Re: [Python-checkins] CVS: python/dist/src/Objects abstract.c,2.49,2.50
Thomas Wouters writes:
Fix grouping: this is how I intended it, misguided as I was in boolean operator associativity.
And to think I spent time digging out my reference material to make sure I didn't change anything! This is why compilers have warnings like that! -Fred -- Fred L. Drake, Jr. <fdrake at beopen.com> BeOpen PythonLabs Team Member
On Thu, Aug 31, 2000 at 03:09:33AM -0400, Fred L. Drake, Jr. wrote:
Thomas Wouters writes:
Fix grouping: this is how I intended it, misguided as I was in boolean operator associativity.
And to think I spent time digging out my reference material to make sure I didn't change anything!
Well, if you'd dug out the PEP, you'd have known what way the parentheses were *intended* to go :-) 'HASINPLACE' is a macro that does a Py_HasFeature() for the _inplace_ struct members, and those struct members shouldn't be dereferenced if HASINPLACE is false :)
This is why compilers have warnings like that!
Definately ! Now if only there was a permanent way to add -Wall.... hmm...
Hey, I got it ! What about we set it by default, if the compiler is gcc ? :)
--
Thomas Wouters
Thomas Wouters writes:
Definately ! Now if only there was a permanent way to add -Wall.... hmm... Hey, I got it ! What about we set it by default, if the compiler is gcc ? :)
I'd be happy with this. -Fred -- Fred L. Drake, Jr. <fdrake at beopen.com> BeOpen PythonLabs Team Member
On Thu, 31 Aug 2000, Fred L. Drake, Jr. wrote:
Thomas Wouters writes:
Definately ! Now if only there was a permanent way to add -Wall.... hmm... Hey, I got it ! What about we set it by default, if the compiler is gcc ? :)
I'd be happy with this.
For 2.1, I suggest going for -Werror too.
--
Moshe Zadka
On Thu, Aug 31, 2000 at 10:45:19AM +0300, Moshe Zadka wrote:
Thomas Wouters writes:
Definately ! Now if only there was a permanent way to add -Wall.... hmm... Hey, I got it ! What about we set it by default, if the compiler is gcc ? :)
For 2.1, I suggest going for -Werror too.
No, don't think so. -Werror is severe: it would cause compile-failures on
systems not quite the same as ours. For instance, when using
Linux-2.4.0-test-kernels (bleeding edge ;) I consistently get a warning
about a redefine in
participants (3)
-
Fred L. Drake, Jr. -
Moshe Zadka -
Thomas Wouters