addressable 2.5.0 → 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 +40 -0
- data/Gemfile +13 -17
- data/README.md +14 -15
- data/Rakefile +5 -3
- data/addressable.gemspec +37 -0
- data/lib/addressable.rb +2 -0
- data/lib/addressable/idna.rb +2 -0
- data/lib/addressable/idna/native.rb +4 -2
- data/lib/addressable/idna/pure.rb +54 -53
- data/lib/addressable/template.rb +63 -83
- data/lib/addressable/uri.rb +128 -64
- data/lib/addressable/version.rb +3 -1
- data/spec/addressable/idna_spec.rb +23 -7
- data/spec/addressable/net_http_compat_spec.rb +2 -0
- data/spec/addressable/security_spec.rb +2 -0
- data/spec/addressable/template_spec.rb +75 -2
- data/spec/addressable/uri_spec.rb +408 -215
- data/spec/spec_helper.rb +12 -0
- data/tasks/clobber.rake +2 -0
- data/tasks/gem.rake +8 -7
- 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 +22 -16
- data/spec/addressable/rack_mount_compat_spec.rb +0 -104
data/lib/addressable/template.rb
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
# encoding:utf-8
|
2
4
|
#--
|
3
5
|
# Copyright (C) Bob Aman
|
@@ -35,7 +37,7 @@ module Addressable
|
|
35
37
|
Addressable::URI::CharacterClasses::DIGIT + '_'
|
36
38
|
|
37
39
|
var_char =
|
38
|
-
"(
|
40
|
+
"(?>(?:[#{variable_char_class}]|%[a-fA-F0-9][a-fA-F0-9])+)"
|
39
41
|
RESERVED =
|
40
42
|
"(?:[#{anything}]|%[a-fA-F0-9][a-fA-F0-9])"
|
41
43
|
UNRESERVED =
|
@@ -410,7 +412,7 @@ module Addressable
|
|
410
412
|
# match.captures
|
411
413
|
# #=> ["a", ["b", "c"]]
|
412
414
|
def match(uri, processor=nil)
|
413
|
-
uri = Addressable::URI.parse(uri)
|
415
|
+
uri = Addressable::URI.parse(uri) unless uri.is_a?(Addressable::URI)
|
414
416
|
mapping = {}
|
415
417
|
|
416
418
|
# First, we need to process the pattern, and extract the values.
|
@@ -488,6 +490,8 @@ module Addressable
|
|
488
490
|
# @param [Hash] mapping The mapping that corresponds to the pattern.
|
489
491
|
# @param [#validate, #transform] processor
|
490
492
|
# An optional processor object may be supplied.
|
493
|
+
# @param [Boolean] normalize_values
|
494
|
+
# Optional flag to enable/disable unicode normalization. Default: true
|
491
495
|
#
|
492
496
|
# The object should respond to either the <tt>validate</tt> or
|
493
497
|
# <tt>transform</tt> messages or both. Both the <tt>validate</tt> and
|
@@ -518,11 +522,11 @@ module Addressable
|
|
518
522
|
# "http://example.com/{?one,two,three}/"
|
519
523
|
# ).partial_expand({"one" => "1", "three" => 3}).pattern
|
520
524
|
# #=> "http://example.com/?one=1{&two}&three=3"
|
521
|
-
def partial_expand(mapping, processor=nil)
|
525
|
+
def partial_expand(mapping, processor=nil, normalize_values=true)
|
522
526
|
result = self.pattern.dup
|
523
527
|
mapping = normalize_keys(mapping)
|
524
528
|
result.gsub!( EXPRESSION ) do |capture|
|
525
|
-
transform_partial_capture(mapping, capture, processor)
|
529
|
+
transform_partial_capture(mapping, capture, processor, normalize_values)
|
526
530
|
end
|
527
531
|
return Addressable::Template.new(result)
