exception; the :class:`.Session` now adds some detail to it
when the exception occurs within an autoflush.  This approach
is taken as opposed to combining :class:`.FlushError` with
a Python 3 style "chained exception" approach so as to maintain
compatibility both with Py2K code as well as code that already
catches ``IntegrityError`` or similar.

11 years ago sqlalchemy/dialects/postgresql/pgjson:
nathan [Wed, 11 Dec 2013 15:27:10 +0000 (10:27 -0500)] 
 sqlalchemy/dialects/postgresql/pgjson:
 - Updated documentation for JSON class

11 years ago sqlalchemy/dialects/postgresql/pgjson:
nathan [Wed, 11 Dec 2013 15:21:08 +0000 (10:21 -0500)] 
 sqlalchemy/dialects/postgresql/pgjson:
 - Added support for additional operators
 - Made return as json default (rather than text)

11 years ago sqlalchemy/dialects/postgresql/psycopg2:
nathan [Tue, 10 Dec 2013 15:09:15 +0000 (10:09 -0500)] 
 sqlalchemy/dialects/postgresql/psycopg2:
 - Removed unneeded import of psycopg2.extensions

11 years ago sqlalchemy/dialects/postgresql/pgjson:
nathan [Tue, 10 Dec 2013 15:01:51 +0000 (10:01 -0500)] 
 sqlalchemy/dialects/postgresql/pgjson:
 - Fixed reference to HSTORE
 - Corrected spelling of SQLAlchemy

 sqlalchemy/dialects/postgresql/psycopg2:
 - Added psycopg2 specific wrapper type for JSON which uses inherent json deserialization facilities
 - Added code to detect and utilize the JSON wrapper if psycopg2 >= 2.5

test/dialect/postgresql/test_types:
- removed reference to use_native_hstore

11 years ago- round trip test
Mike Bayer [Tue, 10 Dec 2013 02:27:14 +0000 (21:27 -0500)] 
- round trip test
- changelog
- some doc rearrangement

11 years agoMerge branch 'tsvector' of https://bitbucket.org/nibrahim/sqlalchemy/branch/tsvector...
Mike Bayer [Tue, 10 Dec 2013 02:01:26 +0000 (21:01 -0500)] 
Merge branch 'tsvector' of https://bitbucket.org/nibrahim/sqlalchemy/branch/tsvector into tsvector

11 years agosqlany dialect moves to github
Mike Bayer [Tue, 10 Dec 2013 01:56:10 +0000 (20:56 -0500)] 
sqlany dialect moves to github

11 years agoAdds test to verify tsvector creation.
Noufal Ibrahim [Mon, 9 Dec 2013 19:17:01 +0000 (00:47 +0530)] 
Adds test to verify tsvector creation.

Signed-off-by: Noufal Ibrahim <noufal@nibrahim.net.in>
11 years agoAdds tsvector to ischema_names for reflection to work.
Noufal Ibrahim [Mon, 9 Dec 2013 19:16:39 +0000 (00:46 +0530)] 
Adds tsvector to ischema_names for reflection to work.

Signed-off-by: Noufal Ibrahim <noufal@nibrahim.net.in>
11 years agoUpdates documentation for tsvector type.
Noufal Ibrahim [Thu, 5 Dec 2013 12:51:49 +0000 (18:21 +0530)] 
Updates documentation for tsvector type.

Signed-off-by: Noufal Ibrahim <noufal@nibrahim.net.in>
11 years agoImplements TSVECTOR type for postgresql.
Noufal Ibrahim [Thu, 5 Dec 2013 12:06:57 +0000 (17:36 +0530)] 
Implements TSVECTOR type for postgresql.

Signed-off-by: Noufal Ibrahim <noufal@nibrahim.net.in>
11 years agoMerge branch 'rel_0_9' of https://github.com/nathan-rice/sqlalchemy into rel_0_9
nathan [Mon, 9 Dec 2013 16:52:23 +0000 (11:52 -0500)] 
Merge branch 'rel_0_9' of https://github.com/nathan-rice/sqlalchemy into rel_0_9

11 years agosqlalchemy/dialects/postgresql/__init__.py:
nathan [Mon, 9 Dec 2013 16:46:36 +0000 (11:46 -0500)] 
sqlalchemy/dialects/postgresql/__init__.py:
- Added import references to JSON class

 sqlalchemy/dialects/postgresql/base.py:
 - Added visitor method for JSON class

 sqlalchemy/dialects/postgresql/pgjson (new):
 - JSON class, supports automatic serialization and deserialization of json data, as well as basic json operators.

