No if one is doing with DB directly or with JDBC, the following will
work:
-- start transaction
INSERT INTO t_1 (name) VALUES ('name1');
INSERT INTO t_1 (name) VALUES ('name2');
INSERT INTO t_2 (name,t_1_id) select 'name1' , t_1.id from t_1 where
t_1.name='name1';
commit; -- end transaction
You might also take a look at the 3.0 lifecycles that Ari mentioned
in the issue, but my guess is that JPA lifecycles won't do what is
needed.
I did. To be inside the same transaction for writing, PrePersist would
be the logical choice
(from what I understood the docs). The very short document
I'd be interested in hearing what others have to say... feasible given
the current cayenne stack?
The problem is that the original poster is trying to use Cayenne like he
uses pure SQL and
No, I'm not.
Like I mentioned in earlier posts, I'm using ObjRelationships and all
the Cayenne goodie
Hm. You're defining a relationship; I've always mapped relationships aas
relationships, and let Cayenne deal with the id specifics. Note that
cayenne won't force the creation foreign key (although you should create
one, if your db supports it).
You say you have "too many tables like that"...
* write the record to database, fetch it back again and
then you'll have the primary key
Than this is not the same transaction :(.
Correct. This is a limitation of how databases work, not Cayenne. You
can't have a primary key until you write the record, unless you do
something tricky like u
Hi,
Thank you very much for the quick response.
> The 3.0 API was relaxed from 2.0. In 2.0, deleteObject took
> an org.apache.cayenne.Persistent. In 3.0, it takes any
> java.lang.Object.
I see...
> See:
> http://cayenne.apache.org/doc/api/org/apache/cayenne/access/DataContext.html#deleteObje
Trying to delete object entities with Cayenne 3M5 doesn't seem to work anymore
- it used to work with 2.0 :(.
I'm getting the error message:
java.lang.NoSuchMethodError:
org.apache.cayenne.access.DataContext.deleteObject(Lorg/apache/cayenne/Persistent;)V
Is there something different in 3.0 abo
> It sounds like you are trying to
> write a record into a log/audit
> table.
Yes, for some entities it is some sort of high level
audit, and for others it is a recent activity table.
> Have you looked at MySQL triggers to do
> this? The advantage
> of using a trigger is that even if someone c
> > Can't get primary key from temporary id
> >
> >
> > while trying to access the 'id' of a newly created
> object entity A.
> > I need this 'id' to write the value together with some
> > other fields in some other object entity B inside the same
> > transaction. Between A and B the
I'm getting the following error:
Can't get primary key from temporary id
while trying to access the 'id' of a newly created object entity A.
I need this 'id' to write the value together with some other fields in some
other object entity B inside the same transaction. Be
> > This would sound cool to me too, but from the Cayenne
> documentation example, this approach seems to have the
> problem that Java has single inheritance :(.
> >
> > If there's only one such table than it would work
> (single inheritance).
> > E.g. In the above example, by doing a superclass
> > http://cayenne.apache.org/doc/jar-files-and-dependencies.html
> > describes some of them but not all :(.
> >
> > What about?:
> > asm-3.0..jar
> > asm-commons-3.0.jar
> > geronimo-jpa_3.0_spec-1.0.jar ?
> >
> > Are they required when using a "pure cayenne" server,
> i.e. no JPA?
> > (they w
The "lib/third-party" contains several jar files required as dependency.
http://cayenne.apache.org/doc/jar-files-and-dependencies.html
describes some of them but not all :(.
What about?:
asm-3.0.jar
asm-commons-3.0.jar
geronimo-jpa_3.0_spec-1.0.jar ?
Are they required when using a "pure caye
Thank you very much for your reply.
> > Is it possible for Cayenne(and the Modeler) to map
> conditional relationships?
> >
> > I have a table, e.g. "tag" that relates to too many
> other tables, but one row is always related only to one
> table. This is done by having
> > instead of many foreign
> Is it possible for Cayenne(and the Modeler) to map
> conditional relationships?
>
> I have a table, e.g. "tag" that relates to too many other
> tables, but one row is always related only to one table.
> This is done by having
> instead of many foreign keys columns, just with 2 in a
> generic wa
Is it possible for Cayenne(and the Modeler) to map conditional relationships?
I have a table, e.g. "tag" that relates to too many other tables, but one row
is always related only to one table. This is done by having
instead of many foreign keys columns, just with 2 in a generic way:
- 'related_t
Sorry, but I'm a confused now :).
> It used to be true
and now it is no true anymore?
Or is this so jut for Cayenne3?
> that the reverse
> db relationship was required, but
> the reverse obj relationship was not required. From
> the application's
> point of view, it doesn't have to be there.
I'
Does Cayenne require the use of a reversed relationship always?
Of course they're practical, but in many schemes e.g. in the case of the 'user'
table, just too many other tables (~2/3 of them) point to it because of the
required "updated_by_user_id".
If reversed relationships would be required
> "Convention over Configuration" can be useful (and fast) in
> simple implementations (ex: JSP form assignments) but may
> still be a bit too "fuzzy" for the larger scaleable
> projects.
>
> These are my current opinions. I am interested in how
> others look at it.
The main issue is however tha
> I am curious: how are your
> related-tables managed in a relational database without
> FK's? Is there no RDB perspective in the design?
The ORM does it. e.g. ActiveRecord for Ruby on Rails applications, or other
ORMs/stacks that work on similar concepts (and there are quite many lately).
If I
> Just reread your question.. Sorry for
> misleading. This is true, Cayenne
> cannot know about relationships in DB without FKs. So you
> have to map them
> manually
Adding manually is not efficient for most cases (too big databases) :(.
(This would not convince users to ditch RoR's ActiveRecord f
> I don't reverse engineer often, but I
> believe if the constraints
> aren't defined in the DB, then the reverse engineering
> process has no
> way of knowing about the relationship,
Well, it could, based on the naming conventions: e.g.
"artist_id" from any table is a relation to "artists" table
> I've used Cayenne with a DB that
> didn't have FK constraints (the DB
> basically only had PK uniqueness constraints). Worked
> fine.
Nice. Good to know :).
> Just map
> things correctly in the modeler.
But the Cayenne Modeler doesn't recognize those relationships when reverse
engineering t
Can Cayenne work (consistently) with relationships where the underlying foreign
key constraint is missing? (e.g. when the fields for foreign key are there and
respect the standard naming convention - artist_id for the relationship
pointing to artists but the constraint is not set ?
I tried to
How to model/map files (with their content) e.g. in webapplications?
What is the best practice?
(1) To put their content as blob fully in the DB plus fields for
properties,
or
(2) to put the metadata about files only in the DB and the content
in a file system?
or
(3) other solution?
The sec
My preference (and it seems to be fairly efficient), is to continue with
the standard design pattern of storing metadata (file-path, etc) in the
database and storing the file in the filesystem. You can them make
programatic references much easier.
Thank you. I used too mostly this is the appro
How to model/map files (with their content) e.g. in webapplications?
What is the best practice?
(1) To put their content as blob fully in the DB plus fields for properties,
or
(2) to put the metadata about files only in the DB and the content
in a file system?
or
(3) other solution?
The secon
27 matches
Mail list logo
