[Numpy-discussion] RE: Python 2.2 seriously crippled for numerical computation?

[Gerhard Häring]

Le 04/03/02 à 20:04, Tim Peters écrivit:
> [Tim]
> >> For the rest of it, it looks like the "1e-200**2 raises OverflowError"
> >> glitch is unique to platforms using glibc.
> 
> [Jon Ribbens]
> > No, it happens on OpenBSD 2.7 also.
> >
> > Python 2.2 (#2, Feb  3 2002, 18:27:17)
> > [GCC 2.95.2 19991024 (release)] on openbsd2
> > Type "help", "copyright", "credits" or "license" for more information.
> > >>> 1e-200**2
> > Traceback (most recent call last):
> >   File "<stdin>", line 1, in ?
> > OverflowError: (34, 'Result too large')
> 
> I don't understand your "no":  the output shows that Python was built using
> GCC 2.95.2, yet you're claiming this does *not* use glibc?  gcc and glibc go
> together like Guido and Van Rossum <wink>.
 
True. In that there is one Guido van Rossum. And a lot of other people
with Guido as a first name. And also a bunch of other van Rossumns.

glibc is just the libc on Linux systems, but gcc runs perfectly well on
a gazillion other operating systems, like the varios *BSD and even
Windows, in the gcc incarnations of mingw32 and Cygwin.

And even on Linux, there are other compilers linking against the glibc,
like the Intel compilers.

Gerhard
-- 
This sig powered by Python!
Außentemperatur in München: 1.5 °C      Wind: 1.7 m/s