Vfmd: Markdown with a spec

#1

Post by roop » Mon, Jun 23, 2014, 3:37 PM UTC

Re: Vfmd: Markdown with a spec

#2

Post by dozzie » Mon, Jun 23, 2014, 3:49 PM UTC

Less specification, more formal grammar. Foo-to-HTML converter is next to useless, especially compared to a well-formed parse tree. (In fact, there are already some PEG parsers for Markdown.)

Re: Vfmd: Markdown with a spec

#3

Post by roop » Mon, Jun 23, 2014, 4:05 PM UTC

dozzie wrote: post #2

Less specification, more formal grammar. Foo-to-HTML converter is next to useless, especially compared to a well-formed parse tree. (In fact, there are already some PEG parsers for Markdown.)

It's not possible to write a formal (non-PEG) grammar for Markdown. Hence, a spec, which is the next best thing (a la HTML5). Of course, vfmd is not a md-to-html converter, it's an md-to-parse-tree converter + a parse-tree-to-html converter.

Re: Vfmd: Markdown with a spec

#4

Post by dozzie » Mon, Jun 23, 2014, 7:04 PM UTC

roop wrote: post #3

dozzie wrote: post #2

Less specification, more formal grammar. Foo-to-HTML converter is next to useless, especially compared to a well-formed parse tree. (In fact, there are already some PEG parsers for Markdown.)

It's not possible to write a formal (non-PEG) grammar for Markdown. Hence, a spec, which is the next best thing (a la HTML5). Of course, vfmd is not a md-to-html converter, it's an md-to-parse-tree converter + a parse-tree-to-html converter.

PEG is not a formal grammar? PEG being not a subset of context-free grammars? That's all new things to me.

And you actually have seen PEG parsers for Markdown? Like this one: https://github.com/jgm/peg-markdown/, with grammar written explicitly and lexer/parser generated?

Re: Vfmd: Markdown with a spec

#5

Post by roop » Tue, Jun 24, 2014, 4:40 AM UTC

dozzie wrote: post #4

roop wrote: post #3

Earlier quoted context omitted.

It's not possible to write a formal (non-PEG) grammar for Markdown. Hence, a spec, which is the next best thing (a la HTML5). Of course, vfmd is not a md-to-html converter, it's an md-to-parse-tree converter + a parse-tree-to-html converter.

PEG is not a formal grammar? PEG being not a subset of context-free grammars? That's all new things to me. And you actually have seen PEG parsers for Markdown? Like this one: https://github.com/jgm/peg-markdown/ , with grammar written explicitly and lexer/parser generated?

Please see: http://www.vfmd.org/introduction/#prior-work