|
528
532
|
end
|
@@ -533,6 +537,8 @@ module Addressable
|
|
533
537
|
# @param [Hash] mapping The mapping that corresponds to the pattern.
|
534
538
|
# @param [#validate, #transform] processor
|
535
539
|
# An optional processor object may be supplied.
|
540
|
+
# @param [Boolean] normalize_values
|
541
|
+
# Optional flag to enable/disable unicode normalization. Default: true
|
536
542
|
#
|
537
543
|
# The object should respond to either the <tt>validate</tt> or
|
538
544
|
# <tt>transform</tt> messages or both. Both the <tt>validate</tt> and
|
@@ -583,11 +589,11 @@ module Addressable
|
|
583
589
|
# ExampleProcessor
|
584
590
|
# ).to_str
|
585
591
|
# #=> Addressable::Template::InvalidTemplateValueError
|
586
|
-
def expand(mapping, processor=nil)
|
592
|
+
def expand(mapping, processor=nil, normalize_values=true)
|
587
593
|
result = self.pattern.dup
|
588
594
|
mapping = normalize_keys(mapping)
|
589
595
|
result.gsub!( EXPRESSION ) do |capture|
|
590
|
-
transform_capture(mapping, capture, processor)
|
596
|
+
transform_capture(mapping, capture, processor, normalize_values)
|
591
597
|
end
|
592
598
|
return Addressable::URI.parse(result)
|
593
599
|
end
|
@@ -647,40 +653,6 @@ module Addressable
|
|
647
653
|
self.to_regexp.named_captures
|
648
654
|
end
|
649
655
|
|
650
|
-
##
|
651
|
-
# Generates a route result for a given set of parameters.
|
652
|
-
# Should only be used by rack-mount.
|
653
|
-
#
|
654
|
-
# @param params [Hash] The set of parameters used to expand the template.
|
655
|
-
# @param recall [Hash] Default parameters used to expand the template.
|
656
|
-
# @param options [Hash] Either a `:processor` or a `:parameterize` block.
|
657
|
-
#
|
658
|
-
# @api private
|
659
|
-
def generate(params={}, recall={}, options={})
|
660
|
-
merged = recall.merge(params)
|
661
|
-
if options[:processor]
|
662
|
-
processor = options[:processor]
|
663
|
-
elsif options[:parameterize]
|
664
|
-
# TODO: This is sending me into fits trying to shoe-horn this into
|
665
|
-
# the existing API. I think I've got this backwards and processors
|
666
|
-
# should be a set of 4 optional blocks named :validate, :transform,
|
667
|
-
# :match, and :restore. Having to use a singleton here is a huge
|
668
|
-
# code smell.
|
669
|
-
processor = Object.new
|
670
|
-
class <<processor
|
671
|
-
attr_accessor :block
|
672
|
-
def transform(name, value)
|
673
|
-
block.call(name, value)
|
674
|
-
end
|
675
|
-
end
|
676
|
-
processor.block = options[:parameterize]
|
677
|
-
else
|
678
|
-
processor = nil
|
679
|
-
end
|
680
|
-
result = self.expand(merged, processor)
|
681
|
-
result.to_s if result
|
682
|
-
end
|
683
|
-
|
684
656
|
private
|
685
657
|
def ordered_variable_defaults
|
686
658
|
@ordered_variable_defaults ||= begin
|
@@ -704,6 +676,8 @@ module Addressable
|
|
704
676
|
# The expression to expand
|
705
677
|
# @param [#validate, #transform] processor
|
706
678
|
# An optional processor object may be supplied.
|
679
|
+
# @param [Boolean] normalize_values
|
680
|
+
# Optional flag to enable/disable unicode normalization. Default: true
|
707
681
|
#
|
708
682
|
# The object should respond to either the <tt>validate</tt> or
|
709
683
|
# <tt>transform</tt> messages or both. Both the <tt>validate</tt> and
|
@@ -718,56 +692,36 @@ module Addressable
|
|
718
692
|
# after sending the value to the transform method.
|
719
693
|
#
|
720
694
|
# @return [String] The expanded expression
|
721
|
-
def transform_partial_capture(mapping, capture, processor = nil
|
695
|
+
def transform_partial_capture(mapping, capture, processor = nil,
|
696
|
+
normalize_values = true)
|
722
697
|
_, operator, varlist = *capture.match(EXPRESSION)
|
723
698
|
|
724
|
-
vars = varlist.split(
|
699
|
+
vars = varlist.split(",")
|
725
700
|
|
726
|
-
if
|
701
|
+
if operator == "?"
|
727
702
|
# partial expansion of form style query variables sometimes requires a
|
728
703
|
# slight reordering of the variables to produce a valid url.
|
729
704
|
first_to_expand = vars.find { |varspec|
|
730
705
|
_, name, _ = *varspec.match(VARSPEC)
|
731
|
-
mapping.key? name
|
706
|
+
mapping.key?(name) && !mapping[name].nil?
|
732
707
|
}
|
733
708
|
|
734
709
|
vars = [first_to_expand] + vars.reject {|varspec| varspec == first_to_expand} if first_to_expand
|
735
710
|
end
|
736
711
|
|
737
|
-
vars
|
738
|
-
|
739
|
-
.reduce("") do |acc, (varspec, op)|
|
712
|
+
vars.
