hibernate-devel — development

Re: [Hibernate] Delete bug (and fix)

[Gavin K. <ga...@ap...>]

The reason it doesn't check list!=null is that find() is never supposed to
return a null list.... How did you make that happen?? Do you have a
stacktrace, etc?


> I've found a bug that causes a NPE in the Session.delete(String query)
> implementation: if you pass a query that return no objects you end up
> with an NPE. The incriminated code follows:
>
> SessionImpl.java (line 1178)
>
> public int delete(String query, Object[] values, Type[] types) throws
> HibernateException, SQLException {
>
> if ( log.isTraceEnabled() ) {
> log.trace( "delete: " + query );
> if (values.length!=0)  log.trace( "parameters: " +
> StringHelper.toString(values) );
> }
>
> List list = find(query, values, types);
> int size = list.size();
> for ( int i=0; i<size; i++ ) delete( list.get(i) );
> return size;
> }
>
> The list.size() call fails if list is null... should look like:
>
> if(list != null)  {
> int size = list.size();
> for ( int i=0; i<size; i++ ) delete( list.get(i) );
> return size;
> } else {
> return 0;
> }
>
> Bye
> Andrea
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:
> Access Your PC Securely with GoToMyPC. Try Free Now
> https://www.gotomypc.com/s/OSND/DD
> _______________________________________________
> hibernate-devel mailing list
> hib...@li...
> https://lists.sourceforge.net/lists/listinfo/hibernate-devel

[Hibernate] Delete bug (and fix)

[Andrea A. <aa...@li...>]

I've found a bug that causes a NPE in the Session.delete(String query)=20
implementation: if you pass a query that return no objects you end up
with an NPE. The incriminated code follows:

SessionImpl.java (line 1178)

public int delete(String query, Object[] values, Type[] types) throws=20
HibernateException, SQLException {

		if ( log.isTraceEnabled() ) {
			log.trace( "delete: " + query );
			if (values.length!=3D0)  log.trace( "parameters: " +=20
StringHelper.toString(values) );
		}

		List list =3D find(query, values, types);
		int size =3D list.size();
		for ( int i=3D0; i<size; i++ ) delete( list.get(i) );
		return size;
	}

The list.size() call fails if list is null... should look like:

	if(list !=3D null)  {
		int size =3D list.size();
		for ( int i=3D0; i<size; i++ ) delete( list.get(i) );
			return size;
	} else {
		return 0;
	}

Bye
Andrea

[Hibernate] Session local interface...

[Max R. A. <ma...@eo...>]

Hi!

Just wondering if anyone have discovered that I uploaded a small patch for
having
Session local interceptors. I think it is worthwile having the option of
session local interceptors,
otherwise one would have to go through loops to support session-specific
interceptors.

/max

[Hibernate] cvs commits notification ?

[Max R. A. <ma...@eo...>]

Hi!

Is there somewhere one can signup for being notified whenever there are
some commits to the cvs repository ? (I expect this is some option on
sf.net, but i just can't find it )

/max

[Hibernate] Re: Jmx error in 1.1.5

[Gavin K. <ga...@ap...>]

Ah. heh. yeah thats a misprint, isnt it. Thanks

----- Original Message -----
From: "Andrea Aime" <aa...@li...>
To: "Gavin King" <ga...@ap...>
Sent: Sunday, October 20, 2002 1:57 AM
Subject: Re: Jmx error in 1.1.5


> Because the line should be:
>
> this.mapResources = mappingResources.trim();
>
> mapResources has not been initialized so it's null and we are
> calling a trim on it... and we are discarding the mappingResources
> parameter...
> Andrea
>
> > Yick! Sorry, I actually wasn't fully aware of this bug.....
> >
> > Are you suggesting the problem is here:
> >
> >
> > public void setMapResources(String mappingResources) {
> >
> > this.mapResources = mapResources.trim();
> >
> > }
> >
> >
> >
> > if so, why exactly is setMapResources() being called with a null
argument.
> > I'm confused....
> >
> >
> > ----- Original Message -----
> > From: "Andrea Aime" <aa...@li...>
> > To: "Gavin King" <ga...@ap...>
> > Sent: Sunday, October 20, 2002 1:43 AM
> > Subject: Jmx error in 1.1.5
> >
> > > Hi Gavin,
> > > the error in cirrus/hibernate/jmx/HibernateService.java
> >
> > (setMapResources.java)
> >
> > > that prevents the use of Hibernate as a jmx service is still there in
> >
> > 1.1.5,
> >
> > > thought it was corrected in cvs some time ago... this is particularly
> > > frustrating since someone trying to deploy hibernate as a service will
> >
> > only
> >
> > > see a nullPointerException again and again (and wondering what he did
> > > wrong...). Hope you can fix it and put out a 1.1.5b...
> > > Best regards
> > > Andrea Aime

[Hibernate] Re: Jmx error in 1.1.5

[Gavin K. <ga...@ap...>]

Yick! Sorry, I actually wasn't fully aware of this bug.....