11 years agoforgot authors!!
Mike Bayer [Sun, 8 Dec 2013 20:00:12 +0000 (15:00 -0500)] 
forgot authors!!

11 years ago- The :func:`.engine_from_config` function has been improved so that
Mike Bayer [Sat, 7 Dec 2013 23:38:15 +0000 (18:38 -0500)] 
- The :func:`.engine_from_config` function has been improved so that
we will be able to parse dialect-specific arguments from string
configuration dictionaries.  Dialect classes can now provide their
own list of parameter types and string-conversion routines.
The feature is not yet used by the built-in dialects, however.
[ticket:2875]

11 years ago- documentation cleanup in ORM including [ticket:2816]
Mike Bayer [Sat, 7 Dec 2013 23:00:34 +0000 (18:00 -0500)] 
- documentation cleanup in ORM including [ticket:2816]

11 years agospecify run as module here
Mike Bayer [Sat, 7 Dec 2013 22:37:36 +0000 (17:37 -0500)] 
specify run as module here

11 years ago- remove verbiage about "namespaces" from the dogpile example, [ticket:2862]
Mike Bayer [Sat, 7 Dec 2013 22:33:13 +0000 (17:33 -0500)] 
- remove verbiage about "namespaces" from the dogpile example, [ticket:2862]
- fix broken py2k/py3k isms

11 years ago- A DBAPI that raises an error on ``connect()`` which is not a subclass
Mike Bayer [Sat, 7 Dec 2013 22:20:05 +0000 (17:20 -0500)] 
- A DBAPI that raises an error on ``connect()`` which is not a subclass
of dbapi.Error (such as ``TypeError``, ``NotImplementedError``, etc.)
will propagate the exception unchanged.  Previously,
the error handling specific to the ``connect()`` routine would both
inappropriately run the exception through the dialect's
:meth:`.Dialect.is_disconnect` routine as well as wrap it in
a :class:`sqlalchemy.exc.DBAPIError`.  It is now propagated unchanged
in the same way as occurs within the execute process. [ticket:2881]
- add tests for this in test_parseconnect, but also add tests in test_execute
to ensure the execute() behavior as well

11 years ago- restore having the connection closed before decrementing the overflow counter
Mike Bayer [Sat, 7 Dec 2013 04:01:08 +0000 (23:01 -0500)] 
- restore having the connection closed before decrementing the overflow counter

11 years ago- The :class:`.QueuePool` has been enhanced to not block new connection
Mike Bayer [Sat, 7 Dec 2013 00:57:19 +0000 (19:57 -0500)] 
- The :class:`.QueuePool` has been enhanced to not block new connection
attempts when an existing connection attempt is blocking.  Previously,
the production of new connections was serialized within the block
that monitored overflow; the overflow counter is now altered within
it's own critical section outside of the connection process itself.
[ticket:2880]

11 years ago- Made a slight adjustment to the logic which waits for a pooled
Mike Bayer [Fri, 6 Dec 2013 20:53:59 +0000 (15:53 -0500)] 
- Made a slight adjustment to the logic which waits for a pooled
connection to be available, such that for a connection pool
with no timeout specified, it will every half a second break out of
the wait to check for the so-called "abort" flag, which allows the
waiter to break out in case the whole connection pool was dumped;
normally the waiter should break out due to a notify_all() but it's
possible this notify_all() is missed in very slim cases.
This is an extension of logic first introduced in 0.8.0, and the
issue has only been observed occasionally in stress tests.

