On Mar 19, Antony Dovgal wrote:
> On 19.03.2009 15:00, Sergei Golubchik wrote:
> > Hi, Antony!
> > On Mar 18, Antony Dovgal wrote:
> >> Hello all.
> >> At the moment MySQL plugin process is overly complicated - in order to
> >> build a plugin one must perform the following actions:
> >> 1) fetch MySQL sources
> >> 2) run `./configure`
> >> 3) run `make`, as my_config.h is created by make, not ./configure.
> > What kind of plugin do you want to build ?
> I'm working on a statistics daemon that provides access to its data
> via read-only storage engine.
Okay, so it's a storage engine plugin. That explains it.
We have, basically, two groups of plugins - created cleanly from
scratch, they have a well-defined api (in plugin.h) and they don't need
MySQL sources to build. Fulltext parser, daemon and audit plugins are of
And we have plugins which use internal server data structures and
calls - storage engine is of the latter kind. This "api" is unstable,
although it depends on what from the server internals you actually use.
We work on providing a well-defined api to storage engines, but this is
far from complete :(
By the way, if you just want to provide read-only access to your
data,may be INFORMATION_SCHEMA plugin would be more appropriate ?
It's way simpler than storage engine.
Regards / Mit vielen GrЭъen,
__ ___ ___ ____ __
/ |/ /_ __/ __/ __ \/ / Sergei Golubchik <serg@stripped>
/ /|_/ / // /\ \/ /_/ / /__ Principal Software Engineer/Server Architect
/_/ /_/\_, /___/\___\_\___/ Sun Microsystems GmbH, HRB MЭnchen 161028
<___/ Sonnenallee 1, 85551 Kirchheim-Heimstetten
GeschДftsfЭhrer: Thomas Schroeder, Wolfgang Engels, Dr. Roland Boemer
Vorsitzender des Aufsichtsrates: Martin HДring