Fix syntax spec in ABNF.

3 files changed, 59 insertions, 83 deletions

diff --git a/spec/syntax.abnf b/spec/syntax.abnf
index 6ee024d..a083eda 100644
--- a/spec/syntax.abnf
+++ b/spec/syntax.abnf
@@ -1,23 +1,39 @@
-Unit          = *Blank Datum [ Blank ]
+;This file follows strict ABNF rules and can be used with abnfgen.
 
 
+File          = [Unit] *( Blank Unit ) *Blank [Trail]
+
+
+Unit          = *Blank Datum
+
 Blank         = HTAB / LF / %x0b / %x0c / CR / Comment
 
-Datum         = OneDatum *( [JoinChar] OneDatum )
+Trail         = SkipLine / SkipUnit
+
+
+Comment       = SkipLine LF / SkipUnit Blank
+
+SkipLine      = ';' [ SkipLStart *AnyButLF ]
 
-JoinChar      = '.' / ':'
+SkipUnit      = ';' '~' Unit
 
 
-Comment       = ';' ( SkipUnit / SkipLine [LF] )
+SkipLStart    = %x00-09 / %x0b-7d / %x7f-ff
+              ; any but LF or '~'
 
-SkipUnit      = '~' Unit
+AnyButLF      = %x00-09 / %x0b-ff
 
-SkipLine      = *( %x00-09 / %x0b-ff )  ; any but LF
 
+Datum         = SingleDatum
+              / JoinedDatum *( [ '.' / ':' ] JoinedDatum )
 
-OneDatum      = BareString / CladDatum
 
-BareString    = 1*BareChar
+SingleDatum   = BareString / CladDatum / DottedString
+
+JoinedDatum   = BareString / CladDatum
+
+
+BareString    = BareChar *( BareChar / Numeric )
 
 CladDatum     = '|' *( PipeStrChar / '\' StringEsc ) '|'
               / '"' *( QuotStrChar / '\' StringEsc ) '"'
@@ -25,28 +41,37 @@ CladDatum     = '|' *( PipeStrChar / '\' StringEsc ) '|'
               / '(' List ')' / '[' List ']' / '{' List '}'
               / "'" Datum / '`' Datum / ',' Datum
 
+DottedString  = ( '.' / Numeric ) *( '.' / Numeric / BareChar )
 
-BareChar      = ALPHA / DIGIT
-              / '!' / '$' / '%' / '*' / '+' / '-' / '.' / '/'
-              / '<' / '=' / '>' / '?' / '@' / '^' / '_' / '~'
 
+BareChar      = '!' / '$' / '%' / '*' / '/' / '<' / '=' / '>'
+              / '?' / '@' / '^' / '_' / '~' / ALPHA
 
-PipeStrChar   = %x00-5b / %x5d-7b / %x7d-ff  ; any but | or \
+Numeric       = '+' / '-' / DIGIT
 
-QuotStrChar   = %x00-21 / %x23-5b / %x5d-ff  ; any but " or \
+
+PipeStrChar   = %x00-5b / %x5d-7b / %x7d-ff
+              ; any but '|' or '\'
+
+QuotStrChar   = %x00-21 / %x23-5b / %x5d-ff
+              ; any but '"' or '\'
 
 HashExpr      = Rune [ '\' BareString / CladDatum ]
               / '\' BareString
               / '%' Label ( '%' / '=' Datum )
               / CladDatum
 
-List          = *Unit [ '&' Unit ] *Blank
+List          = [Unit] *( Blank Unit ) *Blank [Tail] [SkipUnit]
+
+Tail          = '&' Unit *Blank
 
 
 StringEsc     = '\' / '|' / '"' / *( HTAB / SP ) LF *( HTAB / SP )
               / 'a' / 'b' / 't' / 'n' / 'v' / 'f' / 'r' / 'e'
-              / 'x' 1*( 2*HEXDIG ) ';'
-              / 'u' 1*6( HEXDIG ) ';'
+              / 'x' 1*( 2HEXDIG ) ';'
+              / 'u' 1*5HEXDIG ';'
+              / 'u' '0' 1*5HEXDIG ';'
+              / 'u' '1' '0' 1*4HEXDIG ';'
 
 
 Rune          = ALPHA *5( ALPHA / DIGIT )
diff --git a/spec/syntax.md b/spec/syntax.md
index affa7a1..7f3561c 100644
--- a/spec/syntax.md
+++ b/spec/syntax.md
@@ -6,7 +6,7 @@ We use a BNF notation with the following rules:
   followed by `bar`.
 
 * Expressions may be followed by `?`, `*`, `+`, `{N}`, or `{N,M}`,
-  which have meanings analogous to regular expressions.
+  which have the same meanings as in regular expressions.
 
 * The syntax `[foo]` is shorthand for `(foo)?`.
 
@@ -20,21 +20,24 @@ We use a BNF notation with the following rules:
 
 * Ranges of terminal values are expressed as `x...y` (inclusive).
 
-* ABNF "core rules" like `ALPHA` and `HEXDIG` are supported, with the
-  addition of EOF to explicitly demarcate the end of the byte stream.
+* ABNF "core rules" like `ALPHA` and `HEXDIG` are supported.
 
