Subject | [ib-support] Survival of Prepared state was Re: just if you are curious |
---|---|
Author | Roman Rokytskyy <rrokytskyy@acm.org> |
Post date | 2003-02-24T08:24:59Z |
Hi,
call as well.
However prepared statements survive normal commit too and JayBird
supports them. Method supportsOpenStatementsAcrossCommit() must be
corrected to return true. This was already asked here and it was
proved to work correctly.
Correct me if I am wrong, but commit retaining is relevant only to
result set (in other words, result set do not survive the commit,
while statements do survive).
Best regards,
Roman Rokytskyy
> My mistake...the JayBird DOES support commit retaining.....but onlyCommitRetaining is not available in JDBC and we do not use this API
> at the lowest (FB API) level.
> The JDBC functions do not use them, as I see this code somewhere
> in a file called FBDatabaseMetaData.java. Seems like it's hardcoded.
>
> public boolean supportsOpenStatementsAcrossCommit() throws
> SQLException
> {
> return false;//commit retaining only.
> }
>
> AFAIK, JDBC does not have a "CommitRetaining" function. (Neither
> does ADO/OLEDB if I understand correctly). A connection url switch
> to toggle commit retaining function might be useful in this case,
> since I suspect it could help out with performance especially
> if we're using some O/R mapping tool...
call as well.
However prepared statements survive normal commit too and JayBird
supports them. Method supportsOpenStatementsAcrossCommit() must be
corrected to return true. This was already asked here and it was
proved to work correctly.
Correct me if I am wrong, but commit retaining is relevant only to
result set (in other words, result set do not survive the commit,
while statements do survive).
Best regards,
Roman Rokytskyy