Are you suggesting the problem is here:


public void setMapResources(String mappingResources) {

this.mapResources = mapResources.trim();

}



if so, why exactly is setMapResources() being called with a null argument.
I'm confused....


----- Original Message -----
From: "Andrea Aime" <aa...@li...>
To: "Gavin King" <ga...@ap...>
Sent: Sunday, October 20, 2002 1:43 AM
Subject: Jmx error in 1.1.5


> Hi Gavin,
> the error in cirrus/hibernate/jmx/HibernateService.java
(setMapResources.java)
> that prevents the use of Hibernate as a jmx service is still there in
1.1.5,
> thought it was corrected in cvs some time ago... this is particularly
> frustrating since someone trying to deploy hibernate as a service will
only
> see a nullPointerException again and again (and wondering what he did
> wrong...). Hope you can fix it and put out a 1.1.5b...
> Best regards
> Andrea Aime

Re: [Hibernate] ClobType

[Gavin K. <ga...@ap...>]

 
> Gavin, I'm not really sure what you meant by a Holder class that you 
> mentioned in a previous e-mail.  Would you care to go into details?


I don't have a set idea of what we need here, except that:

Use of java.sql.Clob class is non-useful, since it provides no way to
update the data. setCharacterStream() / getCharacterStream() should 
therefore be the methods we use to interact with the JDBC 
PreparedStatement  / ResultSet

Re: [Hibernate] ClobType

[Mark W. <mor...@SM...>]

Gavin King wrote:

>Yes! I like this idea a lotfully. Good thinking :)
>

Ok, unless someone else is working on this, I'm going to try to hammer 
something out that'll handle Clobs nicely over the weekend.  Any 
advice/suggestions are more than welcome.

Gavin, I'm not really sure what you meant by a Holder class that you 
mentioned in a previous e-mail.  Would you care to go into details?

-Mark

>----- Original Message -----
>From: "Jon Lipsky" <jon...@xe...>
>
>  
>
>>Shouldn't it be possible to create wrapper reader/writer's for the real
>>reader/writers that implement some kind dirty checking?  Perhaps the
>>    
>>
>wrapper
>  
>
>>reader/writer to read and write to an and from a char[] arrays, and then
>>when we really to to persist the Clob, the wrapper could then write the
>>char[] to the real writer.?.?
>>    
>>

Re: [Hibernate] Automatically determining dialect

[Chris N. <ch...@si...>]

On Fri, Oct 18, 2002 at 10:57:15AM +0200, Christoph Sturm wrote:
> I think its a good idea because every feature that makes hibernate more
> auto-configuring is a good feature. I wanted to work on the dialects anyway,
> so I will take a look at this too if its ok for the others too :)

It might also be nice if you could specify database-dependent
properties, which will only get picked up if the corresponding dialect
is in use. For example, something like

hibernate.connection.url.mysql = jdbc:mysql:///test
hibernate.connection.url.postgresql = jdbc:postgresql:template1

Maybe there is a better syntax.

Then, the only thing you'd *have* to change would be the driver_class.
Don't know a way around that one.

-Chris

Re: [Hibernate] Automatically determining dialect

[Gavin K. <ga...@ap...>]

Do it :)

----- Original Message -----
From: "Christoph Sturm" <ch...@mc...>
To: "Chris Nokleberg" <ch...@si...>;
<hib...@li...>
Sent: Friday, October 18, 2002 6:57 PM
Subject: Re: [Hibernate] Automatically determining dialect


> I think its a good idea because every feature that makes hibernate more
> auto-configuring is a good feature. I wanted to work on the dialects
anyway,
> so I will take a look at this too if its ok for the others too :)
>
> -chris
> ----- Original Message -----
> From: "Chris Nokleberg" <ch...@si...>
> To: <hib...@li...>
> Sent: Friday, October 18, 2002 8:37 AM
> Subject: [Hibernate] Automatically determining dialect
>
>
> > I'm trying to reduce the number of properties required to switching
> > between databases. I think hibernate could automatically determine the
> > proper dialect once it is connected to the database, via
> >
> >   conn.getMetaData().getDatabaseProductName()
> >
> > It doesn't appear that the dialect needs to be known prior to
> > connection, but I could easily be wrong about that.
> >
> > Another issue is that not all databases may have full jdbc metadata
> > support, but this would be easy enough to find out. Thoughts?
> >
> > -Chris
> >
> >
> > -------------------------------------------------------
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > hibernate-devel mailing list
> > hib...@li...
> > https://lists.sourceforge.net/lists/listinfo/hibernate-devel
> >
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> hibernate-devel mailing list
> hib...@li...
> https://lists.sourceforge.net/lists/listinfo/hibernate-devel

Re: [Hibernate] Automatically determining dialect

[Christoph S. <ch...@mc...>]

I think its a good idea because every feature that makes hibernate more
auto-configuring is a good feature. I wanted to work on the dialects anyway,
so I will take a look at this too if its ok for the others too :)

