Core rules

Every Rule subclass is initialized with the RFC 5234 core rules, so they are available by name in any grammar without being redefined:

from abnf import Rule

Rule("ALPHA")   # already defined in every Rule subclass

Rule	ABNF definition	Matches
ALPHA	%x41-5A / %x61-7A	A–Z, a–z
BIT	"0" / "1"	a binary digit
CHAR	%x01-7F	any 7-bit US-ASCII character except NUL
CR	%x0D	carriage return
CRLF	CR LF	Internet standard newline
CTL	%x00-1F / %x7F	control characters
DIGIT	%x30-39	0–9
DQUOTE	%x22	the double-quote "
HEXDIG	DIGIT / "A" / "B" / "C" / "D" / "E" / "F"	a hexadecimal digit
HTAB	%x09	horizontal tab
LF	%x0A	line feed
LWSP	*(WSP / CRLF WSP)	linear white space (see the RFC 5234 caveat)
OCTET	%x00-FF	any 8-bit byte
SP	%x20	a space
VCHAR	%x21-7E	any visible (printing) character
WSP	SP / HTAB	a space or horizontal tab

Note

HEXDIG matches the hexadecimal letters case-insensitively, per the ABNF convention that literal strings are case-insensitive unless declared otherwise (see Architecture: parser combinators and RFC 7405).