Re: libsvn_client/commit.c:remove_tmpfiles()

mark benedetto king <mbk@lowlatency.com> writes:

> It is certainly possible that the file could not exist: it may have been
> removed by some other process.

It's also possible that the file was opened with the "delete-on-close"
flag, though I haven't the energy to verify this. If that's the case,
though, perhaps this function should just go away, and we'll let APR
do its job.

Of course, there is an issue filed about how updates are less than
perfectly selective about their use of tmpfiles that was originally
filed noticed because commits which left tmpfile lying about caused
serious problems for updates. I say all that to say, if the update
thing is still a real issue (it hasn't been verified in some time now)
then we need to make quite certain that those tmpfiles are getting
removed -- and of course, we need to fix update, too.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Jul 10 01:22:03 2003