|
713
|
+
inject("".dup) do |acc, varspec|
|
740
714
|
_, name, _ = *varspec.match(VARSPEC)
|
741
|
-
|
742
|
-
|
743
|
-
|
744
|
-
|
745
|
-
|
746
|
-
|
747
|
-
|
748
|
-
|
749
|
-
|
750
|
-
|
751
|
-
# Creates a lazy Enumerator of the operators that should be used to expand
|
752
|
-
# variables in a varlist starting with `operator`. For example, an operator
|
753
|
-
# `"?"` results in the sequence `"?","&","&"...`
|
754
|
-
#
|
755
|
-
# @param [String] operator from which to generate a sequence
|
756
|
-
#
|
757
|
-
# @return [Enumerator] sequence of operators
|
758
|
-
def operator_sequence(operator)
|
759
|
-
rest_operator = if "?" == operator
|
760
|
-
"&"
|
761
|
-
else
|
762
|
-
operator
|
763
|
-
end
|
764
|
-
head_operator = operator
|
765
|
-
|
766
|
-
Enumerator.new do |y|
|
767
|
-
y << head_operator.to_s
|
768
|
-
while true
|
769
|
-
y << rest_operator.to_s
|
770
|
-
end
|
715
|
+
next_val = if mapping.key? name
|
716
|
+
transform_capture(mapping, "{#{operator}#{varspec}}",
|
717
|
+
processor, normalize_values)
|
718
|
+
else
|
719
|
+
"{#{operator}#{varspec}}"
|
720
|
+
end
|
721
|
+
# If we've already expanded at least one '?' operator with non-empty
|
722
|
+
# value, change to '&'
|
723
|
+
operator = "&" if (operator == "?") && (next_val != "")
|
724
|
+
acc << next_val
|
771
725
|
end
|
772
726
|
end
|
773
727
|
|
@@ -780,6 +734,9 @@ module Addressable
|
|
780
734
|
# The expression to replace
|
781
735
|
# @param [#validate, #transform] processor
|
782
736
|
# An optional processor object may be supplied.
|
737
|
+
# @param [Boolean] normalize_values
|
738
|
+
# Optional flag to enable/disable unicode normalization. Default: true
|
739
|
+
#
|
783
740
|
#
|
784
741
|
# The object should respond to either the <tt>validate</tt> or
|
785
742
|
# <tt>transform</tt> messages or both. Both the <tt>validate</tt> and
|
@@ -794,7 +751,8 @@ module Addressable
|
|
794
751
|
# after sending the value to the transform method.
|
795
752
|
#
|
796
753
|
# @return [String] The expanded expression
|
797
|
-
def transform_capture(mapping, capture, processor=nil
|
754
|
+
def transform_capture(mapping, capture, processor=nil,
|
755
|
+
normalize_values=true)
|
798
756
|
_, operator, varlist = *capture.match(EXPRESSION)
|
799
757
|
return_value = varlist.split(',').inject([]) do |acc, varspec|
|
800
758
|
_, name, modifier = *varspec.match(VARSPEC)
|
@@ -814,7 +772,7 @@ module Addressable
|
|
814
772
|
"Can't convert #{value.class} into String or Array."
|
815
773
|
end
|
816
774
|
|
817
|
-
value = normalize_value(value)
|
775
|
+
value = normalize_value(value) if normalize_values
|
818
776
|
|
819
777
|
if processor == nil || !processor.respond_to?(:transform)
|
820
778
|
# Handle percent escaping
|
@@ -877,7 +835,9 @@ module Addressable
|
|
877
835
|
end
|
878
836
|
if processor.respond_to?(:transform)
|
879
837
|
transformed_value = processor.transform(name, value)
|
880
|
-
|
838
|
+
if normalize_values
|
839
|
+
transformed_value = normalize_value(transformed_value)
|
840
|
+
end
|
881
841
|
end
|
882
842
|
end
|
883
843
|
acc << [name, transformed_value]
|
@@ -979,15 +939,35 @@ module Addressable
|
|
979
939
|
end
|
980
940
|
end
|
981
941
|
|
942
|
+
##
|
943
|
+
# Generates the <tt>Regexp</tt> that parses a template pattern. Memoizes the
|
944
|
+
# value if template processor not set (processors may not be deterministic)
|
945
|
+
#
|
946
|
+
# @param [String] pattern The URI template pattern.
|
947
|
+
# @param [#match] processor The template processor to use.
|
948
|
+
#
|
949
|
+
# @return [Array, Regexp]
|
950
|
+
# An array of expansion variables nad a regular expression which may be
|
951
|
+
# used to parse a template pattern
|
952
|
+
def parse_template_pattern(pattern, processor = nil)
|
953
|
+
if processor.nil? && pattern == @pattern
|
954
|
+
@cached_template_parse ||=
|
955
|
+
parse_new_template_pattern(pattern, processor)
|
956
|
+
else
|
957
|
+
parse_new_template_pattern(pattern, processor)
