tests/tibdoc.g

   1 x  ="x"
   2 // interactions between !=> and &~ mean that I need to rethink the chartage
   3 // indentation styling...
   4 // literal blocks [[need to ignore bracing]] double-colon style?
   5 // definition -- by prior line indentation, like headings in the original structured text
   6 // tables
   7 // dropcap
   8 // output formats: latex, contex, ps, pdf, html, man, txt, rfc
   9
  10 // escapification
  11 // comment
  12 // math
  13 // image
  14 // figures
  15
  16 // FIXME: these have to go at the top so they have their dropAll bit set before PreSequence.build...
  17 ws      !  = w**
  18 w       !  =  [\r\n\ ]
  19 nw         = ~[\r\n\ ]
  20
  21 //////////////////////////////////////////////////////////////////////////////
  22
  23 s                   = Doc => top
  24
  25 Doc                 = {Header} Body                   /ws => doc
  26 Header              = "header" { kv */ ws }           /ws => header
  27 Body                = Section*/ws                         => body
  28 Section             = { SectionHeader Paragraph* /ws => section }
  29 SectionHeader       = "==" SectionHeaderBody "=="
  30 SectionHeaderBody   =  "=" SectionHeaderBody "="
  31                     >      ws text ws
  32
  33 sp    !  = " "**
  34 blank    = sp "\n" sp "\n" ws
  35
  36 kv           = word "=" text /ws => kv1
  37 wp !  = w++
  38 num !  = [0-9]++
  39 Paragraph    = { "\"\"" ws  text }  => "blockquote"
  40              > { "---" "-"*      }  => "hr"
  41              > { text }             => "p"
  42
  43 onums     !  = nums (". "|") ")
  44 any       !  = ~[]*
  45
  46 uli          = "* "         (ws text &~ any (oli|uli)) => "li"
  47 oli          = ("# "|onums) (ws text &~ any (oli|uli)) => "li"
  48
  49 text         = Item => text
  50 Itemx        = ws Item
  51              | ()
  52 Item         = blockquote
  53              > { uli+/ws => "ul" }       Itemx  => []
  54              | { oli+/ws => "ol" }       Itemx  => []
  55              > pre                       Itemx  => []
  56              > structured                Itemx  => []
  57              > structuredx               Itemx  => []
  58              > styled                    Itemx  => []
  59              > qtext                     Itemx  => []
  60              > (alphanum++ => stringify) Itemx  => []
  61              > symbol                    Itemx  => []
  62              > (sym++ => stringify)      Itemx  => []
  63              > Paragraph                 Itemx  => []
  64
  65 blockquote   = "\"\"" text "\"\""        => "blockquote"
  66              | "\"\"" block              => "blockquote"
  67
  68 qtext        = "\"" text "\""            => "quoted"
  69 pre          = "[verbatim]" { ~[]+ } /ws => "verbatim"   // FIXME doesn't work
  70
  71 styled       = "__" text "__"      => underline
  72              | "((" text "))"      => footnote
  73              | ( "[[" text "]]"    => tt
  74                >  "[" word "]"     => citation
  75                )
  76              | "!!" text "!!"      => strikethrough
  77              | "^^" (word|block)   => superscript
  78              | ",," (word|block)   => subscript
  79              | "\\sc" block        => smallcap
  80              | "**" text "**"      => bold
  81              | "!" (word|block)    => keyword
  82              >  "*" text "*"       => it
  83
  84 block   = { text }
  85 structured   = { text } "->" (url|email) => link
  86               //> alphanum++ "->" (url|email) => link
  87 structuredx   = glyph
  88               > email
  89               > url
  90
  91 glyph        = "(r)" | "(c)" | "(tm)" | "--"  // euro symbol?
  92              | "\\today" -> ~[a-z] => today
  93
  94
  95 // URLs //////////////////////////////////////////////////////////////////////////////
  96
  97 // interesting opportunity to show off boolean grammars here: define other
  98 // subtypes of url (ftp, etc) as conjunctions, but the "master pattern"
  99 // only gets parsed once
 100
 101 urlpath      = urlchar*
 102 username     = [a-zA-Z0-9;/?:&=$\-_.+]++ => stringify
 103 password     = [a-zA-Z0-9;/?:&=$\-_.+]++ => stringify
 104 urlchar      = [a-zA-Z0-9;/?:&=$\-_.+@]
 105              | "%" [0-9] [0-9]       => "%"
 106 url          = "mailto" ":"   email
 107              > method "://" url_login? host (":" nums)? ("/" urlpath)?     => "url"
 108 url_login    = username (":" password) "@"                                 => "login"
 109 method       = [+\-.a-z0-9]+ => stringify
 110 domain       = (part +/ ".") -> ~"." => domain
 111 part         = [a-zA-Z0-9\-]++  => stringify
 112 // interesting use of boolean grammars
 113 //            &~ ([\-0-9] ~[]* | ~[]* [\-0-9])
 114
 115 email        = username "@" host -> ~[.] => emailaddr
 116 nums         = [0-9]++ => stringify
 117 host         = nums "." nums "." nums "." nums => "ip"
 118              | domain
 119
 120
 121
 122 // Tokens ///////////////////////////////////////////////////////////////////
 123
 124 word       = alphanum++ => stringify
 125            | quoted
 126
 127 quoted     = "\"" ((~[\"\\] | escaped)+) "\""
 128            | "\"\"" => ""
 129 escaped    = "\\n" => "\n"
 130            | "\\r" => "\r"
 131            | "\\" ~[nr]
 132
 133
 134 // Chars ///////////////////////////////////////////////////////////////
 135
 136 alpha      = [a-zA-Z]
 137 //num        = [0-9]
 138 alphanum   = [a-zA-Z0-9]
 139 sym        = ~[a-zA-Z0-9\ \r\n=\">]
 140
 141