|
|
|
@ -24,10 +24,9 @@ module Twitter::TwitterText
|
|
|
|
|
)
|
|
|
|
|
\)
|
|
|
|
|
/iox
|
|
|
|
|
REGEXEN[:valid_iri_ucschar] = /[\u{A0}-\u{D7FF}\u{F900}-\u{FDCF}\u{FDF0}-\u{FFEF}\u{10000}-\u{1FFFD}\u{20000}-\u{2FFFD}\u{30000}-\u{3FFFD}\u{40000}-\u{4FFFD}\u{50000}-\u{5FFFD}\u{60000}-\u{6FFFD}\u{70000}-\u{7FFFD}\u{80000}-\u{8FFFD}\u{90000}-\u{9FFFD}\u{A0000}-\u{AFFFD}\u{B0000}-\u{BFFFD}\u{C0000}-\u{CFFFD}\u{D0000}-\u{DFFFD}\u{E1000}-\u{EFFFD}]/iou
|
|
|
|
|
REGEXEN[:valid_iri_iprivate] = /[\u{E000}-\u{F8FF}\u{F0000}-\u{FFFFD}\u{100000}-\u{10FFFD}]/iou
|
|
|
|
|
REGEXEN[:valid_url_query_chars] = /(?:#{REGEXEN[:valid_iri_ucschar]})|(?:#{REGEXEN[:valid_iri_iprivate]})|[a-z0-9!?\*'\(\);:&=\+\$\/%#\[\]\-_\.,~|@]/iou
|
|
|
|
|
REGEXEN[:valid_url_query_ending_chars] = /(?:#{REGEXEN[:valid_iri_ucschar]})|(?:#{REGEXEN[:valid_iri_iprivate]})|[a-z0-9_&=#\/\-]/iou
|
|
|
|
|
UCHARS = '\u{A0}-\u{D7FF}\u{F900}-\u{FDCF}\u{FDF0}-\u{FFEF}\u{10000}-\u{1FFFD}\u{20000}-\u{2FFFD}\u{30000}-\u{3FFFD}\u{40000}-\u{4FFFD}\u{50000}-\u{5FFFD}\u{60000}-\u{6FFFD}\u{70000}-\u{7FFFD}\u{80000}-\u{8FFFD}\u{90000}-\u{9FFFD}\u{A0000}-\u{AFFFD}\u{B0000}-\u{BFFFD}\u{C0000}-\u{CFFFD}\u{D0000}-\u{DFFFD}\u{E1000}-\u{EFFFD}\u{E000}-\u{F8FF}\u{F0000}-\u{FFFFD}\u{100000}-\u{10FFFD}'
|
|
|
|
|
REGEXEN[:valid_url_query_chars] = /[a-z0-9!?\*'\(\);:&=\+\$\/%#\[\]\-_\.,~|@#{UCHARS}]/iou
|
|
|
|
|
REGEXEN[:valid_url_query_ending_chars] = /[a-z0-9_&=#\/\-#{UCHARS}]/iou
|
|
|
|
|
REGEXEN[:valid_url_path] = /(?:
|
|
|
|
|
(?:
|
|
|
|
|
#{REGEXEN[:valid_general_url_path_chars]}*
|
|
|
|
@ -57,23 +56,21 @@ module Twitter::TwitterText
|
|
|
|
|
#{REGEXEN[:validate_url_pct_encoded]}|
|
|
|
|
|
#{REGEXEN[:validate_url_sub_delims]}
|
|
|
|
|
)/iox
|
|
|
|
|
REGEXEN[:xmpp_uri] = %r{
|
|
|
|
|
(xmpp:) # Protocol
|
|
|
|
|
(//#{REGEXEN[:validate_nodeid]}+@#{REGEXEN[:valid_domain]}/)? # Authority (optional)
|
|
|
|
|
(#{REGEXEN[:validate_nodeid]}+@)? # Username in path (optional)
|
|
|
|
|
(#{REGEXEN[:valid_domain]}) # Domain in path
|
|
|
|
|
(/#{REGEXEN[:validate_resid]}+)? # Resource in path (optional)
|
|
|
|
|
(\?#{REGEXEN[:valid_url_query_chars]}*#{REGEXEN[:valid_url_query_ending_chars]})? # Query String
|
|
|
|
|
}iox
|
|
|
|
|
REGEXEN[:magnet_uri] = %r{
|
|
|
|
|
(magnet:) # Protocol
|
|
|
|
|
(\?#{REGEXEN[:valid_url_query_chars]}*#{REGEXEN[:valid_url_query_ending_chars]}) # Query String
|
|
|
|
|
}iox
|
|
|
|
|
REGEXEN[:valid_extended_uri] = %r{
|
|
|
|
|
( # $1 total match
|
|
|
|
|
(#{REGEXEN[:valid_url_preceding_chars]}) # $2 Preceding character
|
|
|
|
|
( # $3 URL
|
|
|
|
|
(#{REGEXEN[:xmpp_uri]}) | (#{REGEXEN[:magnet_uri]})
|
|
|
|
|
(
|
|
|
|
|
(xmpp:) # Protocol
|
|
|
|
|
(//#{REGEXEN[:validate_nodeid]}+@#{REGEXEN[:valid_domain]}/)? # Authority (optional)
|
|
|
|
|
(#{REGEXEN[:validate_nodeid]}+@)? # Username in path (optional)
|
|
|
|
|
(#{REGEXEN[:valid_domain]}) # Domain in path
|
|
|
|
|
(/#{REGEXEN[:validate_resid]}+)? # Resource in path (optional)
|
|
|
|
|
(\?#{REGEXEN[:valid_url_query_chars]}*#{REGEXEN[:valid_url_query_ending_chars]})? # Query String
|
|
|
|
|
) | (
|
|
|
|
|
(magnet:) # Protocol
|
|
|
|
|
(\?#{REGEXEN[:valid_url_query_chars]}*#{REGEXEN[:valid_url_query_ending_chars]}) # Query String
|
|
|
|
|
)
|
|
|
|
|
)
|
|
|
|
|
)
|
|
|
|
|
}iox
|
|
|
|
|