addressable 2.4.0 → 2.8.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +5 -5
- data/CHANGELOG.md +54 -0
- data/Gemfile +13 -12
- data/README.md +31 -15
- data/Rakefile +5 -3
- data/addressable.gemspec +23 -20
- data/lib/addressable/idna/native.rb +11 -5
- data/lib/addressable/idna/pure.rb +61 -55
- data/lib/addressable/idna.rb +3 -1
- data/lib/addressable/template.rb +64 -84
- data/lib/addressable/uri.rb +228 -96
- data/lib/addressable/version.rb +4 -2
- data/lib/addressable.rb +2 -0
- data/spec/addressable/idna_spec.rb +35 -3
- data/spec/addressable/net_http_compat_spec.rb +3 -1
- data/spec/addressable/security_spec.rb +3 -1
- data/spec/addressable/template_spec.rb +77 -3
- data/spec/addressable/uri_spec.rb +663 -203
- data/spec/spec_helper.rb +12 -0
- data/tasks/clobber.rake +2 -0
- data/tasks/gem.rake +9 -14
- data/tasks/git.rake +2 -0
- data/tasks/metrics.rake +2 -0
- data/tasks/profile.rake +72 -0
- data/tasks/rspec.rake +3 -1
- data/tasks/yard.rake +2 -0
- metadata +36 -11
- data/spec/addressable/rack_mount_compat_spec.rb +0 -104
data/lib/addressable/uri.rb
CHANGED
@@ -1,6 +1,8 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
# encoding:utf-8
|
2
4
|
#--
|
3
|
-
# Copyright (C)
|
5
|
+
# Copyright (C) Bob Aman
|
4
6
|
#
|
5
7
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
6
8
|
# you may not use this file except in compliance with the License.
|
@@ -18,6 +20,7 @@
|
|
18
20
|
|
19
21
|
require "addressable/version"
|
20
22
|
require "addressable/idna"
|
23
|
+
require "public_suffix"
|
21
24
|
|
22
25
|
##
|
23
26
|
# Addressable is a library for processing links and URIs.
|
@@ -44,13 +47,22 @@ module Addressable
|
|
44
47
|
UNRESERVED = ALPHA + DIGIT + "\\-\\.\\_\\~"
|
45
48
|
PCHAR = UNRESERVED + SUB_DELIMS + "\\:\\@"
|
46
49
|
SCHEME = ALPHA + DIGIT + "\\-\\+\\."
|
47
|
-
HOST =
|
48
|
-
AUTHORITY = PCHAR
|
50
|
+
HOST = UNRESERVED + SUB_DELIMS + "\\[\\:\\]"
|
51
|
+
AUTHORITY = PCHAR + "\\[\\:\\]"
|
49
52
|
PATH = PCHAR + "\\/"
|
50
53
|
QUERY = PCHAR + "\\/\\?"
|
51
54
|
FRAGMENT = PCHAR + "\\/\\?"
|
52
55
|
end
|
53
56
|
|
57
|
+
module NormalizeCharacterClasses
|
58
|
+
HOST = /[^#{CharacterClasses::HOST}]/
|
59
|
+
UNRESERVED = /[^#{CharacterClasses::UNRESERVED}]/
|
60
|
+
PCHAR = /[^#{CharacterClasses::PCHAR}]/
|
61
|
+
SCHEME = /[^#{CharacterClasses::SCHEME}]/
|
62
|
+
FRAGMENT = /[^#{CharacterClasses::FRAGMENT}]/
|
63
|
+
QUERY = %r{[^a-zA-Z0-9\-\.\_\~\!\$\'\(\)\*\+\,\=\:\@\/\?%]|%(?!2B|2b)}
|
64
|
+
end
|
65
|
+
|
54
66
|
SLASH = '/'
|
55
67
|
EMPTY_STR = ''
|
56
68
|
|
@@ -70,7 +82,7 @@ module Addressable
|
|
70
82
|
"wais" => 210,
|
71
83
|
"ldap" => 389,
|
72
84
|
"prospero" => 1525
|
73
|
-
}
|
85
|
+
}.freeze
|
74
86
|
|
75
87
|
##
|
76
88
|
# Returns a URI object based on the parsed string.
|
@@ -121,9 +133,9 @@ module Addressable
|
|
121
133
|
user = userinfo.strip[/^([^:]*):?/, 1]
|
122
134
|
password = userinfo.strip[/:(.*)$/, 1]
|
123
135
|
end
|
124
|
-
host = authority.
|
136
|
+
host = authority.sub(
|
125
137
|
/^([^\[\]]*)@/, EMPTY_STR
|
126
|
-
).
|
138
|
+
).sub(
|
127
139
|
/:([^:@\[\]]*?)$/, EMPTY_STR
|
128
140
|
)
|
129
141
|
port = authority[/:([^:@\[\]]*?)$/, 1]
|
@@ -176,33 +188,50 @@ module Addressable
|
|
176
188
|
raise TypeError, "Can't convert #{uri.class} into String."
