release 0.9.4-RC4

[Maxime Lévesque]

Hi, I've published 0.9.4-RC4 :

http://nexus.scala-tools.org/content/repositories/releases/org/squeryl/squeryl_2.8.1/0.9.4-RC4/squeryl_2.8.1-0.9.4-RC4.pom

I am announcing it here first, if I don't hear any bad news I'll
announce it on the squeryl
list next week, and if that goes well, I should publish 0.9.4.

Here's the git log since RC3 :


commit 7d97454c42a1ce2e6aba73ccc99712e0ae4dc609
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Feb 2 13:25:25 2011 -0500

0.9.4-RC4 release


commit 140c8767d01fc40b3cd383f573f0553f3aa57786
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Feb 2 08:02:27 2011 -0500

Added Hook for influencing YieldInspection (to be used by lift-
squeryl-record)


commit 559508793963a2c060d8046f926102c785867f5d
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Feb 2 08:00:36 2011 -0500

Crash early when on/declare is given a null table


commit e75ddecf4a0855771dd569b4c4df4e23fde2133e
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sat Jan 29 18:17:06 2011 -0500

Support for subqueries in set clause of partial update, fixes #6


commit d07acf9e6dcea3753d51f2c35e2b22e29dc3cea1
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sat Jan 29 10:53:29 2011 -0500

Added assertion in test case for subselect arg. to 'in' function


commit 314462f9a5157844af9c2e6204d701f514fcec0a
Author: Marco Rico-Gomez <m.ric...@gmail.com>
Date: Tue Jan 25 16:10:55 2011 +0100

Closes #93: Add secondary constructors in custom types

This commit replaces the algorithm of _createCustomTypeFactory in
FieldMetaData.
The former version takes the first one-arg constructor in the
field class
for which it can create a default value and returns a closure for
it.
This algorithm fails if the field's class has multiple one-arg
constructors
for different types.

The new version of the algorithm is able to handle classes with
multiple
one-arg constructors. It chooses the best constructor for the
given
type to create an instance.


commit 3ffe9395d5a3f0213569eb0ab708e93a36de25c9
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sat Jan 22 14:19:58 2011 -0500

Aggregate (groupBy and compute) return Group[ProductX] instead of
Group[TupleX],
same for Measure[TupleX] and GroupWithMeasures,
Bad news : possible breaking change (if the user code was
declaring or assigning results to TupleX)
Good news : the compiler will catch all of these
incompatilibities, and the refactoring is
simply to change TupleX declarations to ProductX.


commit 82f432165c420b5326e6ef95b03f2d362023c998
Merge: 4989b04 26f1bf4
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sat Jan 22 08:13:19 2011 -0500

Merge branch 'master' of git://github.com/wyuenho/Squeryl into
wyuenho-master

commit 26f1bf488048806dfa39080b723690c6a0bab72d
Author: Yuen Ho Wong <wyu...@MobileCat.local>
Date: Sat Jan 22 19:26:11 2011 +0800

reorganized patch for issue 81


commit 326c16e7896212912b4a803bc8f29b3206a477f1
Author: Yuen Ho Wong <wyu...@MobileCat.local>
Date: Sat Jan 22 18:09:05 2011 +0800

Revert "patch for issue 81"

This reverts commit 4d7f8aa0bf60dc047f677329f780b2d972f9905c.


commit 4989b0474345f25fc641c99907bccf527c9d0775
Merge: 3ddd49c 8235eee
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Fri Jan 21 22:29:26 2011 -0500

Merge branch 'master' of git://github.com/max-l/Squeryl

commit 8235eeeb5bcf82de60938d57e05dfcfd31a08efc
Author: Marco Rico-Gomez <m.ric...@gmail.com>
Date: Mon Jan 17 16:22:00 2011 +0100

