Re: Tcl faster than Perl/Python...but only with tricks...
- From: Stephan Kuhagen <stk@xxxxxxxxxx>
- Date: Sun, 31 Dec 2006 01:52:14 +0100
Ian Bell wrote:
I am no expert but presumably, not matter what the language, the gets or
read ends up as an fgets C library call. Is it possible the differences
are due in part to the way each language sets up the channel e.g buffer
size, blocking, no chars read in fgets etc?
I'm starting to suspect this also, as well as what Alexandre Ferrieux wrote
about translation and encoding. Maybe there are some settings for buffering
and encoding/translation that can make [gets] as fast as the readlines of
Perl/Python. Knowing those settings (or some good heuristics depending on
the task and environment) could be a great improvement and mentioned in
some tutorial, the man pages or something.
If it's not that, maybe a look at the implementation of [gets] comparing it
to the same things in Python and Perl (no way!) would reveal some
interesting differences.
Regards
Stephan
.
- References:
- Tcl faster than Perl/Python...but only with tricks...
- From: Stephan Kuhagen
- Re: Tcl faster than Perl/Python...but only with tricks...
- From: George Petasis
- Re: Tcl faster than Perl/Python...but only with tricks...
- From: Stephan Kuhagen
- Re: Tcl faster than Perl/Python...but only with tricks...
- From: Alexandre Ferrieux
- Re: Tcl faster than Perl/Python...but only with tricks...
- From: Stephan Kuhagen
- Re: Tcl faster than Perl/Python...but only with tricks...
- From: Uwe Klein
- Re: Tcl faster than Perl/Python...but only with tricks...
- From: Ian Bell
- Tcl faster than Perl/Python...but only with tricks...
- Prev by Date: Re: Tcl faster than Perl/Python...but only with tricks...
- Next by Date: unicode normalization
- Previous by thread: Re: Tcl faster than Perl/Python...but only with tricks...
- Next by thread: Re: Tcl faster than Perl/Python...but only with tricks...
- Index(es):
Relevant Pages
|