On Mon, Apr 16, 2012 at 11:39 AM, Julian Foad
<julianf...@btopenworld.com> wrote:
> Johan Corveleyn wrote:
>
>> Julian Foad wrote:
>>>  If I can help by changing the conditional compilation of the symmetric
>>> merge code, I'm happy to do so, just let me know.
>>
>> For now, it's ok, I'm using a Debug build for the time being, assuming
>> that this problem will go away eventually, if the symmetric merge
>> becomes standard.
>>
>> I'm wondering though why SVN_WITH_SYMMETRIC_MERGE is defined only when
>> SVN_DEBUG is defined. This is trunk anyway, so it won't be released as
>> is. Why make the symmetric merge code conditional on whether or not
>> it's a debug build?
>
> No hard-and-fast reason.  A minor practical reason: it's helpful to leave 
> some SVN_DBG calls in that code for the time being, and those don't compile 
> in release mode (that's intentional); but we could easily overcome that if we 
> wanted ("#ifdef SYMMETRIC #define SVN_DBG", for example).

Ah ok, I understand. No problem for the time being. I'll continue
using a debug build now. If I find I need to work/test with a release
build again, I'll see if I can just move the SVN_WITH_SYMMETRIC_MERGE
ifdefs to a smaller scope so the needed symbols
(svn_client__do_symmetric_merge and svn_client__find_symmetric_merge)
are always defined, but otherwise empty.

-- 
Johan

Reply via email to