-chris
----- Original Message -----
From: "Chris Nokleberg" <ch...@si...>
To: <hib...@li...>
Sent: Friday, October 18, 2002 8:37 AM
Subject: [Hibernate] Automatically determining dialect


> I'm trying to reduce the number of properties required to switching
> between databases. I think hibernate could automatically determine the
> proper dialect once it is connected to the database, via
>
>   conn.getMetaData().getDatabaseProductName()
>
> It doesn't appear that the dialect needs to be known prior to
> connection, but I could easily be wrong about that.
>
> Another issue is that not all databases may have full jdbc metadata
> support, but this would be easy enough to find out. Thoughts?
>
> -Chris
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> hibernate-devel mailing list
> hib...@li...
> https://lists.sourceforge.net/lists/listinfo/hibernate-devel
>

Re: [Hibernate] hibernate background?

[Brad C. <bra...@wo...>]

>Do you know was hibernate
> ever deployed in a larger commercial project? Say over
> 50 persistable classes?

we use hibernate extensively in our online content management system.  the
only jdbc in the whole application is for streaming blobs.

> Have you considered to make it a jakarta project?

there is already a somewhat similar project there:
http://jakarta.apache.org/ojb/.  while they are similar in the sense that
they are object/relational mappers, they have differenct features and
architectures.  i have never used OBJ so i can't speak for it, but i have
been very happy with hibernate's functionality, simplicity (read: ease of
use) and robustness.

> The second option I am looking at is jboss v3 and
> using their CMP 2.0 persistence. Jboss is a valid
> option because I can run most of the code efficiently
> from within the conatainer and only send the results
> back to the clients when ready. How does hibernate
> compare to jboss CMP2.0 persistence? Do you know?

O/R mapping and CMP are quite different.  off the top of my head, i prefer
O/R mapping because:
1. it lends itself to more agile development
2. is is better suited to applications that model their problem domain in a
very OO manner

hibernate works well in stateless session beans within jboss (and probably
any other j2ee app server) with container managed transactions.

brad

[Hibernate] Automatically determining dialect

[Chris N. <ch...@si...>]

I'm trying to reduce the number of properties required to switching
between databases. I think hibernate could automatically determine the
proper dialect once it is connected to the database, via

  conn.getMetaData().getDatabaseProductName()

It doesn't appear that the dialect needs to be known prior to
connection, but I could easily be wrong about that. 

Another issue is that not all databases may have full jdbc metadata
support, but this would be easy enough to find out. Thoughts?

-Chris

Re: [Hibernate] dynamic schema update

[Christoph S. <ch...@sc...>]

I will look at it when I have the time. I think they dont have alter 
table support too. If they have a nice structure we could add the stuff 
we need and submit it back. There are so many java persistency 
frameworks, and it would be nice if they would share some code that is 
common to them all, and sql generation is for sure.

greetz
 chris


Gavin King wrote:

>Hmmmm the only trouble with this is that they support *way* fewer platforms
>than us.
>
>Also, can you tell if there is support for subqueries in the API?
>
>----- Original Message -----
>From: "Gavin King" <ga...@ap...>
>To: "Christoph Sturm" <ch...@sc...>
>Cc: "hibernate list" <hib...@li...>
>Sent: Friday, October 18, 2002 5:34 AM
>Subject: Re: [Hibernate] dynamic schema update
>
>
>  
>
>>Yeah, funny, I looked at this project about a year ago in the early stages
>>of Hibernate but it didnt seem far enough advanced to actually use.
>>    
>>
>Looking
>  
>
>>pretty good now.
>>
>>----- Original Message -----
>>From: "Christoph Sturm" <ch...@sc...>
>>To: "Gavin King" <ga...@ap...>
>>Cc: "hibernate list" <hib...@li...>
>>Sent: Friday, October 18, 2002 5:30 AM
>>Subject: Re: [Hibernate] dynamic schema update
>>
>>
>>    
>>
>>>Gavin King wrote:
>>>
>>>      
>>>
>>>>>I have now added a Schemaupdater to hibernate. At the moment it only
>>>>>adds missing colums to the schema, and creates tables if they are
>>>>>missing. I have testet it with sql server. All the rest of hibernate
>>>>>seems to still work, so I could commit it to cvs if you have no
>>>>>
>>>>>
>>>>>          
>>>>>
>>>>objections.
>>>>
>>>>Cool!! Go Ahead and check it in. Sounds like something that will really
>>>>be used by a lot of users....
>>>>
>>>>
>>>>
>>>>
>>>>        
>>>>
>>>>>Something we should really do in the near future is create classes for
>>>>>all the sql statements. I think we should think of our database
>>>>>          
>>>>>
>dialects
>  
>
>>>>>as SQL-Statementfactory-factories, then we could deal with different
>>>>>          
>>>>>
>sql
>  
>
>>>>>dialects much easier.
>>>>>
>>>>>
>>>>>          
>>>>>
>>>>Yes, you are absolutely right (and I have mentioned this before). The
>>>>        
>>>>
>>main
>>    
>>
>>>>remaining piece of bad-design is the lack of a proper OO model of SQL.
>>>>If you want to have a go at writing an elegant heirarchy of classes for
>>>>this,
>>>>it would be _highly_ appreciated.
>>>>
>>>>
>>>>        
>>>>
>>>i wonder if
>>>http://sourceforge.net/projects/crossdb/ would be an option. Just saw
>>>that the ojb guys use this.
>>>
>>>regards
>>> chris
>>>      
>>>
>>
>>-------------------------------------------------------
>>This sf.net email is sponsored by: viaVerio will pay you up to
>>$1,000 for every account that you consolidate with us.
>>http://ad.doubleclick.net/clk;4749864;7604308;v?
>>http://www.viaverio.com/consolidator/osdn.cfm
>>_______________________________________________
>>hibernate-devel mailing list
>>hib...@li...
>>https://lists.sourceforge.net/lists/listinfo/hibernate-devel
>>    
>>
>
>
>
>-------------------------------------------------------
>This sf.net email is sponsored by: viaVerio will pay you up to
>$1,000 for every account that you consolidate with us.
>http://ad.doubleclick.net/clk;4749864;7604308;v?
>http://www.viaverio.com/consolidator/osdn.cfm
>_______________________________________________
>hibernate-devel mailing list
>hib...@li...
>https://lists.sourceforge.net/lists/listinfo/hibernate-devel
>
>
>  
>

