David Crawshaw

When I created the query cache I overloaded some of the existing methods. You can avoid the cache with `PrepareTransient`, but I was working on the assumption that caching is...

This would be a wrapper for [sqlite_preupdate_hook](https://www.sqlite.org/c3ref/preupdate_count.html)? Yes I think that's a good feature.

Nasty. It used to be that sqlite3_reset was always necessary before step would re-execute a statement, but that appears to have changed. It looks like we could use SQLITE_OMIT_AUTORESET to...

A `database/sql` driver certainly should live in a separate package (probably sqlite/driver). I'm not working on it right now because I wouldn't be using it day-to-day, so I wouldn't give...

Thanks for the information. It looks like I have a couple of cases I have not covered. I wish we could combine forces on this, but there's a bit of...

`HAVE_UTIME` is there as of a few days ago. Some of the others look reasonable, but I need to confirm what the values should be on different OSs. E.g. should...

Interesting. Any idea how long you're holding transactions for? I've considered doing this, but connections are already using `sqlite3_busy_timeout` with a 10 second timeout, and in everything I'm doing that's...

The first easy thing we should do is pass context.Deadline from sqlite.Pool through to the returned connection, so that if user code wants to use a connection for more than...

A second motivation for moving the `sqlitex.Pool` object to using a full context is that then it may be possible to use runtime/trace inside the sqlite package. (Though it will...

I'm playing with these API changes on https://github.com/crawshaw/sqlite/tree/sqlitex