Re: Looking to improve performance of svn annotate

Hello Julian,

On Montag, 22. März 2010, Julian Foad wrote:
> Hi Philipp. What do you mean exactly? I wonder if you misunderstood
> when I said, "we read in the blamed file's text (just that one revision
> of it)". I meant just the working revision of the file (or whichever
> revision the blame command specified as the operative version). Not
> every revision.
thank you for your clarification.

No, that was a misunderstanding on my side - I wrongly read that after getting
the bytes ranges you're going to the server again to fetch the corresponding
file data.

This is of course bogus, because we should already have the final text.

The idea that should still work is the incremental byte-range building - for
an often-changed file most of the revisions could be skipped, even if the user
didn't specify a revision range.

Thanks again!

Regards,

Phil
Received on 2010-03-22 14:01:03 CET