# frozen_string_literal: false
# Author:: Akira Yamada <akira@ruby-lang.org>
# You can redistribute it and/or modify it under the same term as Ruby.
# See URI for general documentation
# Includes URI::REGEXP::PATTERN
# Patterns used to parse URI's
# RFC 2396 (URI Generic Syntax)
# RFC 2732 (IPv6 Literal Addresses in URL's)
# RFC 2373 (IPv6 Addressing Architecture)
# alpha = lowalpha | upalpha
# alphanum = alpha | digit
# hex = digit | "A" | "B" | "C" | "D" | "E" | "F" |
# "a" | "b" | "c" | "d" | "e" | "f"
# mark = "-" | "_" | "." | "!" | "~" | "*" | "'" |
# unreserved = alphanum | mark
UNRESERVED = "\\-_.!~*'()#{ALNUM}"
# reserved = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" |
# reserved = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" |
# "$" | "," | "[" | "]" (RFC 2732)
RESERVED = ";/?:@&=+$,\\[\\]"
# domainlabel = alphanum | alphanum *( alphanum | "-" ) alphanum
DOMLABEL = "(?:[#{ALNUM}](?:[-#{ALNUM}]*[#{ALNUM}])?)"
# toplabel = alpha | alpha *( alphanum | "-" ) alphanum
TOPLABEL = "(?:[#{ALPHA}](?:[-#{ALNUM}]*[#{ALNUM}])?)"
# hostname = *( domainlabel "." ) toplabel [ "." ]
HOSTNAME = "(?:#{DOMLABEL}\\.)*#{TOPLABEL}\\.?"
# Class that parses String's into URI's.
# It contains a Hash set of patterns and Regexp's that match and validate.
# URI::Parser.new([opts])
# The constructor accepts a hash as options for parser.
# Keys of options are pattern names of URI components
# and values of options are pattern strings.
# The constructor generates set of regexps for parsing URIs.
# You can use the following keys:
# * :ESCAPED (URI::PATTERN::ESCAPED in default)
# * :UNRESERVED (URI::PATTERN::UNRESERVED in default)
# * :DOMLABEL (URI::PATTERN::DOMLABEL in default)
# * :TOPLABEL (URI::PATTERN::TOPLABEL in default)
# * :HOSTNAME (URI::PATTERN::HOSTNAME in default)
# p = URI::Parser.new(:ESCAPED => "(?:%[a-fA-F0-9]{2}|%u[a-fA-F0-9]{4})")
# u = p.parse("http://example.jp/%uABCD") #=> #<URI::HTTP http://example.jp/%uABCD>
# URI.parse(u.to_s) #=> raises URI::InvalidURIError
# s = "http://example.com/ABCD"
# u1 = p.parse(s) #=> #<URI::HTTP http://example.com/ABCD>
# u2 = URI.parse(s) #=> #<URI::HTTP http://example.com/ABCD>
def initialize(opts = {})
@pattern = initialize_pattern(opts)
@pattern.each_value(&:freeze)
@regexp = initialize_regexp(@pattern)
@regexp.each_value(&:freeze)
# See also URI::Parser.initialize_pattern.
# See also URI::Parser.initialize_regexp.
# Returns a split URI against regexp[:ABS_URI].
scheme, opaque, userinfo, host, port,
registry, path, query, fragment = $~[1..-1]
# URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
# absoluteURI = scheme ":" ( hier_part | opaque_part )
# hier_part = ( net_path | abs_path ) [ "?" query ]
# opaque_part = uric_no_slash *uric
# abs_path = "/" path_segments
# net_path = "//" authority [ abs_path ]
# authority = server | reg_name
# server = [ [ userinfo "@" ] hostport ]
"bad URI(absolute but no scheme): #{uri}"
if !opaque && (!path && (!host && !registry))
"bad URI(absolute but no path): #{uri}"
userinfo, host, port, registry,
rel_segment, abs_path, query, fragment = $~[1..-1]
if rel_segment && abs_path
path = rel_segment + abs_path
# URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
# relativeURI = ( net_path | abs_path | rel_path ) [ "?" query ]
# net_path = "//" authority [ abs_path ]
# abs_path = "/" path_segments
# rel_path = rel_segment [ abs_path ]
# authority = server | reg_name
# server = [ [ userinfo "@" ] hostport ]
raise InvalidURIError, "bad URI(is not URI?): #{uri}"
path = '' if !path && !opaque # (see RFC2396 Section 5.2)
