-x::="x"
// interactions between !=> and &~ mean that I need to rethink the chartage
// indentation styling...
// literal blocks [[need to ignore bracing]] double-colon style?
// image
// figures
+// "reference-style" links
+//
+// this[1] is fun
+//
+// [1] http://...
+//
+
+// consider ++bold++ and **italic**?
+// \br
+// nonbreaking text?
+// ellipsis detection (...)
+// degree: 15^o
+// Arrows: <- -> => <= <->
+
+// textblocks:
+// - attention, caution, danger, error, hint, important, note, tip, warning
+// definition
+// sidebar
+// figure-with-caption
+// epigraph (end-of-chapter note)
+// compound paragraph (??)
+// csv-table?
+// table of contents
+// header, footer
+// #include
+
+// simple macros (#define) (\define)
+// today's date
+
+// table representation
+//
+// \table
+// a bbb c
+// ddd e
+//
+// [a] ...
+// [b] ...
+// [c] ...
+
// 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
-
-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 "="
- > 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
-itemx ::= ws item | ()
-item ::= blockquote
- > pre itemx => []
- > structured itemx => []
- > structuredx itemx => []
- > styled itemx => []
- > qtext itemx => []
- > alphanum++ itemx => []
- > symbol itemx => []
-// > sym++ itemx => []
- > Paragraph itemx => []
-
-symbol ::= sym++
-
-blockquote ::= "\"\"" text "\"\"" => "blockquote"
- | "\"\"" block => "blockquote"
+s = top:: Doc
+
+Doc = doc:: {Header} Body /ws
+Header = header:: "header" { kv */ ws } /ws
+Body = body:: Section*/ws
+Section = { section:: SectionHeader Paragraph* /ws }
+SectionHeader = "==" SectionHeaderBody "=="
+SectionHeaderBody = "=" SectionHeaderBody "="
+ > !ws text !ws
+
+sp = " "**
+blank = !sp "\n" !sp "\n" !ws
+
+kv = kv1:: word "=" text /ws
+wp = w++
+num = [0-9]++
+Paragraph = blockquote:: { "\"\"" !ws text }
+ > hr:: { "---" "-"* }
+ > p:: { text }
+
+onums = nums !(". "|") ")
+any = ~[]*
+
+uli = li:: "* " (!ws text &~ any (oli|uli))
+oli = li:: ("# "|onums) (!ws text &~ any (oli|uli))
+
+//
+
+text = text:: Item
+Itemx = !ws Item
+ | ()
+Item = blockquote
+ > "[]":: { ul:: uli+/ws } Itemx
+ | "[]":: { ol:: oli+/ws } Itemx
+ > "[]":: pre Itemx
+ > "[]":: structured Itemx
+ > "[]":: structuredx Itemx
+ > "[]":: styled Itemx
+ > "[]":: qtext Itemx
+ > "[]":: (stringify:: alphanum++) Itemx
+ > "[]":: symbol Itemx
+ > "[]":: (stringify:: sym++) Itemx
+ > "[]":: Paragraph Itemx
+
+blockquote = blockquote:: "\"\"" text "\"\""
+ | blockquote:: "\"\"" block
-qtext ::= "\"" text "\"" => "quoted"
-pre ::= "[verbatim]" { ~[]+ } /ws => "verbatim" // FIXME doesn't work
+qtext = quoted:: "\"" text "\""
+pre = verbatim:: "[verbatim]" { ~[]+ } /ws // FIXME doesn't work
-styled ::= "__" text "__" => ul
- | "((" text "))" => footnote
- | ( "[[" text "]]" => tt
- > "[" word "]" => citation
+styled = underline:: "__" text "__"
+ | footnote:: "((" text "))"
+ | ( tt:: "[[" text "]]"
+ > citation:: "[" word "]"
)
- | "!!" 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
+ | strikethrough:: "!!" text "!!"
+ | superscript:: "^^" (word|block)
+ | subscript:: ",," (word|block)
+ | smallcap:: "\\sc" block
+ | bold:: "**" text "**"
+ | keyword:: "!" (word|block)
+ > it:: "*" text "*"
+
+//
+
+block = { text }
+structured = link:: { text } "->" (url|email)
//> alphanum++ "->" (url|email) => link
-structuredx ::= glyph
+structuredx = glyph
> email
> url
-glyph ::= "(r)" | "(c)" | "(tm)" | "--" // euro symbol?
- | "\\today" -> ~[a-z] => today
+glyph = "(r)" | "(c)" | "(tm)" | "--" // euro symbol?
+ | today:: "\\today" -> ~[a-z]
// 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;/?:&=$\-_.+@]
- | "%" [0-9] [0-9] => "%"
-url ::= "mailto" ":" email
- > method "://" url_login? host (":" port)? ("/" urlpath)? => "url"
-url_login ::= username (":" password) "@" => "login"
-method ::= [+\-.a-z0-9]+
-port ::= [0-9]+
-
-domain ::= (part +/ ".") -> ~"."
-part ::= [a-zA-Z0-9\-]++ // interesting use of boolean grammars
+urlpath = urlchar*
+username = stringify:: [a-zA-Z0-9;/?:&=$\-_.+]++
+password = stringify:: [a-zA-Z0-9;/?:&=$\-_.+]++
+urlchar = [a-zA-Z0-9;/?:&=$\-_.+@]
+ | "%":: "%" [0-9] [0-9]
+url = "mailto" ":" email
+ > url:: method "://" url_login? host (":" nums)? ("/" urlpath)?
+url_login = login:: username (":" password) "@"
+method = stringify:: [+\-.a-z0-9]+
+domain = domain:: (part +/ ".") -> ~"."
+part = stringify:: [a-zA-Z0-9\-]++
+// interesting use of boolean grammars
// &~ ([\-0-9] ~[]* | ~[]* [\-0-9])
-email ::= username "@" host -> ~[.] => emailaddr
-host ::= [0-9]+ "." [0-9]+ "." [0-9]+ "." [0-9]+ => "ip"
+email = emailaddr:: username "@" host -> ~[.]
+nums = stringify:: [0-9]++
+host = ip:: nums "." nums "." nums "." nums
| domain
// Tokens ///////////////////////////////////////////////////////////////////
-word ::= alphanum++
+word = stringify:: alphanum++
| quoted
-quoted ::= "\"" ((~[\"\\] | escaped)+) "\""
- | "\"\"" => ""
-escaped ::= "\\n" => "\n"
- | "\\r" => "\r"
+quoted = "\"" ((~[\"\\] | escaped)+) "\""
+ | "":: "\"\""
+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=\">]