11 years ago- The precedence rules for the :meth:`.ColumnOperators.collate` operator
Mike Bayer [Fri, 6 Dec 2013 00:03:31 +0000 (19:03 -0500)] 
- The precedence rules for the :meth:`.ColumnOperators.collate` operator
have been modified, such that the COLLATE operator is now of lower
precedence than the comparison operators.  This has the effect that
a COLLATE applied to a comparison will not render parenthesis
around the comparison, which is not parsed by backends such as
MSSQL.  The change is backwards incompatible for those setups that
were working around the issue by applying :meth:`.Operators.collate`
to an individual element of the comparison expression,
rather than the comparison expression as a whole. [ticket:2879]

11 years ago- changelog + test for pullreq #7, MSSQL dialect for DROP INDEX
Mike Bayer [Thu, 5 Dec 2013 23:36:59 +0000 (18:36 -0500)] 
- changelog + test for pullreq #7, MSSQL dialect for DROP INDEX

11 years agoMerged in dharland/sqlalchemy/patch-mssql-drop-index (pull request #7)
Mike Bayer [Thu, 5 Dec 2013 23:29:13 +0000 (18:29 -0500)] 
Merged in dharland/sqlalchemy/patch-mssql-drop-index (pull request #7)

Fix MSSQL dialects visit_drop_index to use the correct DDL

11 years ago- add a new example section for "join conditions", start putting
Mike Bayer [Thu, 5 Dec 2013 17:17:16 +0000 (12:17 -0500)] 
- add a new example section for "join conditions", start putting
the primaryjoin examples there

11 years agofix underline
Mike Bayer [Thu, 5 Dec 2013 17:13:37 +0000 (12:13 -0500)] 
fix underline

11 years agoFix MSSQL dialects visit_drop_index to use the correct DDL
donkopotamus [Thu, 5 Dec 2013 03:49:13 +0000 (16:49 +1300)] 
Fix MSSQL dialects visit_drop_index to use the correct DDL

11 years ago- reflection of unique constraints backported to 0.8.4 [ticket:1443]
Mike Bayer [Tue, 3 Dec 2013 20:01:34 +0000 (15:01 -0500)] 
- reflection of unique constraints backported to 0.8.4 [ticket:1443]

11 years ago- The :class:`.DeferredReflection` class has been enhanced to provide
Mike Bayer [Tue, 3 Dec 2013 18:46:41 +0000 (13:46 -0500)] 
- The :class:`.DeferredReflection` class has been enhanced to provide
automatic reflection support for the "secondary" table referred
to by a :func:`.relationship`.   "secondary", when specified
either as a string table name, or as a :class:`.Table` object with
only a name and :class:`.MetaData` object will also be included
in the reflection process when :meth:`.DeferredReflection.prepare`
is called. [ticket:2865]
- clsregistry._resolver() now uses a stateful _class_resolver()
class in order to handle the work of mapping strings to
objects.   This is to provide for simpler extensibility, namely
a ._resolvers collection of ad-hoc name resolution functions;
the DeferredReflection class adds its own resolver here in order
to handle relationship(secondary) names which generate new
Table objects.

11 years ago- Added ORA-02396 "maximum idle time" error code to list of
Mike Bayer [Tue, 3 Dec 2013 18:18:40 +0000 (13:18 -0500)] 
- Added ORA-02396 "maximum idle time" error code to list of
"is disconnect" codes with cx_oracle. [ticket:2864]

11 years ago- Added new argument ``include_backrefs=True`` to the
Mike Bayer [Mon, 2 Dec 2013 17:40:50 +0000 (12:40 -0500)] 
- Added new argument ``include_backrefs=True`` to the
:func:`.validates` function; when set to False, a validation event
will not be triggered if the event was initated as a backref to
an attribute operation from the other side. [ticket:1535]
- break out validation tests into an updated module test_validators

11 years ago- the pronoun removal commit. there was only one instance of a
Mike Bayer [Sat, 30 Nov 2013 22:31:00 +0000 (17:31 -0500)] 
- the pronoun removal commit.   there was only one instance of a
standalone gendered pronoun with a gender-neutral subject, but also
have replaced all occurences of "his/her", "his or her", etc.  The docs have always
strived to account for both genders in any non-specific singular pronoun,
however recent controversy in the community suggests
that a zero-gendered-pronoun policy is probably best going forward.