Closes #43: Equality expression not reversible in OneToMany


commit 9dec46fc7c7dc1ba7b17fc1e470d0a3a477ebbdd
Author: Marco Rico-Gomez <m.ric...@gmail.com>
Date: Thu Jan 20 11:47:09 2011 +0100

fixes #2: Support query pagination in DerbyAdapter


commit 3ddd49c06edb8aaf48a6df55e365094f7ce96b4d
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Thu Jan 20 10:35:54 2011 -0500

Syntax sugar for the 'associate' method in relations


commit 41d67e4efcff5cf188b97bf7d7be4dbe8288290b
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Jan 19 21:43:05 2011 -0500

added gt,lt,etc, to NonNumericalExpression for consistency
http://github.com/max-l/Squeryl/issues#issue/86


commit 167da2b1c387a6aa62bd4334fc671dfd2c9004eb
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Jan 19 21:11:47 2011 -0500

implicit conversion from boolean to LogicalBoolean, to allow this
kind of syntax sugar :
students.where(s => s.isMultilingual)
as opposed to :
students.where(s=> s.isMultilingual===true)

https://github.com/max-l/Squeryl/issues#issue/78


commit 76f838b6b716df6728a0aa366d788efd93206d76
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Jan 19 20:29:45 2011 -0500

fixes #3 (assign method in relations broken when using
CustomTypes)


commit fba30ee928dd3ead5ff3fbc5bbb4f0522ba556e1
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Jan 19 07:04:56 2011 -0500

comments in QueryDsl


commit da64eb8b6ee84c0cf16e57dacd5f586b5b483580
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sun Jan 16 21:59:22 2011 -0500

refined the milliseconds tests, some DBs don't support them :
Oracle : http://forums.oracle.com/forums/thread.jspa?threadID=239634
MySql : http://bugs.mysql.com/bug.php?id=8523
MSSQL : http://stackoverflow.com/questions/2620627/ms-sql-datetime-precision-problem


commit 35f16ac6c5a4f5a1718954d9e9f2392e0a9b88c2
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sun Jan 16 14:03:41 2011 -0500

fix the test case for millisecond handling of some DBs,
the test should be eventually brokenen down between where clause
usage and
resultset handling. MSSQLServer, Oracle have problem with
milliseconds in
the where clause, MySQL has a problem mapping back the result
(down) to the
millisecond, all other DBs handles it correctly, DB2 remains to be
investigated.


commit 7420e30a00c98e6794adf7e91b10046a80093c2f
Merge: 3b5f670 74c7222
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sun Jan 16 13:34:24 2011 -0500

Merge branch 'master' of github.com:max-l/Squeryl

commit 3b5f67075140300c1ceaa4d86f6bf6bba6b9b66e
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sun Jan 16 13:33:22 2011 -0500

Added test for millisecond handing of timestamps


commit 74c72222e9c9e6278bc06902ff9a6d730ae0c4f0
Author: mrico <m.ric...@gmail.com>
Date: Thu Jan 13 07:37:18 2011 +0100

added Derby support

create mode 100644 src/main/scala/org/squeryl/adapters/
DerbyAdapter.scala

commit 513a2fcc9893289457d6651a1164d2ec5f908669
Merge: d30fe9e bc8c228
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sat Jan 8 07:34:32 2011 -0500

Merge branch 'aw_closeStatements' of git://github.com/max-l/Squeryl

commit d30fe9eb7aa93161e548293351d0e360e7853ef6
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Sat Jan 8 06:30:02 2011 -0500

inhibited testPartialUpdateWithInclusionOperator for MySQL, since
this DB doesn't support subqueries inside IN argument


commit 44a181cba9034ad312d556b7aa25fa2569041d50
Author: Peter Brant <peter...@gmail.com>
Date: Fri Jan 7 02:15:02 2011 +0100

Quote identifiers in emitted SQL

PostgreSQL converts all unquoted identifiers to lower case. For a
manually managed schema that uses mixed case table and column
names,
table and column references need to be quoted for queries to work.

