5210 Commits

Author SHA1 Message Date
Mark Struberg
6f721f69c9 [maven-release-plugin] prepare release 3.2.0 3.2.0 2021-05-10 19:19:33 +02:00
Mark Struberg
bdf3f0a543 OPENJPA-2833 upgrade assembly to contain xbean-asm-shaded 2021-05-10 18:37:24 +02:00
Mark Struberg
6546f259b2 further documentation fixes. up to JPA 2.2 2021-05-10 11:22:55 +02:00
Mark Struberg
eec95cd6cb wip update compat matrix 2021-05-09 00:07:40 +02:00
Mark Struberg
6eb984a375 add java.time Entity field types support 2021-05-09 00:07:12 +02:00
Mark Struberg
61540f00c4 reserved column words should be public
allow configuration
2021-05-09 00:06:13 +02:00
Mark Struberg
9c95b71fcc update docs to reflect changes in OpenJPA-3.2.0 2021-05-08 23:20:58 +02:00
Mark Struberg
8f79c8aba8 up to 2021 2021-05-08 22:50:50 +02:00
Mark Struberg
64d8253578 pimp release notes in preparation for our 3.2.0 release 2021-05-08 22:48:44 +02:00
Mark Struberg
f72f3ca7e0 OPENJPA-2873 add persistence_2_2.xsd handling 2021-05-08 22:39:43 +02:00
Mark Struberg
361da251a9 update docs for OpenJPA-3.2.0 2021-05-08 21:53:12 +02:00
Mark Struberg
271e3e50cf upgrade documentation for the next release 2021-05-08 17:18:33 +02:00
Mark Struberg
e3fd21f0a2 OPENJPA-2871 update to xbean-4.20 2021-05-08 13:58:55 +02:00
Mark Struberg
49116c7f8e OPENJPA-2816 correct jdbc url for herddb-docker 2021-05-08 13:53:49 +02:00
Mark Struberg
25af7c35ee OPENJPA-2849 proper handling of different Date types 2021-05-07 23:07:41 +02:00
Mark Struberg
477d73a996 OPENJPA-2816 fix herddb jdbc driver class name 2021-05-05 07:33:50 +02:00
Enrico Olivelli
f3c65f01ad Start HerdDB profile 2021-05-05 06:50:24 +02:00
Mark Struberg
6a7addea16 OPENJPA-2868 update invalid column names for PostgreSQL 2021-05-04 08:02:40 +02:00
Mark Struberg
cb4e47b82c OPENJPA-2816 had to revert dd9bce0cc9 due to side effect
Changing the delimitedFromDb handling had a negative side effect on PostgreSQL.
Postgres uses all lower internally, thus comparing those from the db to the generated
columns does not get detected as equal, which leads to wrong metadata.

We have to come back to HerdDB and also need to re-evaluate DBDict#delimitAll.
2021-05-03 23:29:48 +02:00
Mark Struberg
813154a570 OPENJPA-2861 fix unit test
Boolean is the correct return value.
This should actually have been returned for all dictionaries.
The problem is that we always only return the _internal_ representation.
The reason is because CASE/WHEN can be used to return values, but also as subquery.
Now imagine a database uses 0 and 1 for false and true. If CASE/WHEN is used
as subquery we really have to return 0/1 (number) because otherwise the WHERE clause
would not fit. When not executing the query on a Entity, we do not know the target type.
So there is probably no way we can later do a BooleanRepresentation call to switch to boolean.
And this would also break existing applications.
2021-05-03 17:06:22 +02:00
Mark Struberg
31d2d862b7 OPENJPA-2868 update invalid columNames for Oracle 2021-05-02 23:14:33 +02:00
Mark Struberg
8e18fdbbbc OPENJPA-2865 add compat check for WITH_TIMEZONE cols 2021-05-02 23:14:33 +02:00
Mark Struberg
c780084ccd OPENJPA-2870 update spec metadata to 2.2 2021-05-02 23:03:10 +02:00
Mark Struberg
f8667f564d OPENJPA-2868 reserved column names for HSQLDB 2021-05-02 22:22:00 +02:00
Mark Struberg
8e3b81521f switch back to dbcp for oracle as well 2021-05-02 21:54:05 +02:00
Mark Struberg
d78bfc47cb OPENJPA-2866 fix identifier unit test
Oracle now supports IDENTITY columns since a very long time.
No need to rely on ancient index creation stuff.
2021-05-02 21:53:12 +02:00
Mark Struberg
5dc56fa17b OPENJPA-2868 update invalidColumnNames for MariaDB 2021-05-02 17:14:58 +02:00
Mark Struberg
d8b3fd87ff OPENJPA-2868 add tested reserved word list for MS SQLServer
no list has previously been maintained.
