On Fri, 27 Feb 2004, D. Richard Hipp wrote:
>[EMAIL PROTECTED] wrote:
> >
> > I'm trying to access the database from two threads: the UI thread mostly
> > reads and occasionally updates rows, while the network thread loads new
> > messages and inserts them. Will that work, or do I just misunderstand how
> > SQLite does multithreading?
> >
>
>Each thread should do its own sqlite_open() and they should not share
>the sqlite* pointers that their respective sqlite_open() calls return.
I've always wondered. Is it safe to access the same sqlite*
pointer sequentially?
I have a construct in which I wrap the sqlite* pointer in a structre with
a mutex, then lock the mutex when I begin a transaction, and unlock the
mutex on commit or rollback. All other access is protected with the mutex
and transaction.
I guess what I'm asking is that there are no thread local assumptions in
the code. I guess not, so the above should be safe?
Cheers,
Christian
--
"The problem with the French is that they don't have a word for entrepreneur"
George W. Bush
----------------------------------------------------------------------
/"\
\ / ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
X - AGAINST MS ATTACHMENTS
/ \
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