|
177
189
|
end
|
178
190
|
# Otherwise, convert to a String
|
179
|
-
uri = uri.to_str.dup
|
191
|
+
uri = uri.to_str.dup.strip
|
180
192
|
hints = {
|
181
193
|
:scheme => "http"
|
182
194
|
}.merge(hints)
|
183
195
|
case uri
|
184
|
-
when /^http
|
185
|
-
uri.
|
186
|
-
when /^https
|
187
|
-
uri.
|
188
|
-
when /^feed:\/+http
|
189
|
-
uri.
|
190
|
-
when /^feed
|
191
|
-
uri.
|
192
|
-
when
|
193
|
-
uri.
|
196
|
+
when /^http:\//i
|
197
|
+
uri.sub!(/^http:\/+/i, "http://")
|
198
|
+
when /^https:\//i
|
199
|
+
uri.sub!(/^https:\/+/i, "https://")
|
200
|
+
when /^feed:\/+http:\//i
|
201
|
+
uri.sub!(/^feed:\/+http:\/+/i, "feed:http://")
|
202
|
+
when /^feed:\//i
|
203
|
+
uri.sub!(/^feed:\/+/i, "feed://")
|
204
|
+
when %r[^file:/{4}]i
|
205
|
+
uri.sub!(%r[^file:/+]i, "file:////")
|
206
|
+
when %r[^file://localhost/]i
|
207
|
+
uri.sub!(%r[^file://localhost/+]i, "file:///")
|
208
|
+
when %r[^file:/+]i
|
209
|
+
uri.sub!(%r[^file:/+]i, "file:///")
|
194
210
|
when /^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}/
|
195
|
-
uri.
|
211
|
+
uri.sub!(/^/, hints[:scheme] + "://")
|
212
|
+
when /\A\d+\..*:\d+\z/
|
213
|
+
uri = "#{hints[:scheme]}://#{uri}"
|
214
|
+
end
|
215
|
+
match = uri.match(URIREGEX)
|
216
|
+
fragments = match.captures
|
217
|
+
authority = fragments[3]
|
218
|
+
if authority && authority.length > 0
|
219
|
+
new_authority = authority.tr("\\", "/").gsub(" ", "%20")
|
220
|
+
# NOTE: We want offset 4, not 3!
|
221
|
+
offset = match.offset(4)
|
222
|
+
uri = uri.dup
|
223
|
+
uri[offset[0]...offset[1]] = new_authority
|
196
224
|
end
|
197
225
|
parsed = self.parse(uri)
|
198
226
|
if parsed.scheme =~ /^[^\/?#\.]+\.[^\/?#]+$/
|
199
227
|
parsed = self.parse(hints[:scheme] + "://" + uri)
|
200
228
|
end
|
201
229
|
if parsed.path.include?(".")
|
202
|
-
|
203
|
-
|
230
|
+
if parsed.path[/\b@\b/]
|
231
|
+
parsed.scheme = "mailto" unless parsed.scheme
|
232
|
+
elsif new_host = parsed.path[/^([^\/]+\.[^\/]*)/, 1]
|
204
233
|
parsed.defer_validation do
|
205
|
-
new_path = parsed.path.
|
234
|
+
new_path = parsed.path.sub(
|
206
235
|
Regexp.new("^" + Regexp.escape(new_host)), EMPTY_STR)
|
207
236
|
parsed.host = new_host
|
208
237
|
parsed.path = new_path
|
@@ -253,24 +282,24 @@ module Addressable
|
|
253
282
|
# Otherwise, convert to a String
|
254
283
|
path = path.to_str.strip
|
255
284
|
|
256
|
-
path.
|
285
|
+
path.sub!(/^file:\/?\/?/, EMPTY_STR) if path =~ /^file:\/?\/?/
|
257
286
|
path = SLASH + path if path =~ /^([a-zA-Z])[\|:]/
|
258
287
|
uri = self.parse(path)
|
259
288
|
|
260
289
|
if uri.scheme == nil
|
261
290
|
# Adjust windows-style uris
|
262
|
-
uri.path.
|
291
|
+
uri.path.sub!(/^\/?([a-zA-Z])[\|:][\\\/]/) do
|
263
292
|
"/#{$1.downcase}:/"
|
264
293
|
end
|
265
|
-
uri.path.
|
294
|
+
uri.path.tr!("\\", SLASH)
|
266
295
|
if File.exist?(uri.path) &&
|
267
296
|
File.stat(uri.path).directory?
|
268
|
-
uri.path.
