[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Bdb strings anomaly

From: C. Michael Pilato <cmpilato_at_collab.net>
Date: Wed, 13 Jan 2010 08:03:47 -0500

Vadim Chekan wrote:
> I'm not sure how to solve it. One way is to modify
> svn_fs_base__get_mutable_rep to create new keys without saving
> strings/representations tables.
> Another way is to return new unsaved string/representation implying
> that consequent calls will modify representation and save it.

Vadim, thanks for the deep dive into the code. Without diving in myself,
let me just say that we have plenty of precedent in the BDB backend for
reserving a key ID without actually storing a row of data keyed on that ID.
 So perhaps you've nailed exactly the right fix for this -- teach the code
to reserve IDs without storing data behind them until there's real data to
add, and of course teach it also not to freak out when dealing with such a
data-less key unless the code knows there should have been something stored
there already.

C. Michael Pilato <cmpilato_at_collab.net>
CollabNet   <>   www.collab.net   <>   Distributed Development On Demand

Received on 2010-01-13 14:04:34 CET

This is an archived mail posted to the Subversion Dev mailing list.