Re: [Hibernate] dynamic schema update

[Gavin K. <ga...@ap...>]

Hmmmm the only trouble with this is that they support *way* fewer platforms
than us.

Also, can you tell if there is support for subqueries in the API?

----- Original Message -----
From: "Gavin King" <ga...@ap...>
To: "Christoph Sturm" <ch...@sc...>
Cc: "hibernate list" <hib...@li...>
Sent: Friday, October 18, 2002 5:34 AM
Subject: Re: [Hibernate] dynamic schema update


> Yeah, funny, I looked at this project about a year ago in the early stages
> of Hibernate but it didnt seem far enough advanced to actually use.
Looking
> pretty good now.
>
> ----- Original Message -----
> From: "Christoph Sturm" <ch...@sc...>
> To: "Gavin King" <ga...@ap...>
> Cc: "hibernate list" <hib...@li...>
> Sent: Friday, October 18, 2002 5:30 AM
> Subject: Re: [Hibernate] dynamic schema update
>
>
> > Gavin King wrote:
> >
> > >>I have now added a Schemaupdater to hibernate. At the moment it only
> > >>adds missing colums to the schema, and creates tables if they are
> > >>missing. I have testet it with sql server. All the rest of hibernate
> > >>seems to still work, so I could commit it to cvs if you have no
> > >>
> > >>
> > >objections.
> > >
> > >Cool!! Go Ahead and check it in. Sounds like something that will really
> > >be used by a lot of users....
> > >
> > >
> > >
> > >
> > >>Something we should really do in the near future is create classes for
> > >>all the sql statements. I think we should think of our database
dialects
> > >>as SQL-Statementfactory-factories, then we could deal with different
sql
> > >>dialects much easier.
> > >>
> > >>
> > >
> > >
> > >Yes, you are absolutely right (and I have mentioned this before). The
> main
> > >remaining piece of bad-design is the lack of a proper OO model of SQL.
> > >If you want to have a go at writing an elegant heirarchy of classes for
> > >this,
> > >it would be _highly_ appreciated.
> > >
> > >
> > i wonder if
> > http://sourceforge.net/projects/crossdb/ would be an option. Just saw
> > that the ojb guys use this.
> >
> > regards
> >  chris
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: viaVerio will pay you up to
> $1,000 for every account that you consolidate with us.
> http://ad.doubleclick.net/clk;4749864;7604308;v?
> http://www.viaverio.com/consolidator/osdn.cfm
> _______________________________________________
> hibernate-devel mailing list
> hib...@li...
> https://lists.sourceforge.net/lists/listinfo/hibernate-devel

Re: [Hibernate] dynamic schema update

[Gavin K. <ga...@ap...>]

Yeah, funny, I looked at this project about a year ago in the early stages
of Hibernate but it didnt seem far enough advanced to actually use. Looking
pretty good now.

