Advanced BNF with regex¶

URI                        = <per RFC 3986>
UTF-GRAPHIC                = `\p{L}|\p{LC}|\p{M}|\p{N}|\p{S}|\p{Zs}`
UTF-FORMAT                 = `\p{Cf}`
UTF-SURROGATE              = `\p{Cs}`
UTF-CONTROL                = `\p{Cc}`
UTF-SPACE                  = `\p{Zs}`
BACKSLASH                  = '\\'
BOOLEAN                    = 'true' | 'false'
OCTDIG                     = `[0-8]`
OCTET                      = OCTDIG{2}
DIGIT                      = [0-9]
BASE64                     = [A-Za-z0-9+/]
BASE64URL                  = [A-Za-z0-9-_]
ALPHA                      = [A-Za-z]
UPPERCASE                  = [A-Z]
LOWERCASE                  = [a-z]
HEXDIG                     = [0-9A-F]
LOWER-HEXDIG               = [0-9a-f]
ALPHANUM                   = ALPHA | DIGIT
SQUOTE                     = "'"
DQUOTE                     = '"'
RFC-3339-DATETIME          = `20\d\d-(12|11|[1-9])-(\d|[12]\d|3[01])\
                              T([01]\d|2[0-3])(:([0-5]\d|60)){2}\
                              (\.{3}|\.{6})?Z`
E-NOTATION-FLOAT           = LITERAL-FLOAT ('E' LITERAL-FLOAT)?
E-NOTATION-NONNEG-FLOAT    = LITERAL-NONNEG-FLOAT ('E' LITERAL-FLOAT)?
E-NOTATION-POSITIVE-FLOAT  = LITERAL-POSITIVE-FLOAT ('E' LITERAL-FLOAT)?
E-NOTATION-NONZERO-FLOAT   = LITERAL-NONZERO-FLOAT ('E' LITERAL-FLOAT)?
E-NOTATION-INT             = LITERAL-FLOAT ('E' LITERAL-FLOAT)?
E-NOTATION-NONNEG-INT      = LITERAL-NONNEG-INT ('E' LITERAL-NONNEG-INT)?
E-NOTATION-POSITIVE-INT    = LITERAL-POSITIVE-INT ('E' LITERAL-NONNEG-INT)?
E-NOTATION-NONZERO-INT     = LITERAL-NONZERO-INT ('E' LITERAL-NONNEG-INT)?
LITERAL-FLOAT              = '-'? DIGIT-STR ('.' DIGIT-STR)?
LITERAL-NONNEG-FLOAT       = DIGIT-STR ('.' DIGIT-STR)?
LITERAL-POSITIVE-FLOAT     = [1-9] DIGIT-STR ('.' DIGIT-STR)?
LITERAL-NONZERO-FLOAT      = '-'? [1-9] DIGIT-STR ('.' DIGIT-STR)?
LITERAL-INT                = '-'? DIGIT-STR
LITERAL-NONNEG-INT         = DIGIT-STR
LITERAL-POSITIVE-INT       = [1-9] DIGIT-STR
LITERAL-NONZERO-INT        = '-'? [1-9] DIGIT-STR
BIN-STR                    = BIN+
OCT-STR                    = OCTDIG+
HEX-STR                    = HEXDIG+
ALPHA-STR                  = ALPHA+
DIGIT-STR                  = DIGIT+
ALPHANUM-STR               = (ALPHA | DIGIT)+
BASE64-STR                 = BASE64+ '='{0,8}
BASE64URL-STR              = BASE64URL+ '='{0,8}
TICK                       = '`'
BIT                        = [01]
CRLF                       = CR LF
CR                         = '\r'
LF                         = '\n'
HTAB                       = '\t'
SP                         = ' '

literal-1     = ' "ab" '
                ; can also use ::= as in XML-MG
literal-2     = " 'ab' "
concatenation = literal-1 'defg'
alternation   = literal-1 | 'xy'
                ; slash is an alterative to |
intersection  = alternation & .{10}
                ; intersection of multiple rules!!
                ; 'intersection' must be exactly ' "ab" xyxyxy'
dot-regex     = .+
                ; regex starting with '.' need not be enclosed in ``
simple-regex  = [^A-Z]{2,4}
                ; regex starting with '[' need not be enclosed in ``
complex-regex = `.+? *\d`
                ; any regex can be enclosed in ``
grouping      = ('ab' | 'cd') 'xy'
complement    = (! 'abc')
                ; complement!!
                ; 'complement' is any text (0+ chars) except 'abc'
set-minus     = .+ - 'abc'
                ; exclusion!
                ; this is identical to 'complement' (above)
unicode-1     = #5F028322
unicode-2     = #'Plus-Minus Sign'
inline-label  = label-1 ([^ ]+)=my-label
                ; declare an inline rule, which can be used anywhere