|
958
|
+
end
|
959
|
+
end
|
960
|
+
|
982
961
|
##
|
983
962
|
# Generates the <tt>Regexp</tt> that parses a template pattern.
|
984
963
|
#
|
985
964
|
# @param [String] pattern The URI template pattern.
|
986
965
|
# @param [#match] processor The template processor to use.
|
987
966
|
#
|
988
|
-
# @return [Regexp]
|
989
|
-
#
|
990
|
-
|
967
|
+
# @return [Array, Regexp]
|
968
|
+
# An array of expansion variables nad a regular expression which may be
|
969
|
+
# used to parse a template pattern
|
970
|
+
def parse_new_template_pattern(pattern, processor = nil)
|
991
971
|
# Escape the pattern. The two gsubs restore the escaped curly braces
|
992
972
|
# back to their original form. Basically, escape everything that isn't
|
993
973
|
# within an expansion.
|
data/lib/addressable/uri.rb
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
# encoding:utf-8
|
2
4
|
#--
|
3
5
|
# Copyright (C) Bob Aman
|
@@ -46,12 +48,21 @@ module Addressable
|
|
46
48
|
PCHAR = UNRESERVED + SUB_DELIMS + "\\:\\@"
|
47
49
|
SCHEME = ALPHA + DIGIT + "\\-\\+\\."
|
48
50
|
HOST = UNRESERVED + SUB_DELIMS + "\\[\\:\\]"
|
49
|
-
AUTHORITY = PCHAR
|
51
|
+
AUTHORITY = PCHAR + "\\[\\:\\]"
|
50
52
|
PATH = PCHAR + "\\/"
|
51
53
|
QUERY = PCHAR + "\\/\\?"
|
52
54
|
FRAGMENT = PCHAR + "\\/\\?"
|
53
55
|
end
|
54
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
|
+
|
55
66
|
SLASH = '/'
|
56
67
|
EMPTY_STR = ''
|
57
68
|
|
@@ -71,7 +82,7 @@ module Addressable
|
|
71
82
|
"wais" => 210,
|
72
83
|
"ldap" => 389,
|
73
84
|
"prospero" => 1525
|
74
|
-
}
|
85
|
+
}.freeze
|
75
86
|
|
76
87
|
##
|
77
88
|
# Returns a URI object based on the parsed string.
|
@@ -122,9 +133,9 @@ module Addressable
|
|
122
133
|
user = userinfo.strip[/^([^:]*):?/, 1]
|
123
134
|
password = userinfo.strip[/:(.*)$/, 1]
|
124
135
|
end
|
125
|
-
host = authority.
|
136
|
+
host = authority.sub(
|
126
137
|
/^([^\[\]]*)@/, EMPTY_STR
|
127
|
-
).
|
138
|
+
).sub(
|
128
139
|
/:([^:@\[\]]*?)$/, EMPTY_STR
|
129
140
|
)
|
130
141
|
port = authority[/:([^:@\[\]]*?)$/, 1]
|
@@ -182,26 +193,33 @@ module Addressable
|
|
182
193
|
:scheme => "http"
|
183
194
|
}.merge(hints)
|
184
195
|
case uri
|
185
|
-
when /^http
|
186
|
-
uri.
|
187
|
-
when /^https
|
188
|
-
uri.
|
189
|
-
when /^feed:\/+http
|
190
|
-
uri.
|
191
|
-
when /^feed
|
192
|
-
uri.
|
193
|
-
when
|
194
|
-
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:///")
|
195
210
|
when /^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}/
|
196
|
-
uri.
|
211
|
+
uri.sub!(/^/, hints[:scheme] + "://")
|
212
|
+
when /\A\d+\..*:\d+\z/
|
213
|
+
uri = "#{hints[:scheme]}://#{uri}"
|
197
214
|
end
|
198
215
|
match = uri.match(URIREGEX)
|
199
216
|
fragments = match.captures
|
200
217
|
authority = fragments[3]
|
201
218
|
if authority && authority.length > 0
|
202
|
-
new_authority = authority.
|
219
|
+
new_authority = authority.tr("\\", "/").gsub(" ", "%20")
|
203
220
|
# NOTE: We want offset 4, not 3!
|
204
221
|
offset = match.offset(4)
|
222
|
+
uri = uri.dup
|
205
223
|
uri[offset[0]...offset[1]] = new_authority
|
206
224
|
end
|
207
225
|
parsed = self.parse(uri)
|
@@ -209,10 +227,11 @@ module Addressable
|
|
209
227
|
parsed = self.parse(hints[:scheme] + "://" + uri)
|
210
228
|
end
|
211
229
|
if parsed.path.include?(".")
