chuckya/config/initializers/twitter_regex.rb
Eugen Rochko 0df91c7b1e
Add dat, dweb, ipfs, ipns, ssb, gopher protocols to URL extractor (#7810)
* Add dat:// and gopher:// to URL extractor

Fix #6072

* Fix comment indent

* Add dweb, ipfs, ipns, ssb
2018-06-15 20:21:47 +02:00

42 lines
1.9 KiB
Ruby

module Twitter
class Regex
REGEXEN[:valid_general_url_path_chars] = /[^\p{White_Space}\(\)\?]/iou
REGEXEN[:valid_url_path_ending_chars] = /[^\p{White_Space}\(\)\?!\*';:=\,\.\$%\[\]~&\|@]|(?:#{REGEXEN[:valid_url_balanced_parens]})/iou
REGEXEN[:valid_url_balanced_parens] = /
\(
(?:
#{REGEXEN[:valid_general_url_path_chars]}+
|
# allow one nested level of balanced parentheses
(?:
#{REGEXEN[:valid_general_url_path_chars]}*
\(
#{REGEXEN[:valid_general_url_path_chars]}+
\)
#{REGEXEN[:valid_general_url_path_chars]}*
)
)
\)
/iox
REGEXEN[:valid_url_path] = /(?:
(?:
#{REGEXEN[:valid_general_url_path_chars]}*
(?:#{REGEXEN[:valid_url_balanced_parens]} #{REGEXEN[:valid_general_url_path_chars]}*)*
#{REGEXEN[:valid_url_path_ending_chars]}
)|(?:#{REGEXEN[:valid_general_url_path_chars]}+\/)
)/iox
REGEXEN[:valid_url] = %r{
( # $1 total match
(#{REGEXEN[:valid_url_preceding_chars]}) # $2 Preceeding chracter
( # $3 URL
((https?|dat|dweb|ipfs|ipns|ssb|gopher):\/\/)? # $4 Protocol (optional)
(#{REGEXEN[:valid_domain]}) # $5 Domain(s)
(?::(#{REGEXEN[:valid_port_number]}))? # $6 Port number (optional)
(/#{REGEXEN[:valid_url_path]}*)? # $7 URL Path and anchor
(\?#{REGEXEN[:valid_url_query_chars]}*#{REGEXEN[:valid_url_query_ending_chars]})? # $8 Query String
)
)
}iox
end
end