[Python-Dev] [Python-checkins] r86355 - python/branches/py3k/Modules/_pickle.c (original) (raw)
Stephen J. Turnbull stephen at xemacs.org
Thu Nov 11 04:04:39 CET 2010
- Previous message: [Python-Dev] [Python-checkins] r86355 - python/branches/py3k/Modules/_pickle.c
- Next message: [Python-Dev] [Python-checkins] r86355 - python/branches/py3k/Modules/_pickle.c
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Alexander Belopolsky writes:
On Wed, Nov 10, 2010 at 7:28 AM, Victor Stinner <victor.stinner at haypocalc.com> wrote: ..
I don't know, but the commit is trivial and cheap. If it improves the support on uncommon compiler, I agree to commit such change.
But it does it at the cost of invalidating the "svn blame" for the last enum entry now and for future additions. The problem is that when you change from
enum { .. X }
to
enum { .. X, Y }
If that bothers you, you can write
enum { A , B /* etc */ , X }
or
enum { A, B, /* etc */ X, enum_bound_otherwise_unused }
I prefer the last; it's a compiler (and debugger) space burden, but shouldn't affect the running python. On the original question, I think it's preferable to keep compilers happy unless you're willing to require C99.
- Previous message: [Python-Dev] [Python-checkins] r86355 - python/branches/py3k/Modules/_pickle.c
- Next message: [Python-Dev] [Python-checkins] r86355 - python/branches/py3k/Modules/_pickle.c
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]