|
212
|
-
|
213
|
-
|
230
|
+
if parsed.path[/\b@\b/]
|
231
|
+
parsed.scheme = "mailto" unless parsed.scheme
|
232
|
+
elsif new_host = parsed.path[/^([^\/]+\.[^\/]*)/, 1]
|
214
233
|
parsed.defer_validation do
|
215
|
-
new_path = parsed.path.
|
234
|
+
new_path = parsed.path.sub(
|
216
235
|
Regexp.new("^" + Regexp.escape(new_host)), EMPTY_STR)
|
217
236
|
parsed.host = new_host
|
218
237
|
parsed.path = new_path
|
@@ -263,24 +282,24 @@ module Addressable
|
|
263
282
|
# Otherwise, convert to a String
|
264
283
|
path = path.to_str.strip
|
265
284
|
|
266
|
-
path.
|
285
|
+
path.sub!(/^file:\/?\/?/, EMPTY_STR) if path =~ /^file:\/?\/?/
|
267
286
|
path = SLASH + path if path =~ /^([a-zA-Z])[\|:]/
|
268
287
|
uri = self.parse(path)
|
269
288
|
|
270
289
|
if uri.scheme == nil
|
271
290
|
# Adjust windows-style uris
|
272
|
-
uri.path.
|
291
|
+
uri.path.sub!(/^\/?([a-zA-Z])[\|:][\\\/]/) do
|
273
292
|
"/#{$1.downcase}:/"
|
274
293
|
end
|
275
|
-
uri.path.
|
294
|
+
uri.path.tr!("\\", SLASH)
|
276
295
|
if File.exist?(uri.path) &&
|
277
296
|
File.stat(uri.path).directory?
|
278
|
-
uri.path.
|
297
|
+
uri.path.chomp!(SLASH)
|
279
298
|
uri.path = uri.path + '/'
|
280
299
|
end
|
281
300
|
|
282
301
|
# If the path is absolute, set the scheme and host.
|
283
|
-
if uri.path
|
302
|
+
if uri.path.start_with?(SLASH)
|
284
303
|
uri.scheme = "file"
|
285
304
|
uri.host = EMPTY_STR
|
286
305
|
end
|
@@ -317,6 +336,21 @@ module Addressable
|
|
317
336
|
return result
|
318
337
|
end
|
319
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
|
+
|
320
354
|
##
|
321
355
|
# Percent encodes a URI component.
|
322
356
|
#
|
@@ -383,18 +417,20 @@ module Addressable
|
|
383
417
|
component.force_encoding(Encoding::ASCII_8BIT)
|
384
418
|
# Avoiding gsub! because there are edge cases with frozen strings
|
385
419
|
component = component.gsub(character_class) do |sequence|
|
386
|
-
|
420
|
+
SEQUENCE_UPCASED_PERCENT_ENCODING_TABLE[sequence]
|
387
421
|
end
|
388
422
|
if upcase_encoded.length > 0
|
389
|
-
|
390
|
-
char
|
391
|
-
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)
|
392
428
|
end
|
393
429
|
return component
|
394
430
|
end
|
395
431
|
|
396
432
|
class << self
|
397
|
-
alias_method :
|
433
|
+
alias_method :escape_component, :encode_component
|
398
434
|
end
|
399
435
|
|
400
436
|
##
|
@@ -436,7 +472,11 @@ module Addressable
|
|
436
472
|
uri = uri.dup
|
437
473
|
# Seriously, only use UTF-8. I'm really not kidding!
|
438
474
|
uri.force_encoding("utf-8")
|
439
|
-
|
475
|
+
|
476
|
+
unless leave_encoded.empty?
|
477
|
+
leave_encoded = leave_encoded.dup.force_encoding("utf-8")
|
478
|
+
end
|
479
|
+
|
440
480
|
result = uri.gsub(/%[0-9a-f]{2}/iu) do |sequence|
|
441
481
|
c = sequence[1..3].to_i(16).chr
|
442
482
|
c.force_encoding("utf-8")
|
@@ -522,12 +562,16 @@ module Addressable
|
|
522
562
|
character_class = "#{character_class}%" unless character_class.include?('%')
|
523
563
|
|
524
564
|
"|%(?!#{leave_encoded.chars.map do |char|
|
525
|
-
seq = char
|
565
|
+
seq = SEQUENCE_ENCODING_TABLE[char]
|
526
566
|
[seq.upcase, seq.downcase]
|
527
567
|
end.flatten.join('|')})"
