On Thu, Jan 19, 2012 at 8:25 AM, John Elrick <john.elr...@fenestra.com>wrote:
>
>
> On Thu, Jan 19, 2012 at 8:15 AM, Richard Hipp <d...@sqlite.org> wrote:
>
>> On Thu, Jan 19, 2012 at 8:10 AM, John Elrick <john.elr...@fenestra.com
>> >wrote:
>>
>> >
>> > I've been reading through the code. Do I understand correctly that if
>> one
>> > prepared statement binding is changed in such a way as may influence the
>> > choice of the query plan that all the prepared statements for that
>> database
>> > connection are flagged for being re-prepared?
>> >
>> >
>> No. Only the one prepared statement whose binding changed is reprepared.
>>
>
>
> Thanks. Now this becomes weirder. I put breakpoints on all the locations
> you mentioned and they were activated precisely as expected. When the
> initial database is created, the breakpoints hit, but are not hit during
> the actual load process. Yet, the first while condition in sqlite3_step:
>
> while( (rc = sqlite3Step(v))==SQLITE_SCHEMA
> && cnt++ < SQLITE_MAX_SCHEMA_RETRY
> && (rc2 = rc = sqlite3Reprepare(v))==SQLITE_OK ){
> sqlite3_reset(pStmt);
> assert( v->expired==0 );
> }
>
> is triggering sqlite3Reprepare like clockwork. So, given that none of the
> listed conditions are occurring, what else could be causing sqlite3Step to
> return SQLITE_SCHEMA?
>
Ok. It's NOT triggering sqlite3Reprepare like clockwork. I added a
function to allow me to check the return value and sqlite3Reprepare is
being called an appropriate number of times. So we're back to the original
question, "why is sqlite3Parser calling _malloc so many times?" I have an
idea of how to determine if any specific queries are to blame, but I have
to work on something unrelated this morning.
_______________________________________________
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
