Re: svn commit: rev 2035 - trunk/subversion/libsvn_repos

sussman@tigris.org writes:

> 1. if the dumpstream parser successfully svn_fs_copy()'s a subtree,
> then make sure it *ignores* all child nodes of the subtree (...as they
> continue to come from the stream.)

Um...I didn't read your patch or the code, but do we still have a way
for the dump/load process to propogate a copy + mods ? That is, if I
copy /trunk to /branches/foo and then modify
/branches/foo/subversion/libsvn_fs/err.c, that's all good, right?

> 2. the dumper was forgetting to add a blank line after outputting a
> set of 'delete' headers, so the headers were blending into the next
> set of 'new revision' headers, appearing to be one giant header block.

Heh...I noticed this the other day, and meant to ask you about it.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Jun 1 14:23:20 2002