Re: svn commit: r35250 - trunk/subversion/mod_dav_svn/reports

C. Michael Pilato wrote:
> Greg Stein wrote:
>> Why not just flush unconditionally? This seems to add a little extra
>> complication. And if you remove .sent_opener anyways, then you'll fall
>> back to an unconditional flush anyways.
>
> See above. No opener, nothing to flush, one less opportunity for an error
> to occur. But I'm not particularly tied to this line of thinking -- I
> merely replicated what the 'svn log' REPORT logic did.

Hrm. It turns out that if I flush the buffer unconditionally, the primary
error returned by svn_repos_node_location_segments() gets lost, and I'm back
to square one. I don't fully understand why, but I can toggle the situation
at will with the addition and removal of that single "if" statement.

Ah! (Thank you, Google.)

See http://subversion.tigris.org/issues/show_bug.cgi?id=3102 for more info!

-- 
C. Michael Pilato <cmpilato_at_collab.net>
CollabNet   <>   www.collab.net   <>   Distributed Development On Demand
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1028784