Advice/help wanted on bzr fast-export-from-cvs

[Michael Haggerty]

Robert Collins wrote:
> On Wed, 2009-08-12 at 00:45 +0200, Michael Haggerty wrote:
>> Even faster than those two options, by a significant factor, is a
>> --use-internal-co option that I have prototyped on my hard disk but
>> not
>> yet released to the wild.  The analogous option is the default for
>> cvs2svn and would probably be the best default for the other
>> converters,
>> if it can get released in time.
> 
> I'd be inclined to start up a CVS server and talk directly to it to
> extract content. That should be pretty darn fast and avoid all the
> startup-teardown overhead involved in invoking cvs repeatedly. cscvs has
> generic python bindings for the CVS protocol in its CVS module.

That's a good idea, and thanks for the reference.

My code does have one advantage compared to the CVS server code
though--it generates the fulltext of all revisions within a file at
once, meaning that it only has to read the file once per branch rather
than once per revision.  Whether that advantage allows it to beat the
(presumably optimized) C code in the CVS server can probably only be
resolved by benchmarking.

Michael