On Fri, Feb 18, 2011 at 12:33 AM, Eric Deplagne <[email protected]> wrote:
> On Thu, 17 Feb 2011 19:18:20 +0100, Arve Knudsen wrote: > > 2011/2/17 Jürgen Hermann <[email protected]> > > > > > > It has to? Why? For religious reasons? > > > > > > No. It's because you can easily turn off what you see, but it's hard to > > > turn on what you don't see. > > > > > > > After programming a lot of C/C++, this is the first time I've heard > anyone > > complain that gcc (or any other compiler) isn't super strict by default. > How > > hard is it anyway to put -Wall in your CFLAGS?? > > The trouble being -Wall doesn't even enable everything... > > > I definitely think it's better to let people enable especially strict > > warnings if/when they see the need; besides, static checks aren't by any > > stretch perfect, they can merely indicate possible code improvements. > > Consider also that Python being a dynamic language makes it notoriously > > difficult to get a tool like pylint right, meaning that there will be a > > certain amount of false positives, which result in extra work for the > > programmer and uglier code (pylint directives in comments). It's better > for > > pylint not to be overly ambitious, considering it's a means to an end, > not > > an end in itself (to some of us anyway). > > > > Arve > > The fact that it is horribly difficult in python to keep track of things > just makes it all the more important that the tool is indeed as fascist as > it can... > > I prefer false positives that I can think about than false negatives that > will let me do horrors... > Having given some thought to this philosophy of checking code as strictly as possible by default, with painful side effects and all; it strikes me as similar to greeting everyone you meet with a smack on the mouth, on the off chance the receiver find it pleasurable ;) Arve Arve
_______________________________________________ Python-Projects mailing list [email protected] http://lists.logilab.org/mailman/listinfo/python-projects