11 years agoremove whitespace
Mike Bayer [Sat, 30 Nov 2013 22:26:28 +0000 (17:26 -0500)] 
remove whitespace

11 years agomake sure thread.join() is used completely here
Mike Bayer [Sat, 30 Nov 2013 21:54:44 +0000 (16:54 -0500)] 
make sure thread.join() is used completely here

11 years ago.tox
Mike Bayer [Sat, 30 Nov 2013 21:45:53 +0000 (16:45 -0500)] 
.tox

11 years agoMerge pull request #49 from msabramo/add_note_to_README_unittests_about_tox
mike bayer [Sat, 30 Nov 2013 21:44:12 +0000 (13:44 -0800)] 
Merge pull request #49 from msabramo/add_note_to_README_unittests_about_tox

README.unittests.rst: Add blurb about tox

11 years ago- changelog
Mike Bayer [Sat, 30 Nov 2013 21:26:55 +0000 (16:26 -0500)] 
- changelog
- put list.clear() instrumentation under "if not py2k"

11 years agoMerge branch 'orm-collections-list-clear' of github.com:schettino72/sqlalchemy into...
Mike Bayer [Sat, 30 Nov 2013 21:17:27 +0000 (16:17 -0500)] 
Merge branch 'orm-collections-list-clear' of github.com:schettino72/sqlalchemy into list_clear

11 years agonot new anymore...
Mike Bayer [Sat, 30 Nov 2013 21:15:58 +0000 (16:15 -0500)] 
not new anymore...

11 years ago- move additional enum compilation tests to postgresql/test_compiler.py
Mike Bayer [Sat, 30 Nov 2013 21:12:20 +0000 (16:12 -0500)] 
- move additional enum compilation tests to postgresql/test_compiler.py

11 years ago- selected documentation issues
Mike Bayer [Sat, 30 Nov 2013 21:07:14 +0000 (16:07 -0500)] 
- selected documentation issues
- add glossary terms

11 years ago- re-document synonyms and remove warnings about "superseded"; synonyms
Mike Bayer [Sat, 30 Nov 2013 20:30:24 +0000 (15:30 -0500)] 
- re-document synonyms and remove warnings about "superseded"; synonyms
are still useful, just include notes that for more complex descriptor operations,
hybrids are probably preferable

11 years ago- Fixed bug where values within an ENUM weren't escaped for single
Mike Bayer [Sat, 30 Nov 2013 18:53:26 +0000 (13:53 -0500)] 
- Fixed bug where values within an ENUM weren't escaped for single
quote signs.   Note that this is backwards-incompatible for existing
workarounds that manually escape the single quotes. [ticket:2878]

11 years agoforgot to add this, oopsie
Mike Bayer [Sat, 30 Nov 2013 15:34:09 +0000 (10:34 -0500)] 
forgot to add this, oopsie

11 years agosimplify this
Mike Bayer [Sat, 30 Nov 2013 15:26:11 +0000 (10:26 -0500)] 
simplify this

11 years agoconvert to a list here for py3k
Mike Bayer [Sat, 30 Nov 2013 00:35:44 +0000 (19:35 -0500)] 
convert to a list here for py3k

11 years ago- add support for bindparam() called from AsFromText
Mike Bayer [Sat, 30 Nov 2013 00:06:33 +0000 (19:06 -0500)] 
- add support for bindparam() called from AsFromText
- get PG dialect to work around "no nonexistent binds" rule for now,
though we might want to reconsider this behavior

11 years ago- start reworking examples to include more code from the wiki.
Mike Bayer [Fri, 29 Nov 2013 23:53:53 +0000 (18:53 -0500)] 
- start reworking examples to include more code from the wiki.
- add the other versioning examples from the wiki
- modernize the dictlike examples

11 years agoremove this
Mike Bayer [Fri, 29 Nov 2013 23:53:37 +0000 (18:53 -0500)] 
remove this

11 years agoFixed bug where in Py2K a unicode literal would not be accepted
Mike Bayer [Fri, 29 Nov 2013 23:15:52 +0000 (18:15 -0500)] 
Fixed bug where in Py2K a unicode literal would not be accepted
as the string name of a class or other argument within
declarative using :func:`.relationship`.