|
528
568
|
end
|
529
569
|
|
530
|
-
character_class =
|
570
|
+
character_class = if leave_re
|
571
|
+
/[^#{character_class}]#{leave_re}/
|
572
|
+
else
|
573
|
+
/[^#{character_class}]/
|
574
|
+
end
|
531
575
|
end
|
532
576
|
# We can't perform regexps on invalid UTF sequences, but
|
533
577
|
# here we need to, so switch to ASCII.
|
@@ -847,11 +891,11 @@ module Addressable
|
|
847
891
|
return nil unless self.scheme
|
848
892
|
@normalized_scheme ||= begin
|
849
893
|
if self.scheme =~ /^\s*ssh\+svn\s*$/i
|
850
|
-
"svn+ssh"
|
894
|
+
"svn+ssh".dup
|
851
895
|
else
|
852
896
|
Addressable::URI.normalize_component(
|
853
897
|
self.scheme.strip.downcase,
|
854
|
-
Addressable::URI::
|
898
|
+
Addressable::URI::NormalizeCharacterClasses::SCHEME
|
855
899
|
)
|
856
900
|
end
|
857
901
|
end
|
@@ -871,7 +915,7 @@ module Addressable
|
|
871
915
|
new_scheme = new_scheme.to_str
|
872
916
|
end
|
873
917
|
if new_scheme && new_scheme !~ /\A[a-z][a-z0-9\.\+\-]*\z/i
|
874
|
-
raise InvalidURIError, "Invalid scheme format: #{new_scheme}"
|
918
|
+
raise InvalidURIError, "Invalid scheme format: '#{new_scheme}'"
|
875
919
|
end
|
876
920
|
@scheme = new_scheme
|
877
921
|
@scheme = nil if @scheme.to_s.strip.empty?
|
@@ -906,7 +950,7 @@ module Addressable
|
|
906
950
|
else
|
907
951
|
Addressable::URI.normalize_component(
|
908
952
|
self.user.strip,
|
909
|
-
Addressable::URI::
|
953
|
+
Addressable::URI::NormalizeCharacterClasses::UNRESERVED
|
910
954
|
)
|
911
955
|
end
|
912
956
|
end
|
@@ -963,7 +1007,7 @@ module Addressable
|
|
963
1007
|
else
|
964
1008
|
Addressable::URI.normalize_component(
|
965
1009
|
self.password.strip,
|
966
|
-
Addressable::URI::
|
1010
|
+
Addressable::URI::NormalizeCharacterClasses::UNRESERVED
|
967
1011
|
)
|
968
1012
|
end
|
969
1013
|
end
|
@@ -1032,9 +1076,9 @@ module Addressable
|
|
1032
1076
|
if !current_user && !current_password
|
1033
1077
|
nil
|
1034
1078
|
elsif current_user && current_password
|
1035
|
-
"#{current_user}:#{current_password}"
|
1079
|
+
"#{current_user}:#{current_password}".dup
|
1036
1080
|
elsif current_user && !current_password
|
1037
|
-
"#{current_user}"
|
1081
|
+
"#{current_user}".dup
|
1038
1082
|
end
|
1039
1083
|
end
|
1040
1084
|
# All normalized values should be UTF-8
|
@@ -1087,6 +1131,7 @@ module Addressable
|
|
1087
1131
|
# @return [String] The host component, normalized.
|
1088
1132
|
def normalized_host
|
1089
1133
|
return nil unless self.host
|
1134
|
+
|
1090
1135
|
@normalized_host ||= begin
|
1091
1136
|
if !self.host.strip.empty?
|
1092
1137
|
result = ::Addressable::IDNA.to_ascii(
|
@@ -1098,14 +1143,17 @@ module Addressable
|
|
1098
1143
|
end
|
1099
1144
|
result = Addressable::URI.normalize_component(
|
1100
1145
|
result,
|
1101
|
-
|
1146
|
+
NormalizeCharacterClasses::HOST
|
1147
|
+
)
|
1102
1148
|
result
|
1103
1149
|
else
|
1104
|
-
EMPTY_STR
|
1150
|
+
EMPTY_STR.dup
|
1105
1151
|
end
|
1106
1152
|
end
|
1107
1153
|
# All normalized values should be UTF-8
|
1108
|
-
@normalized_host
|
1154
|
+
if @normalized_host && !@normalized_host.empty?
|
1155
|
+
@normalized_host.force_encoding(Encoding::UTF_8)