|
297
|
+
uri.path.chomp!(SLASH)
|
269
298
|
uri.path = uri.path + '/'
|
270
299
|
end
|
271
300
|
|
272
301
|
# If the path is absolute, set the scheme and host.
|
273
|
-
if uri.path
|
302
|
+
if uri.path.start_with?(SLASH)
|
274
303
|
uri.scheme = "file"
|
275
304
|
uri.host = EMPTY_STR
|
276
305
|
end
|
@@ -307,6 +336,21 @@ module Addressable
|
|
307
336
|
return result
|
308
337
|
end
|
309
338
|
|
339
|
+
##
|
340
|
+
# Tables used to optimize encoding operations in `self.encode_component`
|
341
|
+
# and `self.normalize_component`
|
342
|
+
SEQUENCE_ENCODING_TABLE = Hash.new do |hash, sequence|
|
343
|
+
hash[sequence] = sequence.unpack("C*").map do |c|
|
344
|
+
format("%02x", c)
|
345
|
+
end.join
|
346
|
+
end
|
347
|
+
|
348
|
+
SEQUENCE_UPCASED_PERCENT_ENCODING_TABLE = Hash.new do |hash, sequence|
|
349
|
+
hash[sequence] = sequence.unpack("C*").map do |c|
|
350
|
+
format("%%%02X", c)
|
351
|
+
end.join
|
352
|
+
end
|
353
|
+
|
310
354
|
##
|
311
355
|
# Percent encodes a URI component.
|
312
356
|
#
|
@@ -367,26 +411,26 @@ module Addressable
|
|
367
411
|
if character_class.kind_of?(String)
|
368
412
|
character_class = /[^#{character_class}]/
|
369
413
|
end
|
370
|
-
|
371
|
-
|
372
|
-
|
373
|
-
|
374
|
-
component.force_encoding(Encoding::ASCII_8BIT)
|
375
|
-
end
|
414
|
+
# We can't perform regexps on invalid UTF sequences, but
|
415
|
+
# here we need to, so switch to ASCII.
|
416
|
+
component = component.dup
|
417
|
+
component.force_encoding(Encoding::ASCII_8BIT)
|
376
418
|
# Avoiding gsub! because there are edge cases with frozen strings
|
377
419
|
component = component.gsub(character_class) do |sequence|
|
378
|
-
|
420
|
+
SEQUENCE_UPCASED_PERCENT_ENCODING_TABLE[sequence]
|
379
421
|
end
|
380
422
|
if upcase_encoded.length > 0
|
381
|
-
|
382
|
-
char
|
383
|
-
end
|
423
|
+
upcase_encoded_chars = upcase_encoded.chars.map do |char|
|
424
|
+
SEQUENCE_ENCODING_TABLE[char]
|
425
|
+
end
|
426
|
+
component = component.gsub(/%(#{upcase_encoded_chars.join('|')})/,
|
427
|
+
&:upcase)
|
384
428
|
end
|
385
429
|
return component
|
386
430
|
end
|
387
431
|
|
388
432
|
class << self
|
389
|
-
alias_method :
|
433
|
+
alias_method :escape_component, :encode_component
|
390
434
|
end
|
391
435
|
|
392
436
|
##
|
@@ -427,14 +471,18 @@ module Addressable
|
|
427
471
|
end
|
428
472
|
uri = uri.dup
|
429
473
|
# Seriously, only use UTF-8. I'm really not kidding!
|
430
|
-
uri.force_encoding("utf-8")
|
431
|
-
|
474
|
+
uri.force_encoding("utf-8")
|
475
|
+
|
476
|
+
unless leave_encoded.empty?
|
477
|
+
leave_encoded = leave_encoded.dup.force_encoding("utf-8")
|
478
|
+
end
|
479
|
+
|
432
480
|
result = uri.gsub(/%[0-9a-f]{2}/iu) do |sequence|
|
433
481
|
c = sequence[1..3].to_i(16).chr
|
434
|
-
c.force_encoding("utf-8")
|
482
|
+
c.force_encoding("utf-8")
|
435
483
|
leave_encoded.include?(c) ? sequence : c
|
436
484
|
end
|
437
|
-
result.force_encoding("utf-8")
|
485
|
+
result.force_encoding("utf-8")
|
438
486
|
if return_type == String
|
439
487
|
return result
|
440
488
|
elsif return_type == ::Addressable::URI
|
@@ -514,19 +562,21 @@ module Addressable
|
|
514
562
|
character_class = "#{character_class}%" unless character_class.include?('%')
|
515
563
|
|
516
564
|
"|%(?!#{leave_encoded.chars.map do |char|
|
517
|
-
seq = char
|
565
|
+
seq = SEQUENCE_ENCODING_TABLE[char]
|
518
566
|
[seq.upcase, seq.downcase]
|
519
567
|
end.flatten.join('|')})"
|
520
568
|
end
|
521
569
|
|
522
|
-
character_class =
|
523
|
-
|
524
|
-
|
525
|
-
|
526
|
-
|
527
|
-
component = component.dup
|
528
|
-
component.force_encoding(Encoding::ASCII_8BIT)