----- Original Message -----
From: "Christoph Sturm" <ch...@sc...>
To: "Gavin King" <ga...@ap...>
Cc: "hibernate list" <hib...@li...>
Sent: Friday, October 18, 2002 5:30 AM
Subject: Re: [Hibernate] dynamic schema update


> Gavin King wrote:
>
> >>I have now added a Schemaupdater to hibernate. At the moment it only
> >>adds missing colums to the schema, and creates tables if they are
> >>missing. I have testet it with sql server. All the rest of hibernate
> >>seems to still work, so I could commit it to cvs if you have no
> >>
> >>
> >objections.
> >
> >Cool!! Go Ahead and check it in. Sounds like something that will really
> >be used by a lot of users....
> >
> >
> >
> >
> >>Something we should really do in the near future is create classes for
> >>all the sql statements. I think we should think of our database dialects
> >>as SQL-Statementfactory-factories, then we could deal with different sql
> >>dialects much easier.
> >>
> >>
> >
> >
> >Yes, you are absolutely right (and I have mentioned this before). The
main
> >remaining piece of bad-design is the lack of a proper OO model of SQL.
> >If you want to have a go at writing an elegant heirarchy of classes for
> >this,
> >it would be _highly_ appreciated.
> >
> >
> i wonder if
> http://sourceforge.net/projects/crossdb/ would be an option. Just saw
> that the ojb guys use this.
>
> regards
>  chris

Re: [Hibernate] dynamic schema update

[Christoph S. <ch...@sc...>]

Gavin King wrote:

>>I have now added a Schemaupdater to hibernate. At the moment it only
>>adds missing colums to the schema, and creates tables if they are
>>missing. I have testet it with sql server. All the rest of hibernate
>>seems to still work, so I could commit it to cvs if you have no
>>    
>>
>objections.
>
>Cool!! Go Ahead and check it in. Sounds like something that will really
>be used by a lot of users....
>
>
>  
>
>>Something we should really do in the near future is create classes for
>>all the sql statements. I think we should think of our database dialects
>>as SQL-Statementfactory-factories, then we could deal with different sql
>>dialects much easier.
>>    
>>
>
>
>Yes, you are absolutely right (and I have mentioned this before). The main
>remaining piece of bad-design is the lack of a proper OO model of SQL.
>If you want to have a go at writing an elegant heirarchy of classes for
>this,
>it would be _highly_ appreciated.
>  
>
i wonder if
http://sourceforge.net/projects/crossdb/ would be an option. Just saw 
that the ojb guys use this.

regards
 chris

Re: [Hibernate] Suggestion: move cirrus into a src directory

[Gavin K. <ga...@ap...>]

> Where does the name cirrus actually come from ? (anything to do with
cirrus
> logic ??)

My previous employer:

http://cirrustech.com.au

I ditched the ".com.au" bit because it seemed a bit parochial.....

> The full CVS history can actually be handled by physically moving the
cirrus
> directory into a src directory on
> the actual cvs server (All the history is in the .v files, and CVS will
just
> return all directories found in the root.
> (see http://www.gnu.org/manual/cvs-1.9/html_node/cvs_69.html  for the gory
> details :)

Yeah, unfortunately I don't have access to the actual CVS server ;) I could
post a sourceforge support request, I suppose.....

> What may it be changed into instead ? net.sf.hibernate, or ?

Yeah maybe. I have been indecisive about this. Hence nothing has got done.
If
people start shouting at me loud enough with a  particular suggestion, maybe
that'll wake me out of this complacency ;)

I must admit, "net.sf" don't sound so bad ....

Re: [Hibernate] Suggestion: move cirrus into a src directory

[Max R. A. <ma...@eo...>]

Ok, great :)

3 things...

Where does the name cirrus actually come from ? (anything to do with cirrus
logic ??)

The full CVS history can actually be handled by physically moving the cirrus
directory into a src directory on
the actual cvs server (All the history is in the .v files, and CVS will just
return all directories found in the root.
(see http://www.gnu.org/manual/cvs-1.9/html_node/cvs_69.html  for the gory
details :)

What may it be changed into instead ? net.sf.hibernate, or ?

/max

----- Original Message -----
From: "Gavin King" <ga...@ap...>
To: "Max Rydahl Andersen" <ma...@eo...>
Cc: "hibernate list" <hib...@li...>
Sent: Thursday, October 17, 2002 8:19 PM
Subject: Re: [Hibernate] Suggestion: move cirrus into a src directory


> I plan to do this when/if the packages get renamed to something other than
> cirrus.hibernate.*. For now, its very useful to have the full CVS history
> for the files......
>
> ----- Original Message -----
> From: "Max Rydahl Andersen" <ma...@eo...>
> To: <hib...@li...>
> Sent: Thursday, October 17, 2002 7:56 PM
> Subject: [Hibernate] Suggestion: move cirrus into a src directory
>
>
> > Hi!
> >
> > I was just wondering why you have placed the .java files package
directory
> > structure in the root of the
> > Hibernate directory. Why not have a seperate src directory for it ?
> >
> > The current setup confuses e.g. eclipse especically if you build
hibernate
> > via ant - then it starts
> > to see the build directory as a set of classes too :)
> >
> > /max
> >
> >
> >
> > -------------------------------------------------------
> > This sf.net email is sponsored by: viaVerio will pay you up to
> > $1,000 for every account that you consolidate with us.
> > http://ad.doubleclick.net/clk;4749864;7604308;v?
> > http://www.viaverio.com/consolidator/osdn.cfm
> > _______________________________________________
> > hibernate-devel mailing list
> > hib...@li...
> > https://lists.sourceforge.net/lists/listinfo/hibernate-devel
>
>

Re: [Hibernate] hibernate background?

