fixed tibdoc
[sbp.git] / tests / tibdoc.g
index 5926194..6287afa 100644 (file)
@@ -1,4 +1,4 @@
-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
@@ -6,80 +6,132 @@ x::="x"
 // 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
 
+// "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]++ => "stringify"
-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
-             > 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?
+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        = quoted:: "\"" text "\""
+pre          = verbatim:: "[verbatim]" { ~[]+ } /ws   // FIXME doesn't work
+
+styled       = underline:: "__" text "__"      
+             | footnote:: "((" text "))"      
+             | ( tt:: "[[" text "]]"    
+             > citation::   "[" word "]"     
+               )
+             | 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
+              > email
+              > url
+
+glyph        = "(r)" | "(c)" | "(tm)" | "--"  // euro symbol?
+             | today:: "\\today" -> ~[a-z]
 
 
 // URLs //////////////////////////////////////////////////////////////////////////////
@@ -88,44 +140,44 @@ glyph      ::= "(r)" | "(c)" | "(tm)"  // euro symbol?
 // subtypes of url (ftp, etc) as conjunctions, but the "master pattern"
 // only gets parsed once
 
-urlpath    ::= urlchar*
-username   ::= [a-zA-Z0-9;/?:&=$\-_.+]++ => "stringify"
-password   ::= [a-zA-Z0-9;/?:&=$\-_.+]++ => "stringify"
-urlchar    ::= [a-zA-Z0-9;/?:&=$\-_.+@]
-             | "%" [0-9] [0-9]       => "%"
-url        ::= "mailto" ":"   email
-             > method "://" url_login? host (":" port)? ("/" urlpath)?     => "url"
-url_login  ::= username (":" password) "@" => "login"
-method     ::= [+\-.a-z0-9]+ 
-port       ::= [0-9]+
-
-domain     ::= part +/ "."
-part       ::= [a-zA-Z0-9\-]++ => "stringify"   // interesting use of boolean grammars
+urlpath      = urlchar*
+username     = stringify:: [a-zA-Z0-9;/?:&=$\-_.+]++
+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      => email
-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++ => "stringify"
+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=\">]