|
570
|
+
character_class = if leave_re
|
571
|
+
/[^#{character_class}]#{leave_re}/
|
572
|
+
else
|
573
|
+
/[^#{character_class}]/
|
574
|
+
end
|
529
575
|
end
|
576
|
+
# We can't perform regexps on invalid UTF sequences, but
|
577
|
+
# here we need to, so switch to ASCII.
|
578
|
+
component = component.dup
|
579
|
+
component.force_encoding(Encoding::ASCII_8BIT)
|
530
580
|
unencoded = self.unencode_component(component, String, leave_encoded)
|
531
581
|
begin
|
532
582
|
encoded = self.encode_component(
|
@@ -537,9 +587,7 @@ module Addressable
|
|
537
587
|
rescue ArgumentError
|
538
588
|
encoded = self.encode_component(unencoded)
|
539
589
|
end
|
540
|
-
|
541
|
-
encoded.force_encoding(Encoding::UTF_8)
|
542
|
-
end
|
590
|
+
encoded.force_encoding(Encoding::UTF_8)
|
543
591
|
return encoded
|
544
592
|
end
|
545
593
|
|
@@ -843,14 +891,17 @@ module Addressable
|
|
843
891
|
return nil unless self.scheme
|
844
892
|
@normalized_scheme ||= begin
|
845
893
|
if self.scheme =~ /^\s*ssh\+svn\s*$/i
|
846
|
-
"svn+ssh"
|
894
|
+
"svn+ssh".dup
|
847
895
|
else
|
848
896
|
Addressable::URI.normalize_component(
|
849
897
|
self.scheme.strip.downcase,
|
850
|
-
Addressable::URI::
|
898
|
+
Addressable::URI::NormalizeCharacterClasses::SCHEME
|
851
899
|
)
|
852
900
|
end
|
853
901
|
end
|
902
|
+
# All normalized values should be UTF-8
|
903
|
+
@normalized_scheme.force_encoding(Encoding::UTF_8) if @normalized_scheme
|
904
|
+
@normalized_scheme
|
854
905
|
end
|
855
906
|
|
856
907
|
##
|
@@ -864,7 +915,7 @@ module Addressable
|
|
864
915
|
new_scheme = new_scheme.to_str
|
865
916
|
end
|
866
917
|
if new_scheme && new_scheme !~ /\A[a-z][a-z0-9\.\+\-]*\z/i
|
867
|
-
raise InvalidURIError, "Invalid scheme format: #{new_scheme}"
|
918
|
+
raise InvalidURIError, "Invalid scheme format: '#{new_scheme}'"
|
868
919
|
end
|
869
920
|
@scheme = new_scheme
|
870
921
|
@scheme = nil if @scheme.to_s.strip.empty?
|
@@ -899,10 +950,13 @@ module Addressable
|
|
899
950
|
else
|
900
951
|
Addressable::URI.normalize_component(
|
901
952
|
self.user.strip,
|
902
|
-
Addressable::URI::
|
953
|
+
Addressable::URI::NormalizeCharacterClasses::UNRESERVED
|
903
954
|
)
|
904
955
|
end
|
905
956
|
end
|
957
|
+
# All normalized values should be UTF-8
|
958
|
+
@normalized_user.force_encoding(Encoding::UTF_8) if @normalized_user
|
959
|
+
@normalized_user
|
906
960
|
end
|
907
961
|
|
908
962
|
##
|
@@ -953,10 +1007,15 @@ module Addressable
|
|
953
1007
|
else
|
954
1008
|
Addressable::URI.normalize_component(
|
955
1009
|
self.password.strip,
|
956
|
-
Addressable::URI::
|
1010
|
+
Addressable::URI::NormalizeCharacterClasses::UNRESERVED
|
957
1011
|
)
|
958
1012
|
end
|
959
1013
|
end
|
1014
|
+
# All normalized values should be UTF-8
|
1015
|
+
if @normalized_password
|
1016
|
+
@normalized_password.force_encoding(Encoding::UTF_8)
|
1017
|
+
end
|
1018
|
+
@normalized_password
|
960
1019
|
end
|
961
1020
|
|
962
1021
|
##
|
@@ -1017,11 +1076,16 @@ module Addressable
|
|
1017
1076
|
if !current_user && !current_password
|
1018
1077
|
nil
|
1019
1078
|
elsif current_user && current_password
|
1020
|
-
"#{current_user}:#{current_password}"
|
1079
|
+
"#{current_user}:#{current_password}".dup
|
1021
1080
|
elsif current_user && !current_password
|
1022
|
-
"#{current_user}"
|
1081
|
+
"#{current_user}".dup
|
1023
1082
|
end
|
1024
1083
|
end
|
1084
|
+
# All normalized values should be UTF-8
|
1085
|
+
if @normalized_userinfo
|
1086
|
+
@normalized_userinfo.force_encoding(Encoding::UTF_8)
|
1087
|
+
end
|
1088
|
+
@normalized_userinfo
|
1025
1089
|
end
|
1026
1090
|
|
1027
1091
|
##
|
@@ -1067,6 +1131,7 @@ module Addressable
|
|
1067
1131
|
# @return [String] The host component, normalized.
|
1068
1132
|
def normalized_host
|
1069
1133
|
return nil unless self.host
|
1134
|
+
|
1070
1135
|
@normalized_host ||= begin
|
1071
1136
|
if !self.host.strip.empty?
