Parrot-0.0.1

Isaac irclark at latveria.castledoom.org
Mon Aug 23 17:00:41 EDT 1999


On Mon, 23 Aug 99 19:41:03 GMT, Phil Hunt <philh at vision25.demon.co.uk> wrote:
>In article <slrn7s1cug.1vk.irclark at latveria.castledoom.org>
>           irclark at latveria.castledoom.org "Isaac" writes:
>> Very interesting indeed!  I was working on a similar program whose
>> functionality would fit much of the description you give here.  My
>> program reads in a description of a gui that's similar in concept
>> to your par files, and outputs code that implements the gui, in
>> either Motif, GTK, or the EZWGL
>
>What's that?
>
Just some toolkit I found on the net.  I found it useful to have several
different toolkits around in order to make sure the language for the 
backend was general enough to get the job done.  I was hoping to allow
specifying backends by parsing a target file describing the target
tool kit.
>
>My next step, once I'm happy with the parser and the HTML backend,
>will be to write a Python/Tkinter backend. Then, I'll have written two
>backends so I'll have a better idea of how to architect a backend 
>framework -- I want to make backends easy to write, to encourage 
>other people to contribute them.
>
Looks like the right approach.  I think it's important to pick disparate
targets for the first few backends.  GTK with C or C++ might be a good 
choice to add to the ones your already trying.  Not too hard to learn,
and fairly popular.
>
>What language is your tool written in? AFAIK, PCCTS is a C parser
>generator -- is this correct? How does it compare to yacc?
>
PCCTS is a superior tool to yacc/bison.  I don't think I'll ever use yacc
again if I get to pick the tool.  PCCTS generates C or C++.  There is a 
second generation tool, ANTLR, that uses improved technology and generates
java and C++ parsers.

Isaac




More information about the Python-list mailing list