##! Please refer to the documentation at ##! https://coreruleset.org/docs/development/regexp_assemble/. ##! The expression generated from this file matches a full HTTP ##! `Accept` header by following the specification as far as ##! necessary, while restricting the `charset` parameter ##! to a list of explicitly allowed values. ##! Where possible, the expression matches tokens "loosely", ##! to reduce complexity and the risk of false positives. ##! Specifications: ##! https://httpwg.org/specs/rfc7231.html#request.conneg ##! https://httpwg.org/specs/rfc7230.html ##! This is a duplicate of the 920600.data file. ##! There will be support for inclusion in the regexp_assemble tool in the future. ##! In the meantime, please keep both files in sync. ##!^ ^ ##!$ $ ##! Helpers ##!> template non-token-with-dquote-chars (),/:;<=>?![\x5c\]{} ##!> template non-token-chars \"{{non-token-with-dquote-chars}} ##!> template token-chars [^{{non-token-chars}}] ##!> template token-with-dquote-chars [^{{non-token-with-dquote-chars}}] ##!> template type-subtype (?:\*|{{token-chars}}+) ##!> template media-type {{type-subtype}}/{{type-subtype}} ##! list of allowed charsets ##!> assemble (?:"? ##!=> ##!> include allowed-charsets ##!=> \b"?)) ##!=< allowed-charsets ##!< ##! Main assembly ##!> assemble (?:{{media-type}}) ##!=> (?:\s*+;\s*+ ##!=> (?:(?:charset\s*+=\s*+ ##!=> ##!=> allowed-charsets ##! If the first part wasn't a "charset", then ##! anything is allowed here that is not "charset". ##! Note that this doesn't follow the RFC strictly. |(?: ##!=> [^c{{non-token-chars}}]{{token-chars}}* c[^h{{non-token-chars}}]{{token-chars}}* ch[^a{{non-token-chars}}]{{token-chars}}* cha[^r{{non-token-chars}}]{{token-chars}}* char[^s{{non-token-chars}}]{{token-chars}}* chars[^e{{non-token-chars}}]{{token-chars}}* charse[^t{{non-token-chars}}]{{token-chars}}* ##!=> )\s*+=\s*+{{token-with-dquote-chars}}+) ##! Clients like to violate the RFC, be lenient with ##! terminating semi-colons. ;? ##!=> )* ##!=> ##! Multiple "media-range" expressions can be ##! specified, comma separated. (?:\s*+,\s*+ ##!=> (?:{{media-type}}) ##!=> (?:\s*+;\s*+ ##!=> (?:(?:charset\s*+=\s*+ ##!=> ##!=> allowed-charsets |(?: ##!=> [^c{{non-token-chars}}]{{token-chars}}* c[^h{{non-token-chars}}]{{token-chars}}* ch[^a{{non-token-chars}}]{{token-chars}}* cha[^r{{non-token-chars}}]{{token-chars}}* char[^s{{non-token-chars}}]{{token-chars}}* chars[^e{{non-token-chars}}]{{token-chars}}* charse[^t{{non-token-chars}}]{{token-chars}}* ##!=> )\s*+=\s*+{{token-with-dquote-chars}}+) ##! Clients like to violate the RFC, be lenient with ##! terminating semi-colons. ;? ##!=> )*)* ##!=> ##!<