+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...
-ws !::= w**
-w !::= [\r\n\ ]
-nw ::= ~[\r\n\ ]
+ws ! = w**
+w ! = [\r\n\ ]
+nw = ~[\r\n\ ]
//////////////////////////////////////////////////////////////////////////////
-s ::= {Doc} => top
+s = Doc => top
-Doc ::= Header Body /ws => doc
-Header ::= "header" { kv */ ws } /ws => header
-Body ::= Section*/ws => body
-Section ::= SectionHeader Paragraph*/ws /ws => section
-SectionHeader ::= "==" SectionHeaderBody "=="
-SectionHeaderBody ::= "=" SectionHeaderBody "="
+Doc = {Header} Body /ws => doc
+Header = "header" { kv */ ws } /ws => header
+Body = Section*/ws => body
+Section = { SectionHeader Paragraph* /ws => section }
+SectionHeader = "==" SectionHeaderBody "=="
+SectionHeaderBody = "=" SectionHeaderBody "="
> ws text ws
-kv ::= word "=" text /ws => kv1
-
-num !::= [0-9]++
-Paragraph ::= { "\"\"" ws text } => "blockquote"
- > { "*" " " ws text } => "ul"
- > { "#" " " ws text } => "ol"
- > { num " " ws text } => "ol"
- > { "---" "-"* } => "hr"
- > { text } => "p"
-
-text ::= item */ ws
-pre ::= "[verbatim]" { ~[]+ } /ws => "verbatim" // FIXME doesn't work
-item ::= pre
- > email
- > structured
- > styled
- > "\"" text "\"" => quoted
- > [a-zA-Z0-9]++
- > 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?
+sp ! = " "**
+blank = sp "\n" sp "\n" ws
+
+kv = word "=" text /ws => kv1
+wp ! = w++
+num ! = [0-9]++
+Paragraph = { "\"\"" ws text } => "blockquote"
+ > { "---" "-"* } => "hr"
+ > { text } => "p"
+
+onums ! = nums (". "|") ")
+any ! = ~[]*
+
+uli = "* " (ws text &~ any (oli|uli)) => "li"
+oli = ("# "|onums) (ws text &~ any (oli|uli)) => "li"
+
+text = Item => text
+Itemx = ws Item
+ | ()
+Item = blockquote
+ > { uli+/ws => "ul" } Itemx => []
+ | { oli+/ws => "ol" } Itemx => []
+ > pre Itemx => []
+ > structured Itemx => []
+ > structuredx Itemx => []
+ > styled Itemx => []
+ > qtext Itemx => []
+ > (alphanum++ => stringify) Itemx => []
+ > symbol Itemx => []
+ > (sym++ => stringify) Itemx => []
+ > Paragraph Itemx => []
+
+blockquote = "\"\"" text "\"\"" => "blockquote"
+ | "\"\"" block => "blockquote"
+
+qtext = "\"" text "\"" => "quoted"
+pre = "[verbatim]" { ~[]+ } /ws => "verbatim" // FIXME doesn't work
+
+styled = "__" text "__" => underline
+ | "((" 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 //////////////////////////////////////////////////////////////////////////////
// subtypes of url (ftp, etc) as conjunctions, but the "master pattern"
// only gets parsed once
-urlpath ::= urlchar*
-username ::= [a-zA-Z0-9;/?:&=$\-_.+]++
-password ::= [a-zA-Z0-9;/?:&=$\-_.+]++
-urlchar ::= [a-zA-Z0-9;/?:&=$\-_.+@]
+urlpath = urlchar*
+username = [a-zA-Z0-9;/?:&=$\-_.+]++ => stringify
+password = [a-zA-Z0-9;/?:&=$\-_.+]++ => stringify
+urlchar = [a-zA-Z0-9;/?:&=$\-_.+@]
| "%" [0-9] [0-9] => "%"
-url ::= "mailto" ":" email
- > method "://" url_login? host (":" port)? ("/" urlpath)? => "url"
-url_login ::= username (":" password) "@" => "login"
-method ::= [+\-.a-z0-9]+
-port ::= [0-9]+
-
-domain ::= part +/ "."
-part ::= [a-zA-Z0-9\-]++ // interesting use of boolean grammars
+url = "mailto" ":" email
+ > method "://" url_login? host (":" nums)? ("/" urlpath)? => "url"
+url_login = username (":" password) "@" => "login"
+method = [+\-.a-z0-9]+ => stringify
+domain = (part +/ ".") -> ~"." => domain
+part = [a-zA-Z0-9\-]++ => stringify
+// interesting use of boolean grammars
// &~ ([\-0-9] ~[]* | ~[]* [\-0-9])
-email ::= username "@" host => email
-host ::= [0-9]+ "." [0-9]+ "." [0-9]+ "." [0-9]+ => "ip"
+email = username "@" host -> ~[.] => emailaddr
+nums = [0-9]++ => stringify
+host = nums "." nums "." nums "." nums => "ip"
| domain
// Tokens ///////////////////////////////////////////////////////////////////
-word ::= alphanum++
+word = alphanum++ => stringify
| quoted
-quoted ::= "\"" ((~[\"\\] | escaped)+) "\""
+quoted = "\"" ((~[\"\\] | escaped)+) "\""
| "\"\"" => ""
-escaped ::= "\\n" => "\n"
+escaped = "\\n" => "\n"
| "\\r" => "\r"
| "\\" ~[nr]
// Chars ///////////////////////////////////////////////////////////////
-alpha ::= [a-zA-Z]
-num ::= [0-9]
-alphanum ::= [a-zA-Z0-9]
-sym ::= ~[a-zA-Z0-9\ \r\n]
+alpha = [a-zA-Z]
+//num = [0-9]
+alphanum = [a-zA-Z0-9]
+sym = ~[a-zA-Z0-9\ \r\n=\">]