tentative checkpoint
[sbp.git] / tests / tibdoc.g
index 442e203..ec31d30 100644 (file)
@@ -1,4 +1,3 @@
-x  ="x"
 // interactions between !=> and &~ mean that I need to rethink the chartage
 // indentation styling...
 // literal blocks [[need to ignore bracing]] double-colon style?
@@ -20,10 +19,7 @@ x  ="x"
 //   [1] http://...
 //
 
-// consider ++bold++ and **italic**?
-// \br
 // nonbreaking text?
-// ellipsis detection (...)
 // degree: 15^o
 // Arrows: <- -> => <= <->
 
@@ -40,7 +36,6 @@ x  ="x"
 // #include
 
 // simple macros (#define) (\define)
-// today's date
 
 // table representation
 //  
@@ -53,83 +48,85 @@ x  ="x"
 //     [c] ...
 
 // FIXME: these have to go at the top so they have their dropAll bit set before PreSequence.build...
-ws      !  = w**
-w       !  =  [\r\n\ ]
+ws         = w**
+w          =  [\r\n\ ]
 nw         = ~[\r\n\ ]
 
 //////////////////////////////////////////////////////////////////////////////
 
-s                   = Doc => top
+s                   = Doc
 
-Doc                 = {Header} Body                   /ws => doc
-Header              = "header" { kv */ ws }           /ws => header
-Body                = Section*/ws                         => body
-Section             = { SectionHeader Paragraph* /ws => section }
+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 text ws
+                    >      !ws (Chars:: text:alphanum++) !ws
 
-sp    !  = " "**
-blank    = sp "\n" sp "\n" ws
+sp       = " "**
+blank    = !sp "\n" !sp "\n" !ws
 
-kv           = word "=" text /ws => kv1
-wp !  = w++
-num !  = [0-9]++
-Paragraph    = { "\"\"" ws  text }  => "blockquote"
-             > { "---" "-"*      }  => "hr"
-             > { text }             => "p"
+kv           = kv:: key:word "=" val:text /ws
+wp           = w++
+num          = [0-9]++
+Paragraph    = Blockquote:: { "\"\" "    text }
+             > HR::         { "---" "-"*      }
+             >              { P:: t:text }
 
-onums     !  = nums (". "|") ")
-any       !  = ~[]*
+onums        = nums !(". "|") ")
+any          = ~[]*
 
-uli          = "* "         (ws text &~ any (oli|uli)) => "li"
-oli          = ("# "|onums) (ws text &~ any (oli|uli)) => "li"
+uli          =  "* "         (!ws text &~ any (oli|uli))
+oli          = !("# "|onums) (!ws text &~ any (oli|uli))
 
-text         = Item => text
-Itemx        = ws Item
+text         = Item
+Itemx        = !ws Item
              | ()
 Item         = blockquote
-             > { uli+/ws => "ul" }       Itemx  => []
-             | { oli+/ws => "ol" }       Itemx  => []
-             > pre                       Itemx  => []
-             > structured                Itemx  => []
-             > structuredx               Itemx  => []
-             > styled                    Itemx  => []
-             > qtext                     Itemx  => []
-             > (alphanum++ => stringify) Itemx  => []
-             > symbol                    Itemx  => []
-             > (sym++ => stringify)      Itemx  => []
-             > Paragraph                 Itemx  => []
-
-blockquote   = "\"\"" text "\"\""        => "blockquote"
-             | "\"\"" block              => "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
              
-qtext        = "\"" text "\""            => "quoted"
-pre          = "[verbatim]" { ~[]+ } /ws => "verbatim"   // FIXME doesn't work
-
-styled       = "__" text "__"      => underline
-             | "((" text "))"      => footnote
-             | ( "[[" text "]]"    => tt
-               >  "[" word "]"     => citation
-               )
-             | "!!" 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
-              //> alphanum++ "->" (url|email) => link
-structuredx   = glyph
+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        = "(r)" | "(c)" | "(tm)" | "--"  // euro symbol?
-             | "\\today" -> ~[a-z] => today
+glyph        = Euro:: "(e)" | "(r)" | "(c)" | "(tm)" | emdash:: "--" | "..."
 
+command      = Today::     "\\today"
+             | LineBreak:: "\\br"
 
 // URLs //////////////////////////////////////////////////////////////////////////////
 
@@ -138,35 +135,37 @@ glyph        = "(r)" | "(c)" | "(tm)" | "--"  // euro symbol?
 // only gets parsed once
 
 urlpath      = urlchar*
-username     = [a-zA-Z0-9;/?:&=$\-_.+]++ => stringify
-password     = [a-zA-Z0-9;/?:&=$\-_.+]++ => stringify
-urlchar      = [a-zA-Z0-9;/?:&=$\-_.+@]
-             | "%" [0-9] [0-9]       => "%"
-url          = "mailto" ":"   email
-             > method "://" url_login? host (":" nums)? ("/" urlpath)?     => "url"
-url_login    = username (":" password) "@"                                 => "login"
-method       = [+\-.a-z0-9]+ => stringify
-domain       = (part +/ ".") -> ~"." => domain
-part         = [a-zA-Z0-9\-]++  => stringify
+username     = [a-zA-Z0-9;/?:&=$\-_.+]++
+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 -> ~[.] => emailaddr
-nums         = [0-9]++ => stringify
-host         = nums "." nums "." nums "." nums => "ip"
-             | domain
+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"
+           | "":: "\"\""
+escaped    = lf:: "\\n"
+           | cr:: "\\r"
            | "\\" ~[nr]