Hi Ron,

using the examples I could quickly get a parser for PHP-serial data format 
running: 

https://github.com/wollmers/Parse-PHP/blob/master/script/read_serial_marpa.pl

Most problems I had with 

- parsing quoted strings
- syntax of SLIF (the regexy part)
- semantics of latm

IMHO the usage of Marpa could be improved with more examples and less 
academic verbosity (keep it in their own documents).

(BTW: Hope I do better myself in implementing Diff/LCS algorithms, e.g. 
https://github.com/wollmers/LCS-BV, based on young scientific papers;-)

In the end I got it with some trial and error.

Maybe I will try something more ASM (approximate string matching) with 
external, ambigous lexing and ASFs for this

https://github.com/wollmers/Taxon-Parse

in Marpa. In the moment I use something like a parse forest at a higher 
level, but got lost in the jungle;-)



Am Freitag, 29. Mai 2015 01:17:43 UTC+2 schrieb Ron Savage:
>
> Hi Helmut
>
> This is a big decision. Since I'm comfortable with Marpa I'd go so far as 
> to say switch to Marpa straight away no matter how much effort/time has 
> been spent on Parse::RecDescent.
>
> There are people on this list, and the IRC channel, who can give you a lot 
> of help.
>
> And there are many Perl modules you can base you code on. See Marpa's 
> homepage: http://savage.net.au/Marpa.html.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"marpa parser" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to