[Biojava-l] ontology exception, addSequence & BioSQLSequenceD B

[Matthew Pocock]

This sounds ideal. I look forward to seeing it in CVS. When you commit, 
could you make sure that the jakarta jar files are in CVS as well (-kb 
as binary)?

Matthew

S. Foote wrote:

>Hi Guys,
>
>I have a modified version of the BioSQL classes that uses the standard
>Jakarta connection pool apis and has been working flawlessly in production
>for the last 4 months.  I have been meaning to check it into the cvs, but
>hadn't got around to it.
>
>If there's no objections I'll update my developer cvs version with the
>latest changes and merge those with my modified connection pool version
>and then check it in.
>
>Cheers,
>Simon Foote
>
> According to Matthew Pocock:
>  
>
>>Matthew Pocock wrote:
>>
>>    
>>
>>>More news when I have it.
>>>      
>>>
>>I have more news :)
>>
>>The version in CVS works for me. I had to do some work to the 
>>OntologySQL class, and modify the biosql schema (which I've attached). 
>>In the end, I removed the symptom by dissabling connection re-use in 
>>JDBCConnectionPool. Could we either a) make JDBCConnectionPool work with 
>>rollback gracefully or b) switch to a standard connection pool API? My 
>>vote is for option 2, but then some of the things I read via google 
>>indicated that we may have the same problems.
>>
>>BioSQL People - could you check the single table I've added 
>>(term_relationship_term) and see if it's sane for folding back into the 
>>trunk of biosql development?
>>
>>Matthew
>>
>>
>>    
>>
>
>[Attachment, skipping...]
>
>[Attachment, skipping...]
>
>  
>