[Gavin K. <ga...@ap...>]

 
> Since hibernate is so clever I
> am wondering what made you write it? 

Frustration with CMP entity beans. urrrghhhh. And 
memory of something vaguely similar that I once
used in SmallTalk.


> There must have
> been a lot of effort put into it. What's the future of
> this project? 

Global Domination, of course :)


> How many people work and contribute to
> it? 

Check the changelog, most contributions from other
people are mentioned explicitly. Also check the mailing 
list archives.


> Have you considered to make it a jakarta project?

Not really seriously. I prefer LGPL and, to be honest,
the status of a lot of the Jakarta subprojects is .... 
hmmm... difficult to judge. Anyway, OJB moved to the 
Jakarta umbrella recently.


> The second option I am looking at is jboss v3 and
> using their CMP 2.0 persistence. Jboss is a valid
> option because I can run most of the code efficiently
> from within the conatainer and only send the results
> back to the clients when ready. How does hibernate
> compare to jboss CMP2.0 persistence? Do you know?

Havn't used JBoss CMP, but since quite a few of our 
users are JBoss users, and from what I know of EJB2,
I would assume that Hibernate is a much more 
sophisticated O/R tool. The basic problem with the
EJB spec is it is trying to do too many things and does
nothing well.

Re: [Hibernate] dynamic schema update

[Gavin K. <ga...@ap...>]

> I have now added a Schemaupdater to hibernate. At the moment it only
> adds missing colums to the schema, and creates tables if they are
> missing. I have testet it with sql server. All the rest of hibernate
> seems to still work, so I could commit it to cvs if you have no
objections.

Cool!! Go Ahead and check it in. Sounds like something that will really
be used by a lot of users....


> Something we should really do in the near future is create classes for
> all the sql statements. I think we should think of our database dialects
> as SQL-Statementfactory-factories, then we could deal with different sql
> dialects much easier.


Yes, you are absolutely right (and I have mentioned this before). The main
remaining piece of bad-design is the lack of a proper OO model of SQL.
If you want to have a go at writing an elegant heirarchy of classes for
this,
it would be _highly_ appreciated.


Thanks

Gavin

[Hibernate] dynamic schema update

[Christoph S. <ch...@sc...>]

Hi All!

I have now added a Schemaupdater to hibernate. At the moment it only 
adds missing colums to the schema, and creates tables if they are 
missing. I have testet it with sql server. All the rest of hibernate 
seems to still work, so I could commit it to cvs if you have no objections.

Something we should really do in the near future is create classes for 
all the sql statements. I think we should think of our database dialects 
as SQL-Statementfactory-factories, then we could deal with different sql 
dialects much easier.

regards
 chris

Re: [Hibernate] ClobType

[Gavin K. <ga...@ap...>]

Yes! I like this idea a lotfully. Good thinking :)

----- Original Message -----
From: "Jon Lipsky" <jon...@xe...>
To: "Gavin King" <ga...@ap...>
Cc: <hib...@li...>
Sent: Thursday, October 17, 2002 3:58 PM
Subject: Re: [Hibernate] ClobType