The quoting scheme is customizable per database adapter.
Currently all
adapters save the PostgreSQL adapter do no identifier quoting
(i.e. they
maintain the old behavior).


commit 5dd819d188a3af81ed482475c1ffb4aa1684fbb9
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Thu Jan 6 20:08:22 2011 -0500

Fix for issue 89 : update(T)(t => where(t.id in (subselect))
set(...))" generates invalid SQL


commit bc8c22856fa36fec71cf389a653f3b8e40d7b5e1
Author: Alec Wysoker <al...@cm1e0-672.broadinstitute.org>
Date: Mon Dec 20 16:45:15 2010 -0500

Fix bug in Session.cleanupResources.
Add ability for Session to print info about unclosed Statements to
aid in fixing statement leaks.
Make the iterator returned by AbstractQuery also implement
Closeable.
Force iterator in View.lookup to be closed after use.


commit 4d7f8aa0bf60dc047f677329f780b2d972f9905c
Author: Yuen Ho Wong <wyu...@MobileCat.local>
Date: Sun Dec 19 03:57:15 2010 +0800

patch for issue 81


commit 0286ac918fd2e74f8f65e1cafe693e4d8f15f4ec
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Fri Dec 17 06:13:05 2010 -0500

IDEA tweaks


commit 2df7b92d1a8b10b8b5721343383e898524caa30e
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Fri Dec 17 06:12:02 2010 -0500

CompositeKey support now up to 9 columns


commit a148ad6109c1f6c17e1c19a6d3db8c7dcb3f5c78
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Thu Dec 16 10:30:34 2010 -0500

Fix for issue 79 (Returning or excepting out of nested transaction
block kills session)


commit 5c3ab25e55b63c04636e0873885e2d1d5c75c672
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Thu Dec 9 20:29:37 2010 -0500

String default lenght is now 128


commit f809ba1f285c46a8b5e9be27ced7f6bf08b955c9
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Thu Dec 9 20:00:57 2010 -0500

Bumped IDEA's project files to 2.8.1


commit 72662a032b3cee7aed1018dcc28bb5fe40fb0ad6
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Wed Dec 8 18:47:26 2010 -0500

Added findTablesFor


commit de0f297439d9556d02613990d7043fc411a6774b
Author: Maxime Lévesque <maxime....@gmail.com>
Date: Tue Nov 30 22:04:33 2010 -0500

eager closing of preparedStatement for DB updates


commit 75ced75995f4168843e992e4c4723847ac89bfc4
Author: Michael Gottschalk <MGott...@gmx.net>
Date: Mon Nov 29 21:01:48 2010 +0100

Removed accidentally added file

delete mode 100644 src/main/scala/org/squeryl/internals/#Utils.scala#

commit 97bd407dfa5f674c7c690239b3f9ebb32538aae7
Author: Michael Gottschalk <MGott...@gmx.net>
Date: Mon Nov 29 19:30:34 2010 +0100

Changed comment at the top of each file so that emacs no longer
recognizes the whole file as a comment

create mode 100644 src/main/scala/org/squeryl/internals/#Utils.scala#

[Marco Rico-Gomez]

+1

Issue #7 is a bit annoying but should be ok for a RC release ?!?

2011/2/2 Maxime Lévesque <maxime....@gmail.com>:

--
\ Marco
--
Twitter: @mricog
Website: http://mrico.eu

[Maxime Lévesque]

Marco, I have some good and bad news :
the bad : RC4 is already declared a dead end, there was a fix in it for Lift, that doesn't completely solves the problem it was intended to solve.
  (http://www.assembla.com/spaces/liftweb/tickets/876-record-squeryl-selecting-unspecified-columns-in-generated-sql)
the good : we have more time to fix #7 ;-)

I have the fix for Lift#876, but It's in a dangerous place, so it'll be a few days since I push it...