Re: Performance with GNU Gettext?



Sophia,

Today at 19:59, Behdad Esfahbod wrote:

> No single profiling data has ever shown anything gettext or
> locale related out of glib taking anything like significant time
> (say 1%).  So, I won't bother with them.  The way gettext is
> implemented is that it keeps the messages in a binary hash table
> on the disk (the .mo file) and looks up messages in it, so as far
> as I know, there's not much initialization overhead involved.

Also note that it's commonly memory mapped (I think: it's been a while
since I looked at that), so it's even faster than that.  It's also
pretty conservative in what files it loads (it was my impression that
Sophia expects it to load all translations, while it loads only one
that is actually needed).

Anyway, for anyone using Glade (think XML parsing), GNU gettext should
not be a performance issue at all. 

Cheers,
Danilo



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