11 years ago- reduce verbiage in mutation.py regarding legacy style, place under versionadded
Mike Bayer [Fri, 29 Nov 2013 21:32:24 +0000 (16:32 -0500)] 
- reduce verbiage in mutation.py regarding legacy style, place under versionadded

11 years agoFixed a syntax error in example code.
Hyunjun Kim [Thu, 21 Nov 2013 07:36:48 +0000 (16:36 +0900)] 
Fixed a syntax error in example code.

11 years agoMerge pull request #45 from timka/patch-1
mike bayer [Fri, 29 Nov 2013 21:12:17 +0000 (13:12 -0800)] 
Merge pull request #45 from timka/patch-1

Fix sessionmaker.__repr__

11 years agoMerge pull request #46 from vrajmohan/master
mike bayer [Fri, 29 Nov 2013 21:11:27 +0000 (13:11 -0800)] 
Merge pull request #46 from vrajmohan/master

More fixes for cross references and reducing warnings (3rd wave)

11 years agomore fix
Mike Bayer [Fri, 29 Nov 2013 20:26:00 +0000 (15:26 -0500)] 
more fix

11 years agofix
Mike Bayer [Fri, 29 Nov 2013 20:23:33 +0000 (15:23 -0500)] 
fix

11 years agomake generativeselect available
Mike Bayer [Fri, 29 Nov 2013 20:22:12 +0000 (15:22 -0500)] 
make generativeselect available

11 years ago- New improvements to the :func:`.text` construct, including
Mike Bayer [Fri, 29 Nov 2013 19:36:24 +0000 (14:36 -0500)] 
- New improvements to the :func:`.text` construct, including
more flexible ways to set up bound parameters and return types;
in particular, a :func:`.text` can now be turned into a full
FROM-object, embeddable in other statements as an alias or CTE
using the new method :meth:`.TextClause.columns`.
[ticket:2877]

11 years ago- changelog, migration
Mike Bayer [Fri, 29 Nov 2013 04:44:11 +0000 (23:44 -0500)] 
- changelog, migration

11 years ago- repair the "lockmode" functionality of load_on_ident(). slightly problematic
Mike Bayer [Fri, 29 Nov 2013 04:31:17 +0000 (23:31 -0500)] 
- repair the "lockmode" functionality of load_on_ident().  slightly problematic
here is that "lockmode" is also public in Session.refresh().

11 years ago- add support for specifying tables or entities for "of"
Mike Bayer [Fri, 29 Nov 2013 04:23:27 +0000 (23:23 -0500)] 
- add support for specifying tables or entities for "of"
- implement Query with_for_update()
- rework docs and tests

11 years ago- fix up rendering of "of"
Mike Bayer [Fri, 29 Nov 2013 03:25:09 +0000 (22:25 -0500)] 
- fix up rendering of "of"
- move out tests, dialect specific out of compiler, compiler tests use new API,
legacy API tests in test_selecatble
- add support for adaptation of ForUpdateArg, alias support in compilers

11 years ago- work in progress, will squash
Mike Bayer [Thu, 28 Nov 2013 17:37:15 +0000 (12:37 -0500)] 
- work in progress, will squash

11 years agoMerge branch 'for_update_of' of github.com:mlassnig/sqlalchemy into for_update_of
Mike Bayer [Thu, 28 Nov 2013 16:52:50 +0000 (11:52 -0500)] 
Merge branch 'for_update_of' of github.com:mlassnig/sqlalchemy into for_update_of

11 years agoadded LockmodeArgs
Mario Lassnig [Thu, 28 Nov 2013 13:50:41 +0000 (14:50 +0100)] 
added LockmodeArgs

11 years ago- add more explicit docs for relationship mixin, [ticket:2876]
Mike Bayer [Wed, 27 Nov 2013 16:18:55 +0000 (11:18 -0500)] 
- add more explicit docs for relationship mixin, [ticket:2876]