|
1156
|
+
end
|
1109
1157
|
@normalized_host
|
1110
1158
|
end
|
1111
1159
|
|
@@ -1163,16 +1211,25 @@ module Addressable
|
|
1163
1211
|
# Returns the top-level domain for this host.
|
1164
1212
|
#
|
1165
1213
|
# @example
|
1166
|
-
# Addressable::URI.parse("www.example.co.uk").tld # => "co.uk"
|
1214
|
+
# Addressable::URI.parse("http://www.example.co.uk").tld # => "co.uk"
|
1167
1215
|
def tld
|
1168
1216
|
PublicSuffix.parse(self.host, ignore_private: true).tld
|
1169
1217
|
end
|
1170
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
|
+
|
1171
1228
|
##
|
1172
1229
|
# Returns the public suffix domain for this host.
|
1173
1230
|
#
|
1174
1231
|
# @example
|
1175
|
-
# Addressable::URI.parse("www.example.co.uk").domain # => "example.co.uk"
|
1232
|
+
# Addressable::URI.parse("http://www.example.co.uk").domain # => "example.co.uk"
|
1176
1233
|
def domain
|
1177
1234
|
PublicSuffix.domain(self.host, ignore_private: true)
|
1178
1235
|
end
|
@@ -1235,9 +1292,9 @@ module Addressable
|
|
1235
1292
|
new_user = new_userinfo.strip[/^([^:]*):?/, 1]
|
1236
1293
|
new_password = new_userinfo.strip[/:(.*)$/, 1]
|
1237
1294
|
end
|
1238
|
-
new_host = new_authority.
|
1295
|
+
new_host = new_authority.sub(
|
1239
1296
|
/^([^\[\]]*)@/, EMPTY_STR
|
1240
|
-
).
|
1297
|
+
).sub(
|
1241
1298
|
/:([^:@\[\]]*?)$/, EMPTY_STR
|
1242
1299
|
)
|
1243
1300
|
new_port =
|
@@ -1421,7 +1478,7 @@ module Addressable
|
|
1421
1478
|
# @return [String] The components that identify a site.
|
1422
1479
|
def site
|
1423
1480
|
(self.scheme || self.authority) && @site ||= begin
|
1424
|
-
site_string = ""
|
1481
|
+
site_string = "".dup
|
1425
1482
|
site_string << "#{self.scheme}:" if self.scheme != nil
|
1426
1483
|
site_string << "//#{self.authority}" if self.authority != nil
|
1427
1484
|
site_string
|
@@ -1440,7 +1497,7 @@ module Addressable
|
|
1440
1497
|
def normalized_site
|
1441
1498
|
return nil unless self.site
|
1442
1499
|
@normalized_site ||= begin
|
1443
|
-
site_string = ""
|
1500
|
+
site_string = "".dup
|
1444
1501
|
if self.normalized_scheme != nil
|
1445
1502
|
site_string << "#{self.normalized_scheme}:"
|
1446
1503
|
end
|
@@ -1501,14 +1558,14 @@ module Addressable
|
|
1501
1558
|
result = path.strip.split(SLASH, -1).map do |segment|
|
1502
1559
|
Addressable::URI.normalize_component(
|
1503
1560
|
segment,
|
1504
|
-
Addressable::URI::
|
1561
|
+
Addressable::URI::NormalizeCharacterClasses::PCHAR
|
1505
1562
|
)
|
1506
1563
|
end.join(SLASH)
|
1507
1564
|
|
1508
1565
|
result = URI.normalize_path(result)
|
1509
1566
|
if result.empty? &&
|
1510
1567
|
["http", "https", "ftp", "tftp"].include?(self.normalized_scheme)
|
1511
|
-
result = SLASH
|
1568
|
+
result = SLASH.dup
|
1512
1569
|
end
|
1513
1570
|
result
|
1514
1571
|
end
|
@@ -1544,7 +1601,7 @@ module Addressable
|
|
1544
1601
|
# @return [String] The path's basename.
|
1545
1602
|
def basename
|
1546
1603
|
# Path cannot be nil
|
1547
|
-
return File.basename(self.path).
|
1604
|
+
return File.basename(self.path).sub(/;[^\/]*$/, EMPTY_STR)
|
1548
1605
|
end
|
1549
1606
|
|
1550
1607
|
##
|
@@ -1576,10 +1633,15 @@ module Addressable
|
|
1576
1633
|
modified_query_class = Addressable::URI::CharacterClasses::QUERY.dup
|
1577
1634
|
# Make sure possible key-value pair delimiters are escaped.
|
1578
1635
|
modified_query_class.sub!("\\&", "").sub!("\\;", "")
|
1579
|
-
pairs = (
|
1636
|
+
pairs = (query || "").split("&", -1)
|
1637
|
+
pairs.delete_if(&:empty?).uniq! if flags.include?(:compacted)
|
1580
1638
|
pairs.sort! if flags.include?(:sorted)
|
1581
1639
|
component = pairs.map do |pair|
|
1582
|
-
Addressable::URI.normalize_component(
|
1640
|
+
Addressable::URI.normalize_component(
|
1641
|
+
pair,
|
1642
|
+
Addressable::URI::NormalizeCharacterClasses::QUERY,
|
1643
|
+
"+"
|
1644
|
+
)
|
1583
1645
|
end.join("&")
