lib/hemingway/latex.treetop in hemingway-0.0.2 vs lib/hemingway/latex.treetop in hemingway-0.0.3

- old
+ new

@@ -32,20 +32,20 @@ # - Okay so this is interesting. When you reference something with a * or # a +, then you're creating a top level array-like structure that you'll # have to dig past to get at the list. elements[0] refers to the array of # tags or texts here. rule paragraph - sequence:( content / footnote )* eop <ParagraphNode> + sequence:( content / footnote / !eop newline )* eop <ParagraphNode> end # The last paragraph in an entry need not end with \n\n. Because this # comes second in the ordered choice for entry, it will only be used # if a paragraph has concluded for some reason and there wasn't an eop # to consume. The only time I can imagine this happening would be at # the end of an entry. rule last_paragraph - sequence:( content / footnote )+ eop? <ParagraphNode> + sequence:( content / footnote / !eop newline )+ eop? <ParagraphNode> end # Example: \tag{text} or just text or $\Delta$ # Notes: # - So I can define methods based on the ordered choice in the top @@ -56,11 +56,11 @@ special / tag / block / math / text end # Treetop does not separate lexing from parsing. Must consume all input. rule whitespace - ( " " / newline )* + ( " " / newline )+ <WhitespaceNode> end rule spaces ( " " )* end @@ -72,10 +72,10 @@ rule eop newline 2.. end rule newline - "\n" + "\n" <NewlineNode> end end end \ No newline at end of file