Re: Externals inconsistencies and feedback

[Blair Zajac <bl...@orcaware.com>]

Blair Zajac wrote:
> Karl Fogel wrote:
>> This was a no-op switch here, right?  That is, you're just running it to
>> show us how 'switch' doesn't restore the missing externals dir, even
>> though 'update' would, right?
> 
> Correct.
> 
>>> At revision 1.
>>> $ ls wc | wc
>>> 0
>>
>>> Another issue, in externals.c compare_external_items(), I don't think 
>>> we can
>>> return that two external items are identical if either the old or new 
>>> external
>>> item has a HEAD peg revision, since a new HEAD revision may have 
>>> replaced the
>>> item with another item.
>>
>> Excellent point; I cannot imagine disagreeing :-).
> 
> OK :)
> 
>>> So I think we should just get rid of the update_unchanged argument to
>>> svn_client__handle_externals().  We should always update externals.  
>>> I can't
>>> think of a time when you wouldn't want to.
>>
>> I wish I could remember why we added it, but what you're saying seems
>> reasonable now, yes.
> 
> I looked through the logs to see if there was any reason to do that, but 
> I couldn't find any.
> 
> I'll make the changes.

Done in r32897.

Blair

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org