On Sep 17, 2009, at 2:34 PM, Stefan Sperling wrote:
> On Thu, Sep 17, 2009 at 11:42:29AM -0700, Dave Brown wrote:
>> I am not sure if error code SVN_ERR_BAD_RELATIVE_PATH is correct
>> here, but seems reasonable.
>>
>> [[[
>> Raise an error in log.c when given paths that are not in the WC,
>> rather
>> than letting the code segfault. (Absolute paths in /tmp seem to be
>> coming from merge operations). Callers of loggy_path() are not
>> prepared
>> to handle a NULL return anyway. Even though this stuff is going
>> away,
>> it is crashing midway through the javahl tests. With this change,
>> the javahl tests complete, getting 1 error and 2 failures.
>>
>> * subversion/libsvn_wc/log.c
>> (loggy_path): return svn_error if path is not a child of
>> adm_abspath
>> ]]]
>>
>> ------------------------------------------------------
>> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2396145
>
>> Index: subversion/libsvn_wc/log.c
>> ===================================================================
>> --- subversion/libsvn_wc/log.c (revision 39409)
>> +++ subversion/libsvn_wc/log.c (working copy)
>> @@ -1727,9 +1727,14 @@
>> SVN_ERR(svn_dirent_get_absolute(&abspath, path, pool));
>> *logy_path = svn_dirent_is_child(adm_abspath, abspath, NULL);
>>
>> - if (! (*logy_path) && strcmp(abspath, adm_abspath) == 0)
>> + if (! (*logy_path) ) {
>
> This does not conform to our coding style.
> Should be:
>
> if (! (*logy_path))
> {
> ...
> }
>
> The fix looks right to me otherwise.
> The docstring says "PATH must not be outside that directory" so
> raising an error if that's the case is perfectly fine.
>
> Thanks!
>
> . o O (maybe this variable should be called "loggy_path"?)
Can't do that. The name of the enclosing function is loggy_path().
-Hyrum
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2396174
Received on 2009-09-17 21:40:23 CEST