[Python-Dev] RE: [Python-checkins] python/dist/src/Objects listobject.c, 2.218, 2.219
Tim Peters
tim.peters at gmail.com
Tue Aug 3 16:53:12 CEST 2004
More information about the Python-Dev mailing list
Tue Aug 3 16:53:12 CEST 2004
- Previous message: [Python-Dev] RE: [Python-checkins] python/dist/src/Objects listobject.c, 2.218, 2.219
- Next message: [Python-Dev] Proposed Python api for testing whether a module exists
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
[Sjoerd Mullender] > This is indeed perverted. Another option is to add a statement > (void) status; > somewhere to the function and to remove the status++ from the return. > This seems to shut up all compilers I've dealt with. If that's enough to shut compilers up, +1 from me. MSVC 7.1 doesn't whine even with no tricks. I do want to keep the two optimizations in list.pop(), which consist of not test+branch'ing the return values of two calls that have been proved safe. But things change over time, and in a debug build I do want to assert that they are in fact safe. I originally did this with a pile of #ifdef Py_DEBUG blocks, but (as the original checkin comment said) that was even uglier than the comma-expression hack.
- Previous message: [Python-Dev] RE: [Python-checkins] python/dist/src/Objects listobject.c, 2.218, 2.219
- Next message: [Python-Dev] Proposed Python api for testing whether a module exists
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Python-Dev mailing list