|
1072
1137
|
result = ::Addressable::IDNA.to_ascii(
|
@@ -1078,12 +1143,18 @@ module Addressable
|
|
1078
1143
|
end
|
1079
1144
|
result = Addressable::URI.normalize_component(
|
1080
1145
|
result,
|
1081
|
-
|
1146
|
+
NormalizeCharacterClasses::HOST
|
1147
|
+
)
|
1082
1148
|
result
|
1083
1149
|
else
|
1084
|
-
EMPTY_STR
|
1150
|
+
EMPTY_STR.dup
|
1085
1151
|
end
|
1086
1152
|
end
|
1153
|
+
# All normalized values should be UTF-8
|
1154
|
+
if @normalized_host && !@normalized_host.empty?
|
1155
|
+
@normalized_host.force_encoding(Encoding::UTF_8)
|
1156
|
+
end
|
1157
|
+
@normalized_host
|
1087
1158
|
end
|
1088
1159
|
|
1089
1160
|
##
|
@@ -1096,14 +1167,6 @@ module Addressable
|
|
1096
1167
|
end
|
1097
1168
|
@host = new_host ? new_host.to_str : nil
|
1098
1169
|
|
1099
|
-
unreserved = CharacterClasses::UNRESERVED
|
1100
|
-
sub_delims = CharacterClasses::SUB_DELIMS
|
1101
|
-
if !@host.nil? && (@host =~ /[<>{}\/\?\#\@"[[:space:]]]/ ||
|
1102
|
-
(@host[/^\[(.*)\]$/, 1] != nil && @host[/^\[(.*)\]$/, 1] !~
|
1103
|
-
Regexp.new("^[#{unreserved}#{sub_delims}:]*$")))
|
1104
|
-
raise InvalidURIError, "Invalid character in host: '#{@host.to_s}'"
|
1105
|
-
end
|
1106
|
-
|
1107
1170
|
# Reset dependent values
|
1108
1171
|
remove_instance_variable(:@authority) if defined?(@authority)
|
1109
1172
|
remove_instance_variable(:@normalized_host) if defined?(@normalized_host)
|
@@ -1144,6 +1207,33 @@ module Addressable
|
|
1144
1207
|
self.host = v
|
1145
1208
|
end
|
1146
1209
|
|
1210
|
+
##
|
1211
|
+
# Returns the top-level domain for this host.
|
1212
|
+
#
|
1213
|
+
# @example
|
1214
|
+
# Addressable::URI.parse("http://www.example.co.uk").tld # => "co.uk"
|
1215
|
+
def tld
|
1216
|
+
PublicSuffix.parse(self.host, ignore_private: true).tld
|
1217
|
+
end
|
1218
|
+
|
1219
|
+
##
|
1220
|
+
# Sets the top-level domain for this URI.
|
1221
|
+
#
|
1222
|
+
# @param [String, #to_str] new_tld The new top-level domain.
|
1223
|
+
def tld=(new_tld)
|
1224
|
+
replaced_tld = host.sub(/#{tld}\z/, new_tld)
|
1225
|
+
self.host = PublicSuffix::Domain.new(replaced_tld).to_s
|
1226
|
+
end
|
1227
|
+
|
1228
|
+
##
|
1229
|
+
# Returns the public suffix domain for this host.
|
1230
|
+
#
|
1231
|
+
# @example
|
1232
|
+
# Addressable::URI.parse("http://www.example.co.uk").domain # => "example.co.uk"
|
1233
|
+
def domain
|
1234
|
+
PublicSuffix.domain(self.host, ignore_private: true)