> Shouldn't it be possible to create wrapper reader/writer's for the real
> reader/writers that implement some kind dirty checking?  Perhaps the
wrapper
> reader/writer to read and write to an and from a char[] arrays, and then
> when we really to to persist the Clob, the wrapper could then write the
> char[] to the real writer.?.?
>
>
> ----- Original Message -----
> From: "Gavin King" <ga...@ap...>
> To: "Mark Woon" <mor...@SM...>
> Cc: <hib...@li...>
> Sent: Thursday, October 17, 2002 7:35 AM
> Subject: Re: [Hibernate] ClobType
>
>
> > The fact that I don't know what I think is the main reason I've never
> added
> > a ClobType before. I'm assuming that the best way would be to somehow
> > provide access to the JDBC Clob object (and similarly for Blobs). Is
this
> > possible? Do you need to provide a seperate Holder class?
> >
> > I like the reader/writer idea but the trouble with that is how on earth
do
> > you do dirty checking? Hibernate can't compare the reader to the writer
to
> > see if it changed.....
> >
> > ----- Original Message -----
> > From: "Mark Woon" <mor...@SM...>
> > To: "Hibernate Mailing List" <hib...@li...>
> > Sent: Wednesday, October 16, 2002 6:11 PM
> > Subject: [Hibernate] ClobType
> >
> >
> > > Gavin,
> > >
> > > I know I've reported that Hibernate works with CLOB's, but I'm afraid
> > > that I was wrong.  Hibernate persists Strings into CLOB columns
> > > perfectly fine, but reading it out of the db yields a null.  I've
added
> > > a new ClobType to handle CLOB columns to treat them as Strings.  I've
> > > attached a diff and the new file.
> > >
> > > I'm not sure that this is the right thing to do, because some may want
> > > direct access to the Clob object, or would prefer a Reader (in which
> > > case the setter ought to be a Writer), which would be a lot more
> > > complicated.  If we go with the former, Hibernate would need to
> > > implement it's own Clob object.  What do you think?
> > >
> > > -Mark
> > >
> > >
> >
> >
>
> --------------------------------------------------------------------------
> --
> > ----
> >
> >
> > > ? build
> > > ? patch.diff
> > > ? cirrus/hibernate/type/ClobType.java
> > > ? lib/j2ee.jar
> > > ? lib/junit.jar
> > > Index: cirrus/hibernate/Hibernate.java
> > > ===================================================================
> > > RCS file:
/cvsroot/hibernate/Hibernate/cirrus/hibernate/Hibernate.java,v
> > > retrieving revision 1.59
> > > diff -c -r1.59 Hibernate.java
> > > *** cirrus/hibernate/Hibernate.java 5 Oct 2002 04:18:30 -0000 1.59
> > > --- cirrus/hibernate/Hibernate.java 16 Oct 2002 08:02:56 -0000
> > > ***************
> > > *** 50,55 ****
> > > --- 50,59 ----
> > >   */
> > >   public static final NullableType STRING = new StringType();
> > >   /**
> > > + * Hibernate <tt>clob</tt> type
> > > + */
> > > + public static final NullableType CLOB = new ClobType();
> > > + /**
> > >   * Hibernate <tt>time</tt> type
> > >   */
> > >   public static final NullableType TIME = new TimeType();
> > > Index: cirrus/hibernate/impl/SessionImpl.java
> > > ===================================================================
> > > RCS file:
> > /cvsroot/hibernate/Hibernate/cirrus/hibernate/impl/SessionImpl.java,v
> > > retrieving revision 1.128
> > > diff -c -r1.128 SessionImpl.java
> > > *** cirrus/hibernate/impl/SessionImpl.java 15 Oct 2002 16:04:05 -0000
> > 1.128
> > > --- cirrus/hibernate/impl/SessionImpl.java 16 Oct 2002 08:03:04 -0000
> > > ***************
> > > *** 1002,1008 ****
> > >   }
> > >   else if ( old!=null ) {
> > >   throw new HibernateException(
> > > ! "Another object was associated with this id (the object with the
given
> > id was already loaded)"
> > >   );
> > >   }
> > >
> > > --- 1002,1009 ----
> > >   }
> > >   else if ( old!=null ) {
> > >   throw new HibernateException(
> > > ! "Another object of " + object.getClass().toString() +
> > > !                                 " has already been loaded with the
id
> "
> > + id
> > >   );
> > >   }
> > >
> > > Index: cirrus/hibernate/sql/OracleDialect.java
> > > ===================================================================
> > > RCS file:
> > /cvsroot/hibernate/Hibernate/cirrus/hibernate/sql/OracleDialect.java,v
> > > retrieving revision 1.25
> > > diff -c -r1.25 OracleDialect.java
> > > *** cirrus/hibernate/sql/OracleDialect.java 2 Oct 2002 06:27:55 -0000
> 1.25
> > > --- cirrus/hibernate/sql/OracleDialect.java 16 Oct 2002 08:03:05 -0000
> > > ***************
> > > *** 31,36 ****
> > > --- 31,37 ----
> > >   register( Types.TIMESTAMP, "DATE" );
> > >   register( Types.VARBINARY, "RAW($l)" );
> > >   register( Types.NUMERIC, "NUMBER(19, $l)" );
> > > +                 register( Types.CLOB, "CLOB" );
> > >
> > >   outerJoinGenerator = new OracleOuterJoinGenerator();
> > >
> > > Index: cirrus/hibernate/type/TypeFactory.java
> > > ===================================================================
> > > RCS file:
> > /cvsroot/hibernate/Hibernate/cirrus/hibernate/type/TypeFactory.java,v
> > > retrieving revision 1.20
> > > diff -c -r1.20 TypeFactory.java
> > > *** cirrus/hibernate/type/TypeFactory.java 5 Oct 2002 09:33:42 -0000
> 1.20
> > > --- cirrus/hibernate/type/TypeFactory.java 16 Oct 2002 08:03:06 -0000
> > > ***************
> > > *** 39,44 ****
> > > --- 39,45 ----
> > >   basics.put( Hibernate.CHARACTER.getName(), Hibernate.CHARACTER);
> > >   basics.put( Hibernate.INTEGER.getName(), Hibernate.INTEGER);
> > >   basics.put( Hibernate.STRING.getName(), Hibernate.STRING);
> > > + basics.put( Hibernate.CLOB.getName(), Hibernate.CLOB);
> > >   basics.put( Hibernate.DATE.getName(), Hibernate.DATE);
> > >   basics.put( Hibernate.TIME.getName(), Hibernate.TIME);
> > >   basics.put( Hibernate.TIMESTAMP.getName(), Hibernate.TIMESTAMP);
> > >
> >
> >
>
> --------------------------------------------------------------------------
> --
> > ----
> >
> >
> > > //$Id: ClobType.java,v 1.17 2002/10/11 05:39:15 oneovthafew Exp $
> > > package cirrus.hibernate.type;
> > >
> > > import java.sql.Clob;
> > > import java.sql.PreparedStatement;
> > > import java.sql.ResultSet;
> > > import java.sql.SQLException;
> > > import java.sql.Types;
> > >
> > > public class ClobType extends ImmutableType implements
DiscriminatorType
> {
> > >
> > >   public Object get(ResultSet rs, String name) throws SQLException {
> > >     Clob clob = rs.getClob(name);
> > >     if (clob != null) {
> > >       long len = clob.length();
> > >       if (len > Integer.MAX_VALUE) {
> > >         long idx = 1;
> > >         StringBuffer buf = new StringBuffer();
> > >         while (idx < len) {
> > >           long diff = len - idx;
> > >           if (diff > Integer.MAX_VALUE) {
> > >             buf.append(clob.getSubString(idx, Integer.MAX_VALUE));
> > >             idx += Integer.MAX_VALUE;
> > >           } else {
> > >             buf.append(clob.getSubString(idx, (int)(diff + 1)));
> > >             idx += diff;
> > >           }
> > >         }
> > >         return buf.toString();
> > >       } else {
> > >         return clob.getSubString(1, (int)len);
> > >       }
> > >     }
> > >     return null;
> > >   }
> > >
> > >   public Class returnedClass() {
> > >     return String.class;
> > >   }
> > >
> > >   public void set(PreparedStatement st, Object value, int index)
throws
> > SQLException {
> > >     st.setString(index, (String) value);
> > >   }
> > >
> > >   public int sqlType() {
> > >     return Types.CLOB;
> > >   }
> > >
> > >   public String getName() { return "clob"; }
> > >
> > >   public String objectToSQLString(Object value) throws Exception {
> > >     return '\'' + (String) value + '\'';
> > >   }
> > >
> > >   public Object stringToObject(String xml) throws Exception {
> > >     return xml;
> > >   }
> > >
> > >   public boolean equals(Object x, Object y) {
> > >     if (x==y) return true;
> > >     if (x==null || y==null) return false;
> > >     // don't have to check class for String
> > >     return x.equals(y);
> > >   }
> > >   public String toXML(Object value) {
> > >     return (String) value;
> > >   }
> > > }
> > >
> >
> >
> >
> > -------------------------------------------------------
> > This sf.net email is sponsored by: viaVerio will pay you up to
> > $1,000 for every account that you consolidate with us.
> > http://ad.doubleclick.net/clk;4749864;7604308;v?
> > http://www.viaverio.com/consolidator/osdn.cfm
> > _______________________________________________
> > hibernate-devel mailing list
> > hib...@li...
> > https://lists.sourceforge.net/lists/listinfo/hibernate-devel
> >
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: viaVerio will pay you up to
> $1,000 for every account that you consolidate with us.
> http://ad.doubleclick.net/clk;4749864;7604308;v?
> http://www.viaverio.com/consolidator/osdn.cfm
> _______________________________________________
> hibernate-devel mailing list
> hib...@li...
> https://lists.sourceforge.net/lists/listinfo/hibernate-devel

