A much simpler way to accomplish the same goal is to stop the bizarre
new trend of, every time svn_api_foo2 is implemented, instantly
"upgrading" every call to svn_api_foo to use svn_api_foo2 instead.
Leaving the old calls in (except for cases where the caller really
does need to use the new functionality) means that our own codebase
will automatically test the backwards-compatibility code. Plus, this
would mean doing less work and reduce pointless code churn that makes
merges more difficult.
--dave
On Sun, Jul 5, 2009 at 11:01 AM, Greg Stein<gstein_at_gmail.com> wrote:
> You keep harping on this. Over and over.
>
> Got some time to fix the situation?
>
> On Sat, Jun 27, 2009 at 07:14, Eric Gillespie<epg_at_pretzelnet.org> wrote:
>> I've seen and fixed this kind of bug before. The real problem
>> here is Subversion's complete lack of unit testing; testing
>> svn(1) doesn't count. People implement these wrappers and
>> obviously do not test them.
>>
>> ------------------------------------------------------------------------
>> r31055 | epg | 2008-05-07 08:38:45 -0700 (Wed, 07 May 2008) | 56 lines
>>
>> Fix the r30694 svn_wc_diff_callbacks2_t compatibility (it mistakenly used
>> existing svn_wc_diff_callbacks_t wrappers, which can't work; the
>> structures are very different).
>>
>> Make svn_wc_diff_callbacks2_t work in Python and add Python test for it.
>>
>> ------------------------------------------------------
>> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2365908
>>
>
> ------------------------------------------------------
> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2368167
>
--
glasser_at_davidglasser.net | langtonlabs.org | flickr.com/photos/glasser/
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2368917
Received on 2009-07-07 20:02:38 CEST