On Nov 19, 2007, at 3:53 PM, C. Michael Pilato wrote:
> Augie Fackler wrote:
>>
>> On Nov 19, 2007, at 3:42 PM, Karl Fogel wrote:
>>
>>> Augie Fackler <durin42@gmail.com> writes:
>>>> http://subversion.tigris.org/issues/show_bug.cgi?id=3026
>>>
>>> One more thing: the code in adm_files.c predates the invention of
>>> subversion/include/private/*.h. It's possible that some of the adm
>>> area handling could be simplified now, by using svn_wc_private.h.
>>> So
>>> don't take the current architecture as gospel -- if things look
>>> twisted, it's because they are, and you're free to untwist them!
>>
>> Excellent. The actual code responsible for silently ignoring .svn and
>> _svn is in subversion/libsvn_subr/opt.c:1011. If nobody beats me to
>> it,
>> I'll add some code there to print a warning to the user. My
>> inclination
>> is that the client should bail out if it sees an argument like this
>> (rather than continue) - does anyone vehemently object to that?
>
> We must continue to allow folks to do:
>
> $ svn propget svn:mergeinfo .*
>
> (where that * will be expanded to include .svn) without error. I
> don't have
> any problem with printing a "hey we skipped .svn on your behalf"
> message,
> but doing so must not interrupt the operation.
Ah, very true. I knew there was a reason I shouldn't do that. :)
Thanks!
Augie
(sorry about the double-send, cmpilato, I had reply-brain-failure)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Nov 20 01:19:29 2007