-* There is no ambiguity, backtracking, or look-ahead beyond one byte.
+* There is no ambiguity, or look-ahead / backtracking beyond one byte.
   Rules match left to right, depth-first, and greedy.  As soon as the
   input matches the first terminal of a rule, it must match that rule
   to the end or it is considered a syntax error.
 
 The last rule means that the BNF is very simple to translate to code.
+It also probably makes it equivalent to PEG.
 
-The parser consumes one `unit` from an input stream every time it's
-called; it returns the `datum` therein, or EOF.
+The parser consumes one `Unit` from an input stream every time it's
+called; it returns the `Datum` therein, or EOF.  The final optional
+`Blank` represents the fact that the parser will consume one more
+blank at the end if it finds one; this is because `Datum` is not
+self-closing so the parser has to check if it goes on.
 
 ```
-Unit          : Blank* ( Datum [Blank] | EOF )
+Unit          : Blank* [ Datum [Blank] ]
 
 
 Blank         : 9...13 | Comment
@@ -44,16 +47,17 @@ Datum         : OneDatum ( [JoinChar] OneDatum )*
 JoinChar      : '.' | ':'
 
 
-Comment       : ';' ( SkipUnit | SkipLine [LF] )
+Comment       : ';' ( SkipUnit | SkipLine )
 
 SkipUnit      : '~' Unit
 
-SkipLine      : ( ~LF )*
+SkipLine      : ( ~LF )* [LF]
 
 
 OneDatum      : BareString | CladDatum
 
-BareString    : BareChar+
+BareString    : ( '.' | '+' | '-' | DIGIT ) ( BareChar | '.' )*
+              | BareChar+
 
 CladDatum     : '|' ( PipeStrChar | '\' StringEsc )* '|'
               | '"' ( QuotStrChar | '\' StringEsc )* '"'
@@ -63,8 +67,9 @@ CladDatum     : '|' ( PipeStrChar | '\' StringEsc )* '|'
 
 
 BareChar      : ALPHA | DIGIT
-              | '!' | '$' | '%' | '*' | '+' | '-' | '.' | '/'
-              | '<' | '=' | '>' | '?' | '@' | '^' | '_' | '~'
+              | '!' | '$' | '%' | '*' | '+'
+              | '-' | '/' | '<' | '=' | '>'
+              | '?' | '@' | '^' | '_' | '~'
 
 
 PipeStrChar   : ~( '|' | '\' )
@@ -76,7 +81,7 @@ HashExpr      : Rune [ '\' BareString | CladDatum ]
               | '%' Label ( '%' | '=' Datum )
               | CladDatum
 
-List          : Unit* [ '&' Unit ] Blank*
+List          : Unit* [ Blank* '&' Unit ] Blank*
 
 
 StringEsc     : '\' | '|' | '"' | ( HTAB | SP )* LF ( HTAB | SP )*
diff --git a/spec/syntax.zbnf b/spec/syntax.zbnf
deleted file mode 100644
index d920845..0000000
--- a/spec/syntax.zbnf
+++ /dev/null
@@ -1,54 +0,0 @@
-Unit          : Blank* ( Datum [Blank] | EOF )
-
-
-Blank         : 9...13 | Comment
-
-Datum         : OneDatum ( [JoinChar] OneDatum )*
-
-JoinChar      : '.' | ':'
-
-
-Comment       : ';' ( SkipUnit | SkipLine [LF] )
-
-SkipUnit      : '~' Unit
-
-SkipLine      : ( ~LF )*
-
-
-OneDatum      : BareString | CladDatum
-
-BareString    : BareChar+
-
-CladDatum     : '|' ( PipeStrChar | '\' StringEsc )* '|'
-              | '"' ( QuotStrChar | '\' StringEsc )* '"'
-              | '#' HashExpr
-              | '(' List ')' | '[' List ']' | '{' List '}'
-              | "'" Datum | '`' Datum | ',' Datum
-
-
-BareChar      : ALPHA | DIGIT
-              | '!' | '$' | '%' | '*' | '+' | '-' | '.' | '/'
-              | '<' | '=' | '>' | '?' | '@' | '^' | '_' | '~'
-
-
-PipeStrChar   : ~( '|' | '\' )
-
-QuotStrChar   : ~( '"' | '\' )
-
-HashExpr      : Rune [ '\' BareString | CladDatum ]
-              | '\' BareString
-              | '%' Label ( '%' | '=' Datum )
-              | CladDatum
-
-List          : Unit* [ '&' Unit ] Blank*
-
-
-StringEsc     : '\' | '|' | '"' | ( HTAB | SP )* LF ( HTAB | SP )*
-              | 'a' | 'b' | 't' | 'n' | 'v' | 'f' | 'r' | 'e'
-              | 'x' ( HEXDIG{2} )+ ';'
-              | 'u' HEXDIG{1,6} ';'
-
-
-Rune          : ALPHA ( ALPHA | DIGIT ){0,5}
-
-Label         : HEXDIG{1,12}