x::="x"
+// interactions between !=> and &~ mean that I need to rethink the chartage
// indentation styling...
// literal blocks [[need to ignore bracing]] double-colon style?
// definition -- by prior line indentation, like headings in the original structured text
// dropcap
// output formats: latex, contex, ps, pdf, html, man, txt, rfc
-// URGENT: why does swapping [a-zA-Z0-9] for alphanum in "item" cause severe breakage?
-// URGENT: why does enabling "url" cause severe breakage? (probably same cause)
-// ... something to do with unit productions
-
-// output formats: latex, contex, ps, pdf, html, man, txt, rfc
-// bullet list
-// numbered list
-// horizontal rule
-// dropcap
-// smallcap
-// strikethrough
-// link
+// escapification
+// comment
// math
// image
-// today's date, last edit date
-// blockquote
-// superscript
-// subscript
-// citations/references
-// typewriter-text
// figures
// FIXME: these have to go at the top so they have their dropAll bit set before PreSequence.build...
//////////////////////////////////////////////////////////////////////////////
-s ::= { Doc } => "top"
+s ::= {Doc} => top
-Doc ::= Header Body /ws => doc
+Doc ::= Header Body /ws => doc
Header ::= "header" { kv */ ws } /ws => header
Body ::= Section*/ws => body
Section ::= SectionHeader Paragraph*/ws /ws => section
kv ::= word "=" text /ws => kv1
+num !::= [0-9]++
Paragraph ::= { "\"\"" ws text } => "blockquote"
> { "*" " " ws text } => "ul"
> { "#" " " ws text } => "ol"
- > { num " " ws text => "ol" }
+ > { num " " ws text } => "ol"
> { "---" "-"* } => "hr"
> { text } => "p"
-text ::= item */ ws
+text ::= item => "t"
+itemx ::= ws item => "w" | ()
+item ::= blockquote => "b"
+ > pre itemx => "a"
+ > structured itemx => "1"
+ > structuredx itemx => "2"
+ > styled itemx => "3"
+ > qtext itemx => "4"
+ > alphanum++ itemx => "5"
+ > symbol itemx => "6"
+// > sym++ itemx => []
+ > Paragraph itemx => "7"
+
+symbol ::= sym++
+
+blockquote ::= "\"\"" text "\"\"" => "blockquote"
+ | "\"\"" block => "blockquote"
+
+qtext ::= "\"" text "\"" => "quoted"
pre ::= "[verbatim]" { ~[]+ } /ws => "verbatim" // FIXME doesn't work
-item ::= pre
- > email
- > structured
- > styled
- > "\"" text "\"" => quoted
- > alphanum++ => "stringify"
- > symbol
-
-symbol ::= symbolx & sym++
-symbolx ::= "--" => emdash
- | ","
- | ":"
- | ";"
-
-styled ::= "**" text "**" => bold
- | "__" text "__" => ul
- | "~~" text "~~" => it // hard to type
- | "((" text "))" => footnote
-
-structured ::= glyph
-// | url
-glyph ::= "(r)" | "(c)" | "(tm)" // euro symbol?
+styled ::= "__" text "__" => ul
+ | "((" text "))" => footnote
+ | ( "[[" text "]]" => tt
+ > "[" word "]" => citation
+ )
+ | "!!" text "!!" => strikethrough
+ | "^^" (word|block) => superscript
+ | ",," (word|block) => subscript
+ | "\\sc" block => smallcap
+ | "**" text "**" => bold
+ | "!" (word|block) => keyword
+ > "*" text "*" => it
+
+block ::= { text }
+structured ::= { text } "->" (url|email) => link
+ //> alphanum++ "->" (url|email) => link
+structuredx ::= glyph
+ > email
+ > url
+
+glyph ::= "(r)" | "(c)" | "(tm)" | "--" // euro symbol?
+ | "\\today" -> ~[a-z] => today
// URLs //////////////////////////////////////////////////////////////////////////////
// only gets parsed once
urlpath ::= urlchar*
-username ::= [a-zA-Z0-9;/?:&=$\-_.+]++ => "stringify"
-password ::= [a-zA-Z0-9;/?:&=$\-_.+]++ => "stringify"
+username ::= [a-zA-Z0-9;/?:&=$\-_.+]++
+password ::= [a-zA-Z0-9;/?:&=$\-_.+]++
urlchar ::= [a-zA-Z0-9;/?:&=$\-_.+@]
| "%" [0-9] [0-9] => "%"
url ::= "mailto" ":" email
method ::= [+\-.a-z0-9]+
port ::= [0-9]+
-domain ::= part +/ "."
-part ::= [A-Za-z0-9\-]++ => "stringify"
+domain ::= (part +/ ".") -> ~"."
+part ::= [a-zA-Z0-9\-]++ // interesting use of boolean grammars
// &~ ([\-0-9] ~[]* | ~[]* [\-0-9])
-email ::= username "@" host => email
-host ::= domain
- | [0-9]+ "." [0-9]+ "." [0-9]+ "." [0-9]+ => "ip"
+email ::= username "@" host -> ~[.] => emailaddr
+host ::= [0-9]+ "." [0-9]+ "." [0-9]+ "." [0-9]+ => "ip"
+ | domain
// Tokens ///////////////////////////////////////////////////////////////////
-word ::= alphanum++ => "stringify"
+word ::= alphanum++
| quoted
quoted ::= "\"" ((~[\"\\] | escaped)+) "\""
// Chars ///////////////////////////////////////////////////////////////
alpha ::= [a-zA-Z]
-num !::= [0-9]++ => "stringify"
//num ::= [0-9]
alphanum ::= [a-zA-Z0-9]
-sym ::= ~[a-zA-Z0-9\ \r\n]
+sym ::= ~[a-zA-Z0-9\ \r\n=\">]