ANTMark, a markdown grammar for ANTLR4

[MAR9000]

I just publish this project https://github.com/mar9000/antmark .
It's my first UNUSABLE parser due to its slowness.
I hope to receive contributions in order to understand ANTLR4 better than now and make ANTMark a usable parser/translator.

Mar

[Terence Parr]

 interesting. Coincidentally, I've been working on one myself that uses a scanner less approach. That is, tokens are actually characters. Its a difficult language to parse because everything is context-sensitive, even \n.

I will take a look at yours when I get back to working on mine. so far it is pretty clean:

https://github.com/parrt/mini-markdown/blob/master/src/org/antlr/md/MarkdownParser.g4

Ter

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

--

Dictation in use. Please excuse homophones, malapropisms, and nonsense. 

[MAR9000]

In fact I took also a look at CSQL.g4 that is a scanner less grammar (made by you) when I realised that I needed several look-ahead sematic predicates and analysing a char stream is probably faster than analysing a token stream.
Also the resulting lexer is not doing that much, this also suggest in my opinion, a scanner less approach. But when I realized this I was already testing some of last tests.

You could be interested in the case of ordered and unordered lists that causes the parsing never ends, in case see tests/README .

Mar

[Terence Parr]

 I will have to look at this later when I get back to that grammar. But I remember needing some predicates to handle when lists terminate.  It actually exposed a few issues with the analysis in ANTLR, which is why we released 4.2.2.

Ter