tentative checkpoint
[sbp.git] / tests / tibdoc.g
index 99bab83..ec31d30 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 +6,127 @@ 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://...
+//
+
+// nonbreaking text?
+// 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)
+
+// 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
-
-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                   = Doc
+
+Doc                 = Doc:: head:{Header} body:Body   /ws
+Header              = Header:: "header" attrs:{ kv */ ws }     /ws
+Body                = Body:: sections:(Section*/ws)
+Section             = { Section:: header:SectionHeader paragraphs:Paragraph* /ws }
+SectionHeader       = "==" SectionHeaderBody "=="
+SectionHeaderBody   =  "=" SectionHeaderBody "="
+                    >      !ws (Chars:: text:alphanum++) !ws
+
+sp       = " "**
+blank    = !sp "\n" !sp "\n" !ws
+
+kv           = kv:: key:word "=" val:text /ws
+wp           = w++
+num          = [0-9]++
+Paragraph    = Blockquote:: { "\"\" "    text }
+             > HR::         { "---" "-"*      }
+             >              { P:: t:text }
+
+onums        = nums !(". "|") ")
+any          = ~[]*
+
+uli          =  "* "         (!ws text &~ any (oli|uli))
+oli          = !("# "|onums) (!ws text &~ any (oli|uli))
+
+text         = Item
+Itemx        = !ws Item
+             | ()
+Item         = blockquote
+             > "[]":: { UL:: uli+/ws }           Itemx
+             | "[]":: { OL:: oli+/ws }           Itemx
+             > "[]":: pre                        Itemx
+             > "[]":: link                       Itemx
+             > "[]":: structured                 Itemx
+             > "[]":: styled                     Itemx
+             > "[]":: (Chars:: text:alphanum++)  Itemx
+             > "[]":: "\"" text "\""             Itemx
+             > "[]":: symbol                     Itemx
+             > "[]":: (Symbol:: sym++)           Itemx
+             > "[]":: Paragraph                  Itemx
+
+blockquote   = Blockquote:: "\"\"" text "\"\""
+             | Blockquote:: "\"\"" block
+             
+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)    
+             | Italic::         "**" text "**"
+
+block         = { text }
+
+link          = Link:: text:({ text })      "->" href:(url|email)
+              > Link:: text:alphanum++  !ws "->" href:(url|email)
+
+structured    = command & "\\" [a-zA-Z0-9]++ block?
+              > glyph
+              > email
+              > url
+
+glyph        = Euro:: "(e)" | "(r)" | "(c)" | "(tm)" | emdash:: "--" | "..."
+
+command      = Today::     "\\today"
+             | LineBreak:: "\\br"
 
 // URLs //////////////////////////////////////////////////////////////////////////////
 
@@ -87,45 +134,46 @@ 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;/?:&=$\-_.+@]
-             | "%" [0-9] [0-9]       => "%"
-url        ::= "mailto" ":"   email
-             > method "://" url_login? host (":" port)? ("/" urlpath)?     => "url"
-url_login  ::= username (":" password) "@" => "login"
-method     ::= [+\-.a-z0-9]+ 
-port       ::= [0-9]+
-
-domain     ::= part +/ "."
-part       ::= [A-Za-z0-9\-]++ => "stringify"
+urlpath      = urlchar*
+username     = [a-zA-Z0-9;/?:&=$\-_.+]++
+password     = [a-zA-Z0-9;/?:&=$\-_.+]++
+urlc         = [a-zA-Z0-9;/?:&=$\-_.+@]
+urlv         = urlc | [%]
+urlchar      = urlc
+             | urlescape:: "%" [0-9] [0-9]
+url          = "mailto" ":"   email -> ~urlv
+             > URL:: method:method "://" login:url_login? host:host port:(":" nums)? path:("/" urlpath)? -> ~urlv
+url_login    = Login:: username:username password:(":" password) "@"
+method       = [+\-.a-z0-9]+
+domain       = (part +/ ".") -> ~"."
+part         = [a-zA-Z0-9\-]++
+// interesting use of boolean grammars
 //            &~ ([\-0-9] ~[]* | ~[]* [\-0-9])
 
-email      ::= username "@" host      => email
-host       ::= domain
-             | [0-9]+ "." [0-9]+ "." [0-9]+ "." [0-9]+ => "ip"
+email        = Email:: user:username "@" host:host -> ~[.]
+nums         = [0-9]++
+host         = IP::  nums "." nums "." nums "." nums
+             | DNS:: domain
 
 
 
 // Tokens ///////////////////////////////////////////////////////////////////
 
-word     ::= alphanum++ => "stringify"
+word       = alphanum++
            | quoted
 
-quoted   ::= "\"" ((~[\"\\] | escaped)+) "\""
-           | "\"\"" => ""
-escaped  ::= "\\n" => "\n"
-           | "\\r" => "\r"
+quoted     = "\"" ((~[\"\\] | escaped)+) "\""
+           | "":: "\"\""
+escaped    = lf:: "\\n"
+           | cr:: "\\r"
            | "\\" ~[nr]
 
 
 // Chars ///////////////////////////////////////////////////////////////
 
-alpha    ::= [a-zA-Z]
-num     !::= [0-9]++ => "stringify"
-//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=\">]