fb259504c9
2007-08-16 20:26:12 +00:00
a794805a86
Another place where it's safest to be explicit about whence the column comes.
2007-08-16 20:24:15 +00:00
94766648e6
We have to carefully convert the user's query input to the boolean lexeme language TO_TSQUERY() expects. We may ultimately have to do more complicated things here, too.
2007-08-16 16:54:52 +00:00
cf8a977778
Add the primary key on columns, not fields.
2007-08-15 22:18:54 +00:00
d414292ac3
Let's not force validation on every Tom, Dick and Harry. This is a little
...
verbose, so maybe I'll change my mind again later and do something yet again
different.
2007-08-15 19:47:16 +00:00
a68a6534a4
More sanity checking.
2007-08-15 19:03:47 +00:00
0175081c54
2007-08-15 19:00:50 +00:00
8e13453362
Added some sanity checks to notice and catch programmer errors as early as
...
possible.
2007-08-15 18:44:35 +00:00
b884f89e80
If a persistent record is abstract, do not generate ColumnExp() constants for
...
it; if it extends another record, generate ColumnExp() constants for persistent
fields inherited from its parents.
2007-08-15 02:39:42 +00:00
a13756eec3
Make sure the record in question is not computed.
2007-08-15 01:16:02 +00:00
ab8ee79dbe
The logging output already contains the enclosing method name. So instead of
...
repeating the method name:
2007/08/14 16:07:52:157 INFO PersistenceContext.cacheStore: cacheStore: entry [key=...
we'll just say what we're doing:
2007/08/14 16:07:52:157 INFO PersistenceContext.cacheStore: storing [key=...
2007-08-14 23:18:22 +00:00
8e916489e0
Brevify our toString() output so that cache debug logging is readable.
2007-08-14 23:15:08 +00:00
34d1e73976
Return the string value of our cached value as our own.
2007-08-14 23:12:53 +00:00
589045232c
Add a mechanism for warning about records that are lazily initialized after an
...
attempt has been made to resolve all registered records.
2007-08-14 21:35:38 +00:00
0b91f0f6a7
Use IllegalArgumentException since those are basically illegal arguments. Also
...
use @exception because it's a nice way to let humans and tools know what's
going on.
2007-08-14 18:47:21 +00:00
f9757c3633
"List to Array to List to Set" becomes "to Set".
2007-08-14 18:37:40 +00:00
4b40140d2d
Tighten up a few places where we return null to callers who really never want to see nulls -- throw exceptions instead.
2007-08-14 17:40:00 +00:00
79024ea85a
Add dialect-agnostic auto-migration of full text indexes.
2007-08-14 17:00:26 +00:00
f80ce4d862
Log the creation of a full text search index on the MySQL side.
2007-08-14 16:09:23 +00:00
e5bf0d9d97
Better comments explaining the hairiest bits of auto-migration.
2007-08-14 15:41:36 +00:00
9edfe28d33
Tidied up SQL execution in liaisons; fixed MySQL column renaming.
2007-08-14 01:11:09 +00:00
3a99decfce
Rename @Index(columns) to @Index(fields) and do the proper translation to column name during table creation / migration.
2007-08-13 22:57:09 +00:00
632b8684be
Classes go after methods.
2007-08-13 21:59:25 +00:00
d72ad79705
Catch marshaller initialiation failure and report it more informatively.
2007-08-13 21:47:56 +00:00
f76ab3592a
More Zell changes:
...
- Re-read the table metadata after we run pre-default migrations, since those
are pretty much guaranteed to change the table.
- Changed EntityMigration.Rename to a pre-default rather than a post-default
migration.
- Made all the static EntityMigration subclasses use the generic liaison rather
than JDBCUtil.
2007-08-13 19:42:46 +00:00
f4dfccbf97
Patch from Zell to quell warnings about mismatched schemas for database-
...
specific auxiliary columns (specifically Postgres's FTS column).
2007-08-13 19:37:09 +00:00
9b87020fe5
Added getCacheAdapter().
2007-08-11 00:48:59 +00:00
f9b6414270
Patch from Zell to properly walk the inheritance tree and aggregate all
...
index-related annotations.
2007-08-08 23:58:43 +00:00
30681979ea
Nixed unused bits.
2007-08-08 23:49:15 +00:00
e21fbd9df1
From Zell: select text and binary column types based on the @Column(length)
...
annotation.
2007-08-08 23:47:50 +00:00
c7150cabdc
Small fix from Zell.
2007-08-07 18:05:49 +00:00
eb9c7c9c74
Reinstate automatic schema migration, per Zell.
2007-08-06 21:52:52 +00:00
8eb5d24be9
Full text index support from Zell, also a fix for deletion.
2007-07-31 21:36:02 +00:00
ddf40eb5c4
Always use an explicit query if we aren't using a cache.
2007-07-31 02:09:32 +00:00
92c54a5325
Synchronize on the connection while we're using it to cooperate with
...
SimpleRepository + StaticConnectionProvider.
2007-07-30 20:10:47 +00:00
b9d14f57c7
Missed a spot.
2007-07-26 19:13:57 +00:00
07666e16a6
Added PersistenceContext.shutdown(), exposed shutdown() in ConnectionProvider.
2007-07-26 19:13:34 +00:00
3628ec528d
Zell patch to factor caching code into pluggable CacheAdapter class. Allows us
...
to disable caching and/or replace it with less industrial strength system
during development.
2007-07-26 19:07:20 +00:00
cf4f640ab3
Deprecation begone!
2007-07-25 23:30:07 +00:00
9f544c9174
El Giganto Depot database dialects support from Zell. PostgresSQL here we
...
come. Automatic schema migration is currently disabled and remains to be
dialectified.
2007-07-25 20:53:40 +00:00
f0bae6ff8d
Header updates.
2007-07-13 00:20:09 +00:00
68d9be57eb
@Override is our friend. It helps to keep us warm and safe.
2007-07-12 21:50:36 +00:00
083cb788d8
Another Zell Depot patch:
...
- refactoring query construction;
- modified collection lookup to first look up the matching primary keys, then
check the cache for hits then look up the remainder from the database;
- other small renamery and cleanup.
2007-07-12 20:02:54 +00:00
e472fc6de8
Check that our schema does not contain stale columns and issue a warning if it
...
does.
2007-06-29 00:25:40 +00:00
5b845c0ca7
2007-06-18 23:14:26 +00:00
15f3d6be4d
Add a delete() that takes a Key.
2007-06-15 01:13:28 +00:00
41acbbdeca
2007-06-12 22:30:48 +00:00
2532edba80
We need to initialize migrations here as well.
2007-06-02 00:07:41 +00:00
9de9835b52
Patch from Zell to fix cache keys.
2007-06-01 19:53:09 +00:00
ffa1f10e6b
Oops, we were using _result to pass the record into the modifier in the first
...
place. We'll continue to do that and clear out _result during invoke() which is
a wee bit clearer anyway.
2007-05-23 18:20:08 +00:00
Par Winzell