11 years ago- the wrapped memoized_property here was not working, as the attribute name
Mike Bayer [Wed, 27 Nov 2013 04:24:13 +0000 (23:24 -0500)] 
- the wrapped memoized_property here was not working, as the attribute name
didn't match.  use straight memoized_props here for now, add a perf test to check it

11 years ago- adjustment, the spec says: "Within the user and password field, any ":",
Mike Bayer [Mon, 25 Nov 2013 19:46:58 +0000 (14:46 -0500)] 
- adjustment, the spec says: "Within the user and password field, any ":",
   "@", or "/" must be encoded." - so re-apply encoding to both password
and username, don't encode spaces as plus signs, don't encode any chars
outside of :, @, / on stringification - but we still parse for any
%XX character (is that right?)

11 years agoREADME.unittests.rst: Add blurb about tox
Marc Abramowitz [Mon, 25 Nov 2013 18:18:24 +0000 (10:18 -0800)] 
README.unittests.rst: Add blurb about tox

11 years agoorm.collection, list.clear(). remove 'before_delete()', added unit-test.
schettino72 [Mon, 25 Nov 2013 01:18:34 +0000 (14:18 +1300)] 
orm.collection, list.clear(). remove 'before_delete()', added unit-test.

11 years ago- The :func:`.create_engine` routine and the related
Mike Bayer [Sun, 24 Nov 2013 23:11:37 +0000 (18:11 -0500)] 
- The :func:`.create_engine` routine and the related
:func:`.make_url` function **no longer URL encode the password**.
Database passwords that include characters like spaces, plus signs
and anything else should now represent these characters directly,
without any URL escaping. [ticket:2873]

11 years agopg8000 fix for decimal return scale
Mike Bayer [Sun, 24 Nov 2013 00:17:22 +0000 (19:17 -0500)] 
pg8000 fix for decimal return scale

11 years ago-be more agnostic of quotes here
Mike Bayer [Sat, 23 Nov 2013 22:22:56 +0000 (17:22 -0500)] 
-be more agnostic of quotes here

11 years ago- evaulate decimal_return_scale statelessly. Don't re-assign to self.decimal_return_...
Mike Bayer [Sat, 23 Nov 2013 22:19:43 +0000 (17:19 -0500)] 
- evaulate decimal_return_scale statelessly.  Don't re-assign to self.decimal_return_scale
so that __repr__() is maintained (for alembic tests)

11 years agoclean up imports
Mike Bayer [Sat, 23 Nov 2013 22:12:54 +0000 (17:12 -0500)] 
clean up imports

11 years ago- this test only for mysql
Mike Bayer [Sat, 23 Nov 2013 22:05:45 +0000 (17:05 -0500)] 
- this test only for mysql

11 years ago- Some refinements to the :class:`.AliasedClass` construct with regards
Mike Bayer [Sat, 23 Nov 2013 22:03:48 +0000 (17:03 -0500)] 
- Some refinements to the :class:`.AliasedClass` construct with regards
to descriptors, like hybrids, synonyms, composites, user-defined
descriptors, etc.  The attribute
adaptation which goes on has been made more robust, such that if a descriptor
returns another instrumented attribute, rather than a compound SQL
expression element, the operation will still proceed.
Addtionally, the "adapted" operator will retain its class; previously,
a change in class from ``InstrumentedAttribute`` to ``QueryableAttribute``
(a superclass) would interact with Python's operator system such that
an expression like ``aliased(MyClass.x) > MyClass.x`` would reverse itself
to read ``myclass.x < myclass_1.x``.   The adapted attribute will also
refer to the new :class:`.AliasedClass` as its parent which was not
always the case before. [ticket:2872]

11 years ago- The precision used when coercing a returned floating point value to
Mike Bayer [Sat, 23 Nov 2013 01:04:19 +0000 (20:04 -0500)] 
- The precision used when coercing a returned floating point value to
Python ``Decimal`` via string is now configurable.  The
flag ``decimal_return_scale`` is now supported by all :class:`.Numeric`
and :class:`.Float` types, which will ensure this many digits are taken
from the native floating point value when it is converted to string.
If not present, the type will make use of the value of ``.scale``, if
the type supports this setting and it is non-None.  Otherwise the original
default length of 10 is used. [ticket:2867]