[Python-Dev] PyErr_SafeFormat

Moshe Zadka Moshe Zadka <moshez@math.huji.ac.il>
Sun, 16 Jul 2000 15:45:02 +0300 (IDT)

On Sun, 16 Jul 2000, M.-A. Lemburg wrote:

> In the end, I would prefer the solution I mentioned a few days
> ago: add checks to configure which determine the availability
> of snprintf() and then put a macro def into pyport.h which
> either uses snprintf() (if available) or simply drops the
> length argument and sticks with sprintf().
> Since snprintf() will become a standard anyway, why try to
> emulate its behaviour ?

I'm -1 on that approach. Python tries to cater to low-end C compilers as
well, and even when snprintf() will become a standard, it will take time
to become a widely available standard. /F's patch works *right* *now* on
almost every C compiler.

python-turned-ansi-only-when-c9x-is-on-the-brink-ly y'rs, Z.

Moshe Zadka <moshez@math.huji.ac.il>
There is no GOD but Python, and HTTP is its prophet.