tests/tibdoc.g

   1 // interactions between !=> and &~ mean that I need to rethink the chartage
   2 // indentation styling...
   3 // literal blocks [[need to ignore bracing]] double-colon style?
   4 // definition -- by prior line indentation, like headings in the original structured text
   5 // tables
   6 // dropcap
   7 // output formats: latex, contex, ps, pdf, html, man, txt, rfc
   8
   9 // escapification
  10 // comment
  11 // math
  12 // image
  13 // figures
  14
  15 // "reference-style" links
  16 //
  17 //   this[1] is fun
  18 //
  19 //   [1] http://...
  20 //
  21
  22 // nonbreaking text?
  23 // degree: 15^o
  24 // Arrows: <- -> => <= <->
  25
  26 // textblocks:
  27 //  - attention, caution, danger, error, hint, important, note, tip, warning
  28 // definition
  29 // sidebar
  30 // figure-with-caption
  31 // epigraph (end-of-chapter note)
  32 // compound paragraph (??)
  33 // csv-table?
  34 // table of contents
  35 // header, footer
  36 // #include
  37
  38 // simple macros (#define) (\define)
  39
  40 // table representation
  41 //
  42 //  \table
  43 //     a bbb c
  44 //     ddd   e
  45 //
  46 //     [a] ...
  47 //     [b] ...
  48 //     [c] ...
  49
  50 // FIXME: these have to go at the top so they have their dropAll bit set before PreSequence.build...
  51 ws         = w**
  52 w          =  [\r\n\ ]
  53 nw         = ~[\r\n\ ]
  54
  55 //////////////////////////////////////////////////////////////////////////////
  56
  57 s                   = Doc
  58
  59 Doc                 = head:Header body:Body  /ws
  60 Header              = H:: { "header" { KeyVal */ ws } /ws }
  61 Body                = B:: {Section}*/ws
  62 Section             = SectionHeader Paragraph* /ws
  63 SectionHeader       = "==" SectionHeaderBody "=="
  64 SectionHeaderBody   =  "=" SectionHeaderBody "="
  65                     >      ws! alphanum++ ws!
  66
  67 sp       = " "**
  68 blank    = sp! "\n" sp! "\n" ws!
  69
  70 KeyVal       = key:word "=" val:text /ws
  71 wp           = w++
  72 num          = [0-9]++
  73
  74 Paragraph   = Blockquote:: { "\"\" "    text }
  75             > HR::         { "---" "-"*      }
  76             > P::          { text            }
  77
  78 onums        = nums (". "|") ")!
  79 any          = ~[]*
  80
  81 uli          =  "* "         (ws! text &~ any (oli|uli)!)
  82 oli          = ("# "|onums)! (ws! text &~ any (oli|uli)!)
  83
  84 text         = Item
  85 Itemx        = ws! Item
  86              | ()
  87 Item
  88 //*/ws
  89      =
  90 // "[]":: blockquote                 ((ws! Item) | ())
  91 //             >
  92  "[]":: { UL:: uli+/ws }           ((ws! Item) | ())
  93              | "[]":: { OL:: oli+/ws }           ((ws! Item) | ())
  94              > "[]":: pre                        ((ws! Item) | ())
  95              > "[]":: link                       ((ws! Item) | ())
  96              > "[]":: structured                 ((ws! Item) | ())
  97              > "[]":: styled                     ((ws! Item) | ())
  98              > "[]":: (Chars:: alphanum++)       ((ws! Item) | ())
  99              > "[]":: "\"" text "\""             ((ws! Item) | ())
 100              > "[]":: (Symbol:: sym++)           ((ws! Item) | ())
 101              > "[]":: { Block:: text }           ((ws! Item) | ())
 102
 103
 104 //blockquote   = Blockquote:: "\"\"" (block | text "\"\"")
 105
 106 pre          = Verbatim:: "[verbatim]" { ~[]+ } /ws   // FIXME doesn't work
 107
 108 styled       = Underline::     "__" text "__"
 109              | Footnote::      "((" text "))"
 110              | TT::            "[[" text "]]"
 111              | Citation::       "[" word "]"
 112              | Strikethrough:: "!!" text "!!"
 113              | Superscript::   "^^" (word|block)
 114              | Subscript::     ",," (word|block)
 115              | Smallcap::      "\\sc" block
 116              | Bold::           "++" text "++"
 117              | Keyword::         "!" (word|block)
 118              | Italic::         "**" text "**"
 119
 120 block         = { text }
 121
 122 link          = LinkText:: text:({ text })      "->" href:(url|email)
 123               > LinkChars:: text:alphanum++  ws! "->" href:(url|email)
 124
 125 structured    = command & "\\" ([a-zA-Z0-9]++)! block?
 126               > glyph
 127               > email
 128               > url
 129
 130 glyph        = Euro:: "(e)" | "(r)" | "(c)" | "(tm)" | emdash:: "--" | "..."
 131
 132 command      = Today::     "\\today"
 133              | LineBreak:: "\\br"
 134
 135 // URLs //////////////////////////////////////////////////////////////////////////////
 136
 137 // interesting opportunity to show off boolean grammars here: define other
 138 // subtypes of url (ftp, etc) as conjunctions, but the "master pattern"
 139 // only gets parsed once
 140
 141 urlpath      = urlchar*
 142 username     = [a-zA-Z0-9;/?:&=$\-_.+]++
 143 password     = [a-zA-Z0-9;/?:&=$\-_.+]++
 144 urlc         = [a-zA-Z0-9;/?:&=$\-_.+@]
 145 urlv         = urlc | [%]
 146 urlchar      = urlc
 147              | urlescape:: "%" [0-9] [0-9]
 148 url          = Mailto:: "mailto" ":"   email -> ~urlv
 149              > URL::
 150                   method:method
 151                   "://"
 152                   login:url_login?
 153                   host:host
 154                   port:(":" nums)?
 155                   path:("/" urlpath)?
 156                      -> ~urlv
 157 url_login    = Login:: username:username password:(":" password) "@"
 158 method       = [+\-.a-z0-9]+
 159 domain       = (part +/ ".") -> ~"."
 160 part         = [a-zA-Z0-9\-]++
 161 // interesting use of boolean grammars
 162 //            &~ ([\-0-9] ~[]* | ~[]* [\-0-9])
 163
 164 email        = user:username "@" host:host -> ~[.]
 165 nums         = [0-9]++
 166 host         = IP::  nums "." nums "." nums "." nums
 167              | DNS:: domain
 168
 169
 170
 171 // Tokens ///////////////////////////////////////////////////////////////////
 172
 173 word       = alphanum++
 174            | quoted
 175
 176 quoted     = "\"" ((~[\"\\] | escaped)+) "\""
 177            | "\"\"":: "\"\""
 178 escaped    = lf:: "\\n"
 179            | cr:: "\\r"
 180            | "\\" ~[nr]
 181
 182
 183 // Chars ///////////////////////////////////////////////////////////////
 184
 185 alpha      = [a-zA-Z]
 186 alphanum   = [a-zA-Z0-9]
 187 sym        = ~[a-zA-Z0-9\ \r\n=\">]
 188
 189