Re: [pysqlite] tracing
"Edzard Pasma" <[email protected]> Sun, 23 Nov 2008 06:08:02 -0800
| Newsgroups | gmane.comp.python.db.pysqlite.user |
|---|---|
| Message-ID | <[email protected]> |
--- [email protected] wrote: > Other than the connect/Connection thing they are compatible for trivial queries, but beyond that are fundamentally different like all the extra transaction stuff in pysqlite and all the extra SQLite stuff in APSW (vfs, vtables, multiple statements etc). Apsw_dbapi2 mainly adds the auto-transaction mechanism. Further it just maps methods an exception names. It only (?) lacks adapters/converters and row_factory. > What exactly do you mean by "application errors" and "locking mechanism"? Especially the auto-transaction mechanism. You cannot mix it with explicit begin/commit statements (like in Oracle). This can be a source for application errors (see my earlier change proposal: http://osdir.com/ml/python.db.pysqlite.user/2007-02/msg00015.html) Generally an equivalent interface might help debugging even if it just rules out that an error is particular to Pysqlite. It should NOT be used for speed tests as it adds extra ballast. > http://apsw.googlecode.com/svn/publish/connection.html#apsw.Connection.setprofile Great Edzard -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkkpRT4ACgkQmOOfHg372QTnJwCfWIxAERssBBMisvwzTpjySs82 jgIAninFfxBapIIcyRqy/Yla7fJvgK26 =4nP8 -----END PGP SIGNATURE----- _______________________________________________ list-pysqlite mailing list list-pysqlite-FR6EJeJVuqdwc357pe9rcyQmJico6nz3epZhswDD4dQ@public.gmane.org http://itsystementwicklung.de/cgi-bin/mailman/listinfo/list-pysqlite