userinfo, host, port, # X
# Parses +uri+ and constructs either matching URI scheme object
# (File, FTP, HTTP, HTTPS, LDAP, LDAPS, or MailTo) or URI::Generic.
# p.parse("ldap://ldap.example.com/dc=example?user=john")
# #=> #<URI::LDAP ldap://ldap.example.com/dc=example?user=john>
scheme, userinfo, host, port,
registry, path, opaque, query, fragment = self.split(uri)
if scheme && URI.scheme_list.include?(scheme.upcase)
URI.scheme_list[scheme.upcase].new(scheme, userinfo, host, port,
registry, path, opaque, query,
Generic.new(scheme, userinfo, host, port,
registry, path, opaque, query,
# Attempts to parse and merge a set of URIs.
uris[0] = convert_to_uri(uris[0])
# extract( str, schemes )
# extract( str, schemes ) {|item| block }
# Patterns to apply to +str+
# Attempts to parse and merge a set of URIs.
# If no +block+ given, then returns the result,
# else it calls +block+ for each element in result.
# See also URI::Parser.make_regexp.
def extract(str, schemes = nil)
str.scan(make_regexp(schemes)) { yield $& }
str.scan(make_regexp(schemes)) { result.push $& }
# Returns Regexp that is default self.regexp[:ABS_URI_REF],
# unless +schemes+ is provided. Then it is a Regexp.union with self.pattern[:X_ABS_URI].
def make_regexp(schemes = nil)
/(?=#{Regexp.union(*schemes)}:)#{@pattern[:X_ABS_URI]}/x
# Regexp to apply. Defaults to self.regexp[:UNSAFE]
# Constructs a safe String from +str+, removing unsafe characters,
# replacing them with codes.
def escape(str, unsafe = @regexp[:UNSAFE])
unless unsafe.kind_of?(Regexp)
# perhaps unsafe is String object
unsafe = Regexp.new("[#{Regexp.quote(unsafe)}]", false)
tmp << sprintf('%%%02X', uc)
end.force_encoding(Encoding::US_ASCII)
# unescape( str, escaped )
# String to remove escapes from
# Regexp to apply. Defaults to self.regexp[:ESCAPED]
# Removes escapes from +str+.
def unescape(str, escaped = @regexp[:ESCAPED])
enc = Encoding::UTF_8 if enc == Encoding::US_ASCII
str.gsub(escaped) { [$&[1, 2]].pack('H2').force_encoding(enc) }
@@to_s = Kernel.instance_method(:to_s)
# Constructs the default Hash of patterns.
def initialize_pattern(opts = {})
ret[:ESCAPED] = escaped = (opts.delete(:ESCAPED) || PATTERN::ESCAPED)
ret[:UNRESERVED] = unreserved = opts.delete(:UNRESERVED) || PATTERN::UNRESERVED
ret[:RESERVED] = reserved = opts.delete(:RESERVED) || PATTERN::RESERVED
ret[:DOMLABEL] = opts.delete(:DOMLABEL) || PATTERN::DOMLABEL
ret[:TOPLABEL] = opts.delete(:TOPLABEL) || PATTERN::TOPLABEL
ret[:HOSTNAME] = hostname = opts.delete(:HOSTNAME)
# RFC 2396 (URI Generic Syntax)
# RFC 2732 (IPv6 Literal Addresses in URL's)
# RFC 2373 (IPv6 Addressing Architecture)
# uric = reserved | unreserved | escaped
ret[:URIC] = uric = "(?:[#{unreserved}#{reserved}]|#{escaped})"
# uric_no_slash = unreserved | escaped | ";" | "?" | ":" | "@" |
# "&" | "=" | "+" | "$" | ","
ret[:URIC_NO_SLASH] = uric_no_slash = "(?:[#{unreserved};?:@&=+$,]|#{escaped})"
ret[:QUERY] = query = "#{uric}*"
ret[:FRAGMENT] = fragment = "#{uric}*"
# hostname = *( domainlabel "." ) toplabel [ "." ]
# reg-name = *( unreserved / pct-encoded / sub-delims ) # RFC3986
ret[:HOSTNAME] = hostname = "(?:[a-zA-Z0-9\\-.]|%\\h\\h)+"
# IPv6address = hexpart [ ":" IPv4address ]
# IPv4address = 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT
# hexpart = hexseq | hexseq "::" [ hexseq ] | "::" [ hexseq ]
# hexseq = hex4 *( ":" hex4)
# XXX: This definition has a flaw. "::" + IPv4address must be
# allowed too. Here is a replacement.
# IPv4address = 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT
ret[:IPV4ADDR] = ipv4addr = "\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}"
hex4 = "[#{PATTERN::HEX}]{1,4}"
# lastpart = hex4 | IPv4address
lastpart = "(?:#{hex4}|#{ipv4addr})"
# hexseq1 = *( hex4 ":" ) hex4
hexseq1 = "(?:#{hex4}:)*#{hex4}"
# hexseq2 = *( hex4 ":" ) lastpart
hexseq2 = "(?:#{hex4}:)*#{lastpart}"
# IPv6address = hexseq2 | [ hexseq1 ] "::" [ hexseq2 ]
ret[:IPV6ADDR] = ipv6addr = "(?:#{hexseq2}|(?:#{hexseq1})?::(?:#{hexseq2})?)"
