Speeding up i18n

[John Tapsell]

2009/2/3 Chusslove Illich <caslav.ilic at gmx.net>:
>> [: John Tapsell :]
>> Has anyone looked into speeding up i18n calls by generating the string
>> hash at compile time rather than at runtime.
>
> Do you have a real-life situation where they are too slow?

Nope, I was looking for someone like you who would know whether this
is worth investigating.  I didn't want to waste my time trying to
speed it up if there was no point.  It was just a random idea :-)



> For an average message (~40 chars, no placeholders) i18n can now deliver ~50
> kmsg/s (kmsg/s = 1000 messages per second) on my 2 GHz AMD Athlon XP machine
> (average machine in 2004).
>
> If that is not good enough, then it's certainly not hashing that should be
> improved, as raw Gettext call (which does the hashing) delivers >1000 kmsg/s
> for the same message/machine (no, no extra zero here).
>
> --
> Chusslove Illich (Часлав Илић)
>