|
1584
1646
|
component == "" ? nil : component
|
1585
1647
|
end
|
@@ -1638,11 +1700,13 @@ module Addressable
|
|
1638
1700
|
# so it's best to make all changes in-place.
|
1639
1701
|
pair[0] = URI.unencode_component(pair[0])
|
1640
1702
|
if pair[1].respond_to?(:to_str)
|
1703
|
+
value = pair[1].to_str
|
1641
1704
|
# I loathe the fact that I have to do this. Stupid HTML 4.01.
|
1642
1705
|
# Treating '+' as a space was just an unbelievably bad idea.
|
1643
1706
|
# There was nothing wrong with '%20'!
|
1644
1707
|
# If it ain't broke, don't fix it!
|
1645
|
-
|
1708
|
+
value = value.tr("+", " ") if ["http", "https", nil].include?(scheme)
|
1709
|
+
pair[1] = URI.unencode_component(value)
|
1646
1710
|
end
|
1647
1711
|
if return_type == Hash
|
1648
1712
|
accu[pair[0]] = pair[1]
|
@@ -1694,7 +1758,7 @@ module Addressable
|
|
1694
1758
|
end
|
1695
1759
|
|
1696
1760
|
# new_query_values have form [['key1', 'value1'], ['key2', 'value2']]
|
1697
|
-
buffer = ""
|
1761
|
+
buffer = "".dup
|
1698
1762
|
new_query_values.each do |key, value|
|
1699
1763
|
encoded_key = URI.encode_component(
|
1700
1764
|
key, CharacterClasses::UNRESERVED
|
@@ -1724,7 +1788,7 @@ module Addressable
|
|
1724
1788
|
#
|
1725
1789
|
# @return [String] The request URI required for an HTTP request.
|
1726
1790
|
def request_uri
|
1727
|
-
return nil if self.absolute? && self.scheme !~ /^https?$/
|
1791
|
+
return nil if self.absolute? && self.scheme !~ /^https?$/i
|
1728
1792
|
return (
|
1729
1793
|
(!self.path.empty? ? self.path : SLASH) +
|
1730
1794
|
(self.query ? "?#{self.query}" : EMPTY_STR)
|
@@ -1739,12 +1803,12 @@ module Addressable
|
|
1739
1803
|
if !new_request_uri.respond_to?(:to_str)
|
1740
1804
|
raise TypeError, "Can't convert #{new_request_uri.class} into String."
|
1741
1805
|
end
|
1742
|
-
if self.absolute? && self.scheme !~ /^https?$/
|
1806
|
+
if self.absolute? && self.scheme !~ /^https?$/i
|
1743
1807
|
raise InvalidURIError,
|
1744
1808
|
"Cannot set an HTTP request URI for a non-HTTP URI."
|
1745
1809
|
end
|
1746
1810
|
new_request_uri = new_request_uri.to_str
|
1747
|
-
path_component = new_request_uri[/^([^\?]*)
|
1811
|
+
path_component = new_request_uri[/^([^\?]*)\??(?:.*)$/, 1]
|
1748
1812
|
query_component = new_request_uri[/^(?:[^\?]*)\?(.*)$/, 1]
|
1749
1813
|
path_component = path_component.to_s
|
1750
1814
|
path_component = (!path_component.empty? ? path_component : SLASH)
|
@@ -1773,7 +1837,7 @@ module Addressable
|
|
1773
1837
|
@normalized_fragment ||= begin
|
1774
1838
|
component = Addressable::URI.normalize_component(
|
1775
1839
|
self.fragment,
|
1776
|
-
Addressable::URI::
|
1840
|
+
Addressable::URI::NormalizeCharacterClasses::FRAGMENT
|
1777
1841
|
)
|
1778
1842
|
component == "" ? nil : component
|
1779
1843
|
end
|
@@ -1899,8 +1963,8 @@ module Addressable
|
|
1899
1963
|
# Section 5.2.3 of RFC 3986
|
1900
1964
|
#
|
1901
1965
|
# Removes the right-most path segment from the base path.
|
1902
|
-
if base_path
|
1903
|
-
base_path.
|
1966
|
+
if base_path.include?(SLASH)
|
1967
|
+
base_path.sub!(/\/[^\/]+$/, SLASH)
|
1904
1968
|
else
|
1905
1969
|
base_path = EMPTY_STR
|
1906
1970
|
end
|
@@ -2349,10 +2413,10 @@ module Addressable
|
|
2349
2413
|
#
|
2350
2414
|
# @param [Proc] block
|
2351
2415
|
# A set of operations to perform on a given URI.
|
2352
|
-
def defer_validation
|
2353
|
-
raise LocalJumpError, "No block given." unless
|
2416
|
+
def defer_validation
|
2417
|
+
raise LocalