expand PEG definition; avoid too generic NAME

1 files changed, 50 insertions, 6 deletions

diff --git a/syntax/def.peg b/syntax/def.peg
index e40610f..640958f 100644
--- a/syntax/def.peg
+++ b/syntax/def.peg
@@ -48,9 +48,53 @@ NEWLINE        <- '\r\n'
                 / '\r'
 
 # Semantic Markdown
-AnnotatedWordsX <- '[' Words ']'
-                   ('{' (SemWords / ![{}] PlainWords) '}')
-SemWords        <- SEMPREFIX '<' ![<>] PRINTABLES '>'
-                 / SEMPREFIX Curie
-Curie           <- NAME? ':' NAME?
-SEMPREFIX       <- [.#]
+# @see <https://source.jones.dk/semantic-markdown/about/>
+AnnotatedWordsX  <- '[' Words ']'
+                    ('{' (SemWords / ![{}] PlainWords) '}')
+SemWords         <- SEMPREFIX '<' ![<>] PRINTABLES '>'
+                  / SEMPREFIX Curie
+SEMPREFIX        <- [.#]
+
+# RDF CURIE
+# @see <https://www.w3.org/TR/rdfa-core/#s_curies>
+Curie          <- (CuriePrefix? ':')? CurieReference
+CuriePrefix    <- NCName
+CurieReference <- (IriPathAbsolute / IriPathRootless / IriPathEmpty)
+                  ('?' IriQuery)? ('#' IriFragment)?
+
+# RFC3987 - Internationalized Resource Identifiers (IRIs)
+# @see <https://tools.ietf.org/html/rfc3987>
+IriPathAbsolute <- '/' (IriSegment ('/' IriSegment?)*)*
+IriPathRootless <- IriSegment ('/' IriSegment?)*
+IriPathEmpty    <- !IriPathChar
+IriSegment      <- IriPathChar+
+IriQuery        <- (IriPathChar / IriPrivate / '/' / '?')*
+IriFragment     <- (IriPathChar / '/' / '?')*
+IriPathChar     <- IriUnreserved / PctEncoded / SubDelims / ':' / '@'
+IriUnreserved   <- [A-Za-z] / [0-9] / '-' / '.' / '_' / '~' / UcsChar
+PctEncoded      <- '%' [0-9A-F] [0-9A-F]
+SubDelims       <- '!' / '$' / '&' / "'" / '(' / ')'
+                 / '*' / '+' / ',' / ';' / '='
+UcsChar         <- [#xA0-#xD7FF] / [#xF900-#xFDCF] / [#xFDF0-#xFFEF]
+                 / [#x10000-#x1FFFD] / [#x20000-#x2FFFD] / [#x30000-#x3FFFD]
+                 / [#x40000-#x4FFFD] / [#x50000-#x5FFFD] / [#x60000-#x6FFFD]
+                 / [#x70000-#x7FFFD] / [#x80000-#x8FFFD] / [#x90000-#x9FFFD]
+                 / [#xA0000-#xAFFFD] / [#xB0000-#xBFFFD] / [#xC0000-#xCFFFD]
+                 / [#xD0000-#xDFFFD] / [#xE1000-#xEFFFD]
+IriPrivate      <- [#xE000-#xF8FF] / [#xF0000-#xFFFFD] / [#x100000-#x10FFFD]
+
+# XML NCName
+# @see <https://www.w3.org/TR/2009/REC-xml-names-20091208/#NT-NCName>
+NCName <- !':' XMLName
+
+# XML Name
+# @see <https://www.w3.org/TR/REC-xml/#NT-Name>
+XMLName          <- NameStartChar NameChar*
+XMLNameChar      <- NameStartChar / "-" / "." / [0-9]
+                  / #xB7 / [#x0300-#x036F] / [#x203F-#x2040]
+XMLNAMESTARTCHAR <- ":" / "_" / [A-Z] / [a-z]
+                  / [#xC0-#xD6] / [#xD8-#xF6]
+                  / [#xF8-#x2FF] / [#x370-#x37D] / [#x37F-#x1FFF]
+                  / [#x200C-#x200D] / [#x2070-#x218F] / [#x2C00-#x2FEF]
+                  / [#x3001-#xD7FF] / [#xF900-#xFDCF] / [#xFDF0-#xFFFD]
+                  / [#x10000-#xEFFFF]