Re: [Hibernate] Suggestion: move cirrus into a src directory

[Gavin K. <ga...@ap...>]

I plan to do this when/if the packages get renamed to something other than
cirrus.hibernate.*. For now, its very useful to have the full CVS history
for the files......

----- Original Message -----
From: "Max Rydahl Andersen" <ma...@eo...>
To: <hib...@li...>
Sent: Thursday, October 17, 2002 7:56 PM
Subject: [Hibernate] Suggestion: move cirrus into a src directory


> Hi!
>
> I was just wondering why you have placed the .java files package directory
> structure in the root of the
> Hibernate directory. Why not have a seperate src directory for it ?
>
> The current setup confuses e.g. eclipse especically if you build hibernate
> via ant - then it starts
> to see the build directory as a set of classes too :)
>
> /max
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: viaVerio will pay you up to
> $1,000 for every account that you consolidate with us.
> http://ad.doubleclick.net/clk;4749864;7604308;v?
> http://www.viaverio.com/consolidator/osdn.cfm
> _______________________________________________
> hibernate-devel mailing list
> hib...@li...
> https://lists.sourceforge.net/lists/listinfo/hibernate-devel

RE: [Hibernate] hibernate background?

[Anton v. S. <an...@ap...>]

I'm not the author, so I can't answer most of your questions, except for
one:

> Do you know was hibernate ever deployed in a larger
> commercial project? Say over 50 persistable classes?

We're currently using Hibernate on a project with over 100 persisted
classes.  It has been integrated into a reinsurance processing system which
processes over US$1 billion in reinsurance premiums each year.  Most of the
current system was developed before we began to use Hibernate, but with
Hibernate we were able to map the existing legacy database with only a few
minor changes.  This has allowed us to integrate Hibernate into our
application very easily.  We've used Hibernate to add a number of new
features to this application, and it has made development much easier.
We've had no problems related to Hibernate itself.

Anton