|
1235
|
+
end
|
1236
|
+
|
1147
1237
|
##
|
1148
1238
|
# The authority component for this URI.
|
1149
1239
|
# Combines the user, password, host, and port components.
|
@@ -1151,7 +1241,7 @@ module Addressable
|
|
1151
1241
|
# @return [String] The authority component.
|
1152
1242
|
def authority
|
1153
1243
|
self.host && @authority ||= begin
|
1154
|
-
authority =
|
1244
|
+
authority = String.new
|
1155
1245
|
if self.userinfo != nil
|
1156
1246
|
authority << "#{self.userinfo}@"
|
1157
1247
|
end
|
@@ -1170,7 +1260,7 @@ module Addressable
|
|
1170
1260
|
def normalized_authority
|
1171
1261
|
return nil unless self.authority
|
1172
1262
|
@normalized_authority ||= begin
|
1173
|
-
authority =
|
1263
|
+
authority = String.new
|
1174
1264
|
if self.normalized_userinfo != nil
|
1175
1265
|
authority << "#{self.normalized_userinfo}@"
|
1176
1266
|
end
|
@@ -1180,6 +1270,11 @@ module Addressable
|
|
1180
1270
|
end
|
1181
1271
|
authority
|
1182
1272
|
end
|
1273
|
+
# All normalized values should be UTF-8
|
1274
|
+
if @normalized_authority
|
1275
|
+
@normalized_authority.force_encoding(Encoding::UTF_8)
|
1276
|
+
end
|
1277
|
+
@normalized_authority
|
1183
1278
|
end
|
1184
1279
|
|
1185
1280
|
##
|
@@ -1197,9 +1292,9 @@ module Addressable
|
|
1197
1292
|
new_user = new_userinfo.strip[/^([^:]*):?/, 1]
|
1198
1293
|
new_password = new_userinfo.strip[/:(.*)$/, 1]
|
1199
1294
|
end
|
1200
|
-
new_host = new_authority.
|
1295
|
+
new_host = new_authority.sub(
|
1201
1296
|
/^([^\[\]]*)@/, EMPTY_STR
|
1202
|
-
).
|
1297
|
+
).sub(
|
1203
1298
|
/:([^:@\[\]]*?)$/, EMPTY_STR
|
1204
1299
|
)
|
1205
1300
|
new_port =
|
@@ -1383,7 +1478,7 @@ module Addressable
|
|
1383
1478
|
# @return [String] The components that identify a site.
|
1384
1479
|
def site
|
1385
1480
|
(self.scheme || self.authority) && @site ||= begin
|
1386
|
-
site_string = ""
|
1481
|
+
site_string = "".dup
|
1387
1482
|
site_string << "#{self.scheme}:" if self.scheme != nil
|
1388
1483
|
site_string << "//#{self.authority}" if self.authority != nil
|
1389
1484
|
site_string
|
@@ -1402,7 +1497,7 @@ module Addressable
|
|
1402
1497
|
def normalized_site
|
1403
1498
|
return nil unless self.site
|
1404
1499
|
@normalized_site ||= begin
|
1405
|
-
site_string = ""
|
1500
|
+
site_string = "".dup
|
1406
1501
|
if self.normalized_scheme != nil
|
1407
1502
|
site_string << "#{self.normalized_scheme}:"
|
1408
1503
|
end
|
@@ -1411,6 +1506,9 @@ module Addressable
|
|
1411
1506
|
end
|
1412
1507
|
site_string
|
1413
1508
|
end
|
1509
|
+
# All normalized values should be UTF-8
|
1510
|
+
@normalized_site.force_encoding(Encoding::UTF_8) if @normalized_site
|
1511
|
+
@normalized_site
|
1414
1512
|
end
|
1415
1513
|
|
1416
1514
|
##
|
@@ -1460,17 +1558,20 @@ module Addressable
|
|
1460
1558
|
result = path.strip.split(SLASH, -1).map do |segment|
|
1461
1559
|
Addressable::URI.normalize_component(
|
1462
1560
|
segment,
|
1463
|
-
Addressable::URI::
|
1561
|
+
Addressable::URI::NormalizeCharacterClasses::PCHAR
|
1464
1562
|
)
|
1465
1563
|
end.join(SLASH)
|
1466
1564
|
|
1467
1565
|
result = URI.normalize_path(result)
|
1468
1566
|
if result.empty? &&
|
1469
1567
|
["http", "https", "ftp", "tftp"].include?(self.normalized_scheme)
|
1470
|
-
result = SLASH
|
1568
|
+
result = SLASH.dup
|
1471
1569
|
end
|
1472
1570
|
result
|
1473
1571
|
end
|
1572
|
+
# All normalized values should be UTF-8
|
1573
|
+
@normalized_path.force_encoding(Encoding::UTF_8) if @normalized_path
|
1574
|
+
@normalized_path
|
1474
1575
|
end
|
1475
1576
|
|
1476
1577
|
##
|
@@ -1489,6 +1590,9 @@ module Addressable
|
|
1489
1590
|
# Reset dependent values
|
1490
1591
|
remove_instance_variable(:@normalized_path) if defined?(@normalized_path)
|
1491
1592
|
remove_composite_values
|
1593
|
+
|
1594
|
+
# Ensure we haven't created an invalid URI
|
1595
|
+
validate()
|
1492
1596
|
end
|
1493
1597
|
|
1494
1598
|
##
|
@@ -1497,7 +1601,7 @@ module Addressable
|
|
1497
1601
|
# @return [String] The path's basename.
|
1498
1602
|
def basename
|
1499
1603
|
# Path cannot be nil
|
1500
|
-
return File.basename(self.path).