# IPv6prefix = ( hexseq1 | [ hexseq1 ] "::" [ hexseq1 ] ) "/" 1*2DIGIT
# ipv6reference = "[" IPv6address "]" (RFC 2732)
ret[:IPV6REF] = ipv6ref = "\\[#{ipv6addr}\\]"
# host = hostname | IPv4address
# host = hostname | IPv4address | IPv6reference (RFC 2732)
ret[:HOST] = host = "(?:#{hostname}|#{ipv4addr}|#{ipv6ref})"
ret[:PORT] = port = '\d*'
# hostport = host [ ":" port ]
ret[:HOSTPORT] = hostport = "#{host}(?::#{port})?"
# userinfo = *( unreserved | escaped |
# ";" | ":" | "&" | "=" | "+" | "$" | "," )
ret[:USERINFO] = userinfo = "(?:[#{unreserved};:&=+$,]|#{escaped})*"
# pchar = unreserved | escaped |
# ":" | "@" | "&" | "=" | "+" | "$" | ","
pchar = "(?:[#{unreserved}:@&=+$,]|#{escaped})"
# segment = *pchar *( ";" param )
segment = "#{pchar}*(?:;#{param})*"
# path_segments = segment *( "/" segment )
ret[:PATH_SEGMENTS] = path_segments = "#{segment}(?:/#{segment})*"
# server = [ [ userinfo "@" ] hostport ]
server = "(?:#{userinfo}@)?#{hostport}"
# reg_name = 1*( unreserved | escaped | "$" | "," |
# ";" | ":" | "@" | "&" | "=" | "+" )
ret[:REG_NAME] = reg_name = "(?:[#{unreserved}$,;:@&=+]|#{escaped})+"
# authority = server | reg_name
authority = "(?:#{server}|#{reg_name})"
# rel_segment = 1*( unreserved | escaped |
# ";" | "@" | "&" | "=" | "+" | "$" | "," )
ret[:REL_SEGMENT] = rel_segment = "(?:[#{unreserved};@&=+$,]|#{escaped})+"
# scheme = alpha *( alpha | digit | "+" | "-" | "." )
ret[:SCHEME] = scheme = "[#{PATTERN::ALPHA}][\\-+.#{PATTERN::ALPHA}\\d]*"
# abs_path = "/" path_segments
ret[:ABS_PATH] = abs_path = "/#{path_segments}"
# rel_path = rel_segment [ abs_path ]
ret[:REL_PATH] = rel_path = "#{rel_segment}(?:#{abs_path})?"
# net_path = "//" authority [ abs_path ]
ret[:NET_PATH] = net_path = "//#{authority}(?:#{abs_path})?"
# hier_part = ( net_path | abs_path ) [ "?" query ]
ret[:HIER_PART] = hier_part = "(?:#{net_path}|#{abs_path})(?:\\?(?:#{query}))?"
# opaque_part = uric_no_slash *uric
ret[:OPAQUE_PART] = opaque_part = "#{uric_no_slash}#{uric}*"
# absoluteURI = scheme ":" ( hier_part | opaque_part )
ret[:ABS_URI] = abs_uri = "#{scheme}:(?:#{hier_part}|#{opaque_part})"
# relativeURI = ( net_path | abs_path | rel_path ) [ "?" query ]
ret[:REL_URI] = rel_uri = "(?:#{net_path}|#{abs_path}|#{rel_path})(?:\\?#{query})?"
# URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
ret[:URI_REF] = "(?:#{abs_uri}|#{rel_uri})?(?:##{fragment})?"
(#{scheme}): (?# 1: scheme)
(#{opaque_part}) (?# 2: opaque)
(?:(?:(#{userinfo})@)? (?# 3: userinfo)
(?:(#{host})(?::(\\d*))?))? (?# 4: host, 5: port)
(#{reg_name}) (?# 6: registry)
(?!//)) (?# XXX: '//' is the mark for hostport)
(#{abs_path})? (?# 7: path)
)(?:\\?(#{query}))? (?# 8: query)
(?:\\#(#{fragment}))? (?# 9: fragment)
(?:(#{userinfo})@)? (?# 1: userinfo)
(#{host})?(?::(\\d*))? (?# 2: host, 3: port)
(#{reg_name}) (?# 4: registry)
(#{rel_segment}) (?# 5: rel_segment)
(#{abs_path})? (?# 6: abs_path)
(?:\\?(#{query}))? (?# 7: query)
(?:\\#(#{fragment}))? (?# 8: fragment)
It is recommended that you Edit text format, this type of Fix handles quite a lot in one request