2021-05-02 16:45:43 +02:00
Mark Struberg
b99e2dba29 OPENJPA-2868 fix invalidColumnNames for MySQL 2021-05-02 16:11:56 +02:00
Mark Struberg
8f959cc3a5 OPENJPA-2868 update reserved column names for H2 2021-05-02 15:37:06 +02:00
Mark Struberg
90ba3b8d5d fix Spec compat tests to work with all DBs
test should rely on whether KEY is a reserved word or not.
2021-05-02 15:36:05 +02:00
Mark Struberg
d8bb07fe86 OPENJPA-2868 update reserved column words for Derby 2021-05-02 14:31:17 +02:00
Mark Struberg
188ed235dc OPENJPA-2867 generated pastable java code
also add more reserved words
2021-05-02 14:11:41 +02:00
Mark Struberg
7d3c89c515 OPENJPA-2867 add a way to test reserved words 2021-05-02 12:30:34 +02:00
Mark Struberg
adf5ad393d update poms to 3.2.0-SNAPSHOT
as discussed on the list - the next release will be 3.2.0
2021-04-24 11:39:25 +02:00
Francesco Chicchiriccò
a850031d11 Revert "OPENJPA-2800 mark additional fields and methods as synthetic"
This reverts commit 5b67ce51801cb646d772dadd03a623312603a0df.
2021-04-21 12:13:46 +02:00
Mark Struberg
5b67ce5180 OPENJPA-2800 mark additional fields and methods as synthetic
When enhancing an Entity we add a few fields and additional methods.
Those elements should properly get marked as synthetic.
2021-04-19 21:13:07 +02:00
Mark Struberg
919154ce69 remove unused import 2021-04-18 23:19:17 +02:00
Mark Struberg
eb2dbec127 OPENJPA-2789 close connection after bulk delete
while ResultSetResult closes the underlying connection we did loose this handling in
JDBCStoreQuery when XROP sharding got added.
2021-04-18 23:05:50 +02:00
Mark Struberg
d5ac99d402 move back to Oracle11XE docker
the o18 docker image I did try was simply what we needed.
While the o11xe is not suitable for running the full test wuite
it is perfectly fine for running single tests.
2021-04-10 19:10:39 +02:00
Mark Struberg
7607653792 test needs Oracle specific create table stmt 2021-04-10 17:18:41 +02:00
Mark Struberg
3b51096dae OPENJPA-2866 no need to test JDBC2 feature
Oracle comes with JDBC4. We can safely assume that
Statement#getGeneratedKeys is always available.
Oracle relies on it and will fail without.
2021-04-10 17:06:59 +02:00
Mark Struberg
d6a64bebcd OPENJPA-2866 Oracle GenerationType#IDENTITY support
we will now create a column definition
"GENERATED ALWAYS AS IDENTITY"
for auto-increment columns.
2021-04-10 12:27:37 +02:00
Mark Struberg
2408ff7d9e tests which do not commit are not likely to fail... 2021-04-10 12:27:07 +02:00
Mark Struberg
bb214f93d4 fix TestQueryExcludingSubclasses sorting
The sorting behaviour of characters )'a..z, A..Z') and
numbers (0..9) is depending on NLS. For e.g. german NLS
in Oracle 0 comes only after z, so we get esub1,esub2,e1,e2
while on some other databases we get e1,e2,esub1,esub2.
Easy fix is to have the second position also a Character to
force a distinctive order over all different databases and
settings.
2021-04-10 11:39:57 +02:00
Mark Struberg
d6a19dd56a Oracle seems to have changed their error handling.
They now (tested with Oracle 18) behave more like other databases,
so we should validate whether one of the described error situations got
catched.
2021-04-10 11:11:21 +02:00
Mark Struberg
c89a94abd8 reduce loop count because that test takes 2 minutes 2021-04-10 11:10:51 +02:00
Mark Struberg
59a8586855 OPENJPA-2865 use java.time handling from JDBC driver 2021-04-09 14:26:51 +02:00
Mark Struberg
c7539a1171 new oracle docker image
The old OracleX11 was plagued with connection problems,
which seems to be a known problem with that version.
2021-04-09 14:22:53 +02:00
Mark Struberg
683e78045c OPENJPA-2863 use MICROS for TIMESTAMP precision in Oracle
Default type for Oracle is TIMESTAMP(6) but somehow we did only
round to 3 fraction digits in the past.
Can be tweaked to the old value via DBDictionary config in persistence.xml
2021-04-08 14:55:28 +02:00