|
1604
|
+
return File.basename(self.path).sub(/;[^\/]*$/, EMPTY_STR)
|
1501
1605
|
end
|
1502
1606
|
|
1503
1607
|
##
|
@@ -1529,13 +1633,21 @@ module Addressable
|
|
1529
1633
|
modified_query_class = Addressable::URI::CharacterClasses::QUERY.dup
|
1530
1634
|
# Make sure possible key-value pair delimiters are escaped.
|
1531
1635
|
modified_query_class.sub!("\\&", "").sub!("\\;", "")
|
1532
|
-
pairs = (
|
1636
|
+
pairs = (query || "").split("&", -1)
|
1637
|
+
pairs.delete_if(&:empty?).uniq! if flags.include?(:compacted)
|
1533
1638
|
pairs.sort! if flags.include?(:sorted)
|
1534
1639
|
component = pairs.map do |pair|
|
1535
|
-
Addressable::URI.normalize_component(
|
1640
|
+
Addressable::URI.normalize_component(
|
1641
|
+
pair,
|
1642
|
+
Addressable::URI::NormalizeCharacterClasses::QUERY,
|
1643
|
+
"+"
|
1644
|
+
)
|
1536
1645
|
end.join("&")
|
1537
1646
|
component == "" ? nil : component
|
1538
1647
|
end
|
1648
|
+
# All normalized values should be UTF-8
|
1649
|
+
@normalized_query.force_encoding(Encoding::UTF_8) if @normalized_query
|
1650
|
+
@normalized_query
|
1539
1651
|
end
|
1540
1652
|
|
1541
1653
|
##
|
@@ -1588,11 +1700,13 @@ module Addressable
|
|
1588
1700
|
# so it's best to make all changes in-place.
|
1589
1701
|
pair[0] = URI.unencode_component(pair[0])
|
1590
1702
|
if pair[1].respond_to?(:to_str)
|
1703
|
+
value = pair[1].to_str
|
1591
1704
|
# I loathe the fact that I have to do this. Stupid HTML 4.01.
|
1592
1705
|
# Treating '+' as a space was just an unbelievably bad idea.
|
1593
1706
|
# There was nothing wrong with '%20'!
|
1594
1707
|
# If it ain't broke, don't fix it!
|
1595
|
-
|
1708
|
+
value = value.tr("+", " ") if ["http", "https", nil].include?(scheme)
|
1709
|
+
pair[1] = URI.unencode_component(value)
|
1596
1710
|
end
|
1597
1711
|
if return_type == Hash
|
1598
1712
|
accu[pair[0]] = pair[1]
|
@@ -1644,7 +1758,7 @@ module Addressable
|
|
1644
1758
|
end
|
1645
1759
|
|
1646
1760
|
# new_query_values have form [['key1', 'value1'], ['key2', 'value2']]
|
1647
|
-
buffer = ""
|
1761
|
+
buffer = "".dup
|
1648
1762
|
new_query_values.each do |key, value|
|
1649
1763
|
encoded_key = URI.encode_component(
|
1650
1764
|
key, CharacterClasses::UNRESERVED
|
@@ -1674,7 +1788,7 @@ module Addressable
|
|
1674
1788
|
#
|
1675
1789
|
# @return [String] The request URI required for an HTTP request.
|
1676
1790
|
def request_uri
|
1677
|
-
return nil if self.absolute? && self.scheme !~ /^https?$/
|
1791
|
+
return nil if self.absolute? && self.scheme !~ /^https?$/i
|
1678
1792
|
return (
|
1679
1793
|
(!self.path.empty? ? self.path : SLASH) +
|
1680
1794
|
(self.query ? "?#{self.query}" : EMPTY_STR)
|
@@ -1689,12 +1803,12 @@ module Addressable
|
|
1689
1803
|
if !new_request_uri.respond_to?(:to_str)
|
1690
1804
|
raise TypeError, "Can't convert #{new_request_uri.class} into String."
|
1691
1805
|
end
|
1692
|
-
if self.absolute? && self.scheme !~ /^https?$/
|
1806
|
+
if self.absolute? && self.scheme !~ /^https?$/i
|
1693
1807
|
raise InvalidURIError,
|
1694
1808
|
"Cannot set an HTTP request URI for a non-HTTP URI."
