db-ojb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Martin Kalén <mka...@apache.org>
Subject Re: [rfc] website docs minor changes
Date Sat, 30 Apr 2005 12:44:35 GMT
Armin Waibel wrote:
>> Also:
>>  * status: +warning re OTM, please help updating the docs!
> 
> Isn't a normal note instead of a warning sufficient or do think a 
> warning is needed because of the unsettled status?

That was more of an "eye-catcher" for us, I planned to remove both
notice/warning before commiting anything. I just put this up big
where I had no clue what to write. :)

>>  * status: +notice/fixme re S.O.D.A., please help updating the docs!
> 
> AFAIK long time ago it was decided that the SODA support will be removed 
> (maybe I'm wrong), because no one ever used this api.

OK, I'll drop it from the doco again until further notice, but I think:
*) if it is usable, why not mention it?
*) if it us unsuable, why not remove it from both docs and code?

>> Update the texts for the "Testing" section.
> 
> +1
> 
> Suggest to remove in all summary.xml files the <section> element to 
> suppress the TOC at the beginning of the document. On a summary site a 
> TOC is not needed.

Yes, I thought so too- it looked a bit strange with a redundand TOC like so:

  TOC TITLE
   link to boxed title

  [BOXED TITLE]

   paragraph about title

It's really rubbing it in, in the readers face? :)

I'll have a look a changing the boxed title into just title for all of those.

>> I took the latest descriptions from Commons Pool and Commons DBCP 
>> websites and updated
>> repository.dtd (what you see on the webpage is copied from 
>> repository.dtd).
>>
>> (Note that there is a new 'minIdle' marked "since OJB 1.0.4", since it 
>> only exists in
>> my local codebase yet. When upgrading commons pool we got this one 
>> "secretly" with
>> a backwards compatible default of 0. I will add it to repository.dtd 
>> and connection conf.)
> 
> Could we add these properties too (to enable prepared statement caching 
> in DBCP, as suggested by you in another thread)?
> poolPreparedStatements
> maxOpenPreparedStatements

Yup, they are in my code and I am running tests at the moment. I can't
test those very well because they conflict with the same concept
(PreparedStatement caching) in the Oracle JDBC-driver, I will run it
against HSQLDB also since the DBCP solution is pure POJO-based and
has no driver requierments at all. Ie, I _think_ it should be good
for MaxDB/SapDB.

I'm finishing the tests/comitts for a few smaller updates in connection
factories in a few hours.

>>  http://people.apache.org/~mkalen/ojb/site/docu/guides/deployment.html#Introduction

> 
> +0, I run my tests against JBoss 3.2.3. Think this version is J2EE 1.3 
> compliant, never tested with JBoss 4.0. Do we have to make a 
> recommendation?

Not at all, I just scanned many documents quickly and thought that
"hey, J2SDK 1.4 is not exactly the void and unknown - do we really
mean what we say here?".

If you say we do, I agree by default. :)

> Think after the first stable release we should use Geronimo as default 
> testing platform for OJB's J2EE integration.

Forget what I said, this was just a guess based on those compile-time
dependency replacements that we have done. I am not running OJB in
any J2EE container, just Tomcat - that's another reason why I didn't
want to change this without asking.

> I must admit that I didn't find time to setup Geronimo with the OJB 
> session bean samples (last time I looked on the web-site Geronimo 
> doesn't support hot/auto-deployment of resources). Do you have 
> experience in using Geronimo?

Very little, we will just keep the documented solution that works IMO!

Regards,
  Martin

---------------------------------------------------------------------
To unsubscribe, e-mail: ojb-dev-unsubscribe@db.apache.org
For additional commands, e-mail: ojb-dev-help@db.apache.org


Mime
View raw message