On Tuesday, September 3, 2013 8:22:42 AM UTC-4, Alister wrote:
> On Mon, 02 Sep 2013 22:13:27 +0000, Joseph L. Casale wrote:
>
>
>
> > I have been battling an issue hopefully someone here has insight with.
>
> >
>
> > I have a database with a few tables I perform a query against with some
>
> > joins against columns collated with NOCASE that leverage = comparisons.
>
> >
>
> > Running the query on the database opened in sqlitestudio returns the
>
> > results in under a minute. Running the query in Python with sqlite3
>
> > doesn't return results for several hours. I haven't figured out what
>
> > pragmas or other shortcuts sqlitestudio uses to provide the results so
>
> > fast.
>
> >
>
> > Using apsw returns the dataset nearly instantaneously but the
>
> > connection/cursor/commit differences are too drastic and would force far
>
> > too large a rewrite for the module change.
>
> >
>
> > Anyone by chance know the underlying changes required in the sqlite3
>
> > module to replicate what sqlitestudio is doing behind the scenes?
>
> >
>
> > Thanks,
>
> > jlc
>
>
>
> you are almost certainly doing something drastically wrong
>
> can you provides examples of your code & the data structure otherwise I
>
> doubt that anyone will be able to assist.
>
>
> --
>
> To err is human, to forgive, beyond the scope of the Operating System.
This pragma speeds up most processes 10-20 times (yes 10-20):
pragma synchronous=OFF
See the SQLITE documentation for an explanation.
I've found no problems with this setting.
--
https://mail.python.org/mailman/listinfo/python-list