|
1695
1809
|
end
|
1696
1810
|
new_request_uri = new_request_uri.to_str
|
1697
|
-
path_component = new_request_uri[/^([^\?]*)
|
1811
|
+
path_component = new_request_uri[/^([^\?]*)\??(?:.*)$/, 1]
|
1698
1812
|
query_component = new_request_uri[/^(?:[^\?]*)\?(.*)$/, 1]
|
1699
1813
|
path_component = path_component.to_s
|
1700
1814
|
path_component = (!path_component.empty? ? path_component : SLASH)
|
@@ -1723,10 +1837,15 @@ module Addressable
|
|
1723
1837
|
@normalized_fragment ||= begin
|
1724
1838
|
component = Addressable::URI.normalize_component(
|
1725
1839
|
self.fragment,
|
1726
|
-
Addressable::URI::
|
1840
|
+
Addressable::URI::NormalizeCharacterClasses::FRAGMENT
|
1727
1841
|
)
|
1728
1842
|
component == "" ? nil : component
|
1729
1843
|
end
|
1844
|
+
# All normalized values should be UTF-8
|
1845
|
+
if @normalized_fragment
|
1846
|
+
@normalized_fragment.force_encoding(Encoding::UTF_8)
|
1847
|
+
end
|
1848
|
+
@normalized_fragment
|
1730
1849
|
end
|
1731
1850
|
|
1732
1851
|
##
|
@@ -1844,8 +1963,8 @@ module Addressable
|
|
1844
1963
|
# Section 5.2.3 of RFC 3986
|
1845
1964
|
#
|
1846
1965
|
# Removes the right-most path segment from the base path.
|
1847
|
-
if base_path
|
1848
|
-
base_path.
|
1966
|
+
if base_path.include?(SLASH)
|
1967
|
+
base_path.sub!(/\/[^\/]+$/, SLASH)
|
1849
1968
|
else
|
1850
1969
|
base_path = EMPTY_STR
|
1851
1970
|
end
|
@@ -2246,15 +2365,13 @@ module Addressable
|
|
2246
2365
|
"Cannot assemble URI string with ambiguous path: '#{self.path}'"
|
2247
2366
|
end
|
2248
2367
|
@uri_string ||= begin
|
2249
|
-
uri_string =
|
2368
|
+
uri_string = String.new
|
2250
2369
|
uri_string << "#{self.scheme}:" if self.scheme != nil
|
2251
2370
|
uri_string << "//#{self.authority}" if self.authority != nil
|
2252
2371
|
uri_string << self.path.to_s
|
2253
2372
|
uri_string << "?#{self.query}" if self.query != nil
|
2254
2373
|
uri_string << "##{self.fragment}" if self.fragment != nil
|
2255
|
-
|
2256
|
-
uri_string.force_encoding(Encoding::UTF_8)
|
2257
|
-
end
|
2374
|
+
uri_string.force_encoding(Encoding::UTF_8)
|
2258
2375
|
uri_string
|
2259
2376
|
end
|
2260
2377
|
end
|
@@ -2296,10 +2413,10 @@ module Addressable
|
|
2296
2413
|
#
|
2297
2414
|
# @param [Proc] block
|
2298
2415
|
# A set of operations to perform on a given URI.
|
2299
|
-
def defer_validation
|
2300
|
-
raise LocalJumpError, "No block given." unless
|
2416
|
+
def defer_validation
|
2417
|
+
raise LocalJumpError, "No block given." unless block_given?
|
2301
2418
|
@validation_deferred = true
|
2302
|
-
|
2419
|
+
yield
|
2303
2420
|
@validation_deferred = false
|
2304
2421
|
validate
|
2305
2422
|
return nil
|
@@ -2371,6 +2488,19 @@ module Addressable
|
|
2371
2488
|
raise InvalidURIError,
|
2372
2489
|
"Cannot have a relative path with an authority set: '#{self.to_s}'"
|
2373
2490
|
end
|
2491
|
+
if self.path != nil && !self.path.empty? &&
|
2492
|
+
self.path[0..1] == SLASH + SLASH && self.authority == nil
|
2493
|
+
raise InvalidURIError,
|
2494
|
+
"Cannot have a path with two leading slashes " +
|
2495
|
+
"without an authority set: '#{self.to_s}'"
|
2496
|
+
end
|
2497
|
+
unreserved = CharacterClasses::UNRESERVED
|
2498
|
+
sub_delims = CharacterClasses::SUB_DELIMS
|
2499
|
+
if !self.host.nil? && (self.host =~ /[<>{}\/\\\?\#\@"[[:space:]]]/ ||
|
2500
|
+
(self.host[/^\[(.*)\]$/, 1] != nil && self.host[/^\[(.*)\]$/, 1] !~
|
2501
|
+
Regexp.new("^[#{unreserved}#{sub_delims}:]*$")))
|
2502
|
+
raise InvalidURIError, "Invalid character in host: '#{self.host.to_s}'"
|
2503
|
+
end
|
2374
2504
|
return nil
|
2375
2505
|
end
|
2376
2506
|
|
@@ -2384,7 +2514,9 @@ module Addressable
|
|
2384
2514
|
def replace_self(uri)
|
2385
2515
|
# Reset dependent values
|
2386
2516
|
instance_variables.each do |var|
|
2387
|
-
|
2517
|
+
if instance_variable_defined?(var) && var != :@validation_deferred
|
2518
|
+
remove_instance_variable(var)
|
2519
|
+
end
|
2388
2520
|
end
|
2389
2521
|
|
2390
2522
|
@scheme = uri.scheme
|