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

Re: svn commit: r893267 - in /subversion/trunk/subversion/libsvn_fs_base: bdb/node-origins-table.c bdb/node-origins-table.h dag.c revs-txns.c

From: Julian Foad <julianfoad_at_btopenworld.com>
Date: Tue, 22 Dec 2009 19:26:42 +0000

C. Michael Pilato wrote:
> julianfoad_at_apache.org wrote:
> > Author: julianfoad
> > Date: Tue Dec 22 18:44:27 2009
> > New Revision: 893267
> >
> > URL: http://svn.apache.org/viewvc?rev=893267&view=rev
> > Log:
> > For obliterate in BDB, implement updating the "node-origins" table.
> >
> > * subversion/libsvn_fs_base/bdb/node-origins-table.c,
> > subversion/libsvn_fs_base/bdb/node-origins-table.h
> > (svn_fs_bdb__change_node_origin): New function to overwrite an existing
> > node-origins row.
>
> Why not just re-use the existing __delete_node_origin() and
> __set_node_origin() functions?

Good idea.

> Or a boolean 'allow_overwrite' flag to
> __set_node_origin()?

I had a perfectly good solution that just disabled that function's check
for not changing an existing entry, but I didn't want to change that
function in that way. I was reluctant to change it at all. But yes, I
could add such a flag.

> Disclaimer: I haven't examined in detail the nuances of the calling
> requirements of these various functions or your new one.

Thanks for the suggestions. I should do one or the other.

- Julian

p.s. Do we really want follow-ups on both the dev@ list and the commits@
list? I don't. I'm removing commits@.
Received on 2009-12-22 20:28:06 CET

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

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.