I'd like to build Markdown as Xtext DSL. But it seems tricky to parse headings like # Introduction, because it has no definite end-symbol. Is there any way to express this? Or is it generally (also not alone a Xtext limitation) not possible?
Here is my Xtext:
grammar markdown.Markdown with org.eclipse.xtext.common.Terminals
generate markdown "http://www.Markdown.markdown"
Model:
entities+=Entity*;
Entity:
Section | Subsection | Paragraph
;
Section:
'#'
content+=TextPart
'::'
;
Subsection:
'##'
content+=TextPart
'::'
;
Paragraph:
content+=TextPart
;
TextPart:
text=Text
;
Text:
(ID | WS | SINGLE_NL | MULTI_NL | ANY_OTHER | '\\[' | '\\]' | ',' | "-" | '\\:' | '\\%' | '\\#' | '\\##' )+;
terminal ID:
('a'..'z' | 'A'..'Z' | '_') ('a'..'z' | 'A'..'Z' | '_' | '0'..'9')*;
terminal SL_COMMENT:
'%%' !('\n' | '\r')* ('\r'? '\n');
terminal MULTI_NL:
'\r'? '\n' (/*(' ' | '\t')**/ '\r'? '\n')+;
terminal SINGLE_NL:
'\r'? '\n';
terminal WS:
' ' | '\t';
terminal ANY_OTHER:
.;
The terminals are from Xdoc. With this grammar-rules is this possible:
# Introduction ::
Lorem ipsum.
## Other chapter ::
Lorem ipsum.
But I'd like Markdown like this:
# Introduction
Lorem ipsum.
## Other chapter
Lorem ipsum.
Therefore we need \n instead of :: as ending. But is this possible?
Furthermore antlr produces warnings caused by the terminal-rules. But this warnings don't occur when building Xdoc.xtext. What is my fault?
warning(200): ../markdown/src-gen/markdown/parser/antlr/internal/InternalMarkdown.g:436:1: Decision can match input such as "'-'" using multiple alternatives: 9, 14
As a result, alternative(s) 14 were disabled for that input
warning(200): ../markdown/src-gen/markdown/parser/antlr/internal/InternalMarkdown.g:436:1: Decision can match input such as "'\\['" using multiple alternatives: 6, 14
As a result, alternative(s) 14 were disabled for that input
...
