On Feb 13, 2009, at 10:36 AM, Stefan Küng wrote:
> Hyrum K. Wright wrote:
>>
>> On Feb 13, 2009, at 9:40 AM, Stefan Küng wrote:
>>
>>> Hyrum K. Wright wrote:
>>>> Author: hwright
>>>> Date: Fri Feb 13 07:34:40 2009
>>>> New Revision: 35846
>>>>
>>>> Log:
>>>> Revert r35748, in accordance with the sentiment described in this
>>>> thread:
>>>> http://svn.haxx.se/dev/archive-2009-02/0151.shtml
>>>>
>>>> Modified:
>>>> trunk/Makefile.in
>>>> trunk/build.conf
>>>> trunk/build/generator/gen_win.py
>>>> trunk/subversion/libsvn_fs_fs/ (props changed)
>>>> trunk/subversion/libsvn_fs_fs/rep-cache.c
>>>> trunk/subversion/tests/cmdline/commit_tests.py
>>>
>>> shouldn't https://svn.collab.net/repos/svn/trunk/build/transform_sql.py
>>> also get changed? Right now, the main function:
>>>
>>> if __name__ == '__main__':
>>> if len(sys.argv) != 2:
>>> usage_and_exit("Incorrect number of arguments")
>>> main(sys.argv[1], sys.argv[1] + ".h")
>>>
>>> simply appens ".h" to the filename, which results in rep-cache-
>>> db.sql.h
>>> instead of rep-cache-db.h.
>>> Maybe changing it to
>>>
>>> if __name__ == '__main__':
>>> if len(sys.argv) != 3:
>>> usage_and_exit("Incorrect number of arguments")
>>> main(sys.argv[1], sys.argv[2])
>>>
>>> ?
>>>
>>> Just asking because I'm now using a patched version of that file
>>> in my
>>> build script.
>>
>> What problem are you seeing? I don't have a windows box handy to
>> test
>> this, but Paul Burba reported a few minutes ago in IRC that a clean
>> build of r35846 works just fine for him.
>
> I'm using a custom build script (I'm using NAnt to build all the libs
> required for TSVN). So I'm calling that python script from the command
> line - the official svn build script calls the internal main method
> instead of the __main__ method.
Ah. Do know what fix should be made, and if so, could you share it
with us? :)
-Hyrum
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1151812
Received on 2009-02-13 17:42:42 CET