Writing a parser

Zielony has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: Writing a parser by kyle (Abbot) on Jan 29, 2008 at 20:05 UTC
A somewhat more complicated parser written without modules: A parser for a limited grammar. You could also look at how something like Parse::RecDescent works internally and mimic the parts you need.	[reply]
Re: Writing a parser by apl (Monsignor) on Jan 29, 2008 at 19:37 UTC
You have a lot of work ahead of you. The very first thing you should do (after considering what you want the parser to produce, as moritz asked) is to develop the Backus-Naur Form you wish to parse. If you don't know what BNF is, you should ask yourself why you wish to develop a parser.	[reply]
Re: Writing a parser by moritz (Cardinal) on Jan 29, 2008 at 19:14 UTC
It looks like you can just iterate over the lines, and check with regexes if the line starts with `#`, and split at whitespaces. But you have to be a bit more specific about the syntax, and what you want the parser to do with it (generate a parse tree? or an abstract syntax tree? Or execute it straight away?)	[reply] [d/l]
Re^2: Writing a parser by pc88mxer (Vicar) on Jan 29, 2008 at 20:21 UTC
Just based on what you are telling us, I would second this approach. I've written the following code many times: `while (<FH>) { s/#.*$//; my @args = split(' ', $_); next unless @args; ... process command ... }` [download] If, however, you have any recursive syntactical elements (like begin/end grouping symbols which need to be matched), a proper parser written with `Parser::RecDescent` is probably in order.	[reply] [d/l] [select]
A reply falls below the community's threshold of quality. You may see it by logging in.