Karim BELABAS on Mon, 16 Dec 2002 14:42:32 +0100 (MET)

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: Vec vs Vecsmall

On Sun, 15 Dec 2002 at 13:17:19 -0800, Ilya wrote:
> Sorry, I meant utf-8.  There should a way for Vec("utf-8 string") to
> break the string into unicode chars, *and* produce "integer
> codepoints" for these characters, *and* convert integer codepoints to
> characters.  *This* is why, IMO, we better use flags for conversion
> functions than proliferate (broken implementations, as the last one
> with s* prototype is?) of new and new functions.

The 's*' prototype for Str() conversion is something different. Do you prefer
"string context" and flags [ previous behaviour ] or decent handling of
multiple arguments, but impossibility of flags [ current ] ?

Besides that, flags are a bad thing when they provide deeper functionality
than simple variants [ like different algorithms, verbosity, etc. ]. Most
code should function properly by using default flags only, i.e no flags at all.


P.S: We're still lacking true printf() / sprintf() equivalents but it's yet
another question.

P.S2: There's still no way to convert integers from base B to base B'.
Karim Belabas                    Tel: (+33) (0)1 69 15 57 48
Dép. de Mathématiques, Bât. 425  Fax: (+33) (0)1 69 15 60 19
Université Paris-Sud             Email: Karim.Belabas@math.u-psud.fr
F-91405 Orsay (France)           http://www.math.u-psud.fr/~belabas/
PARI/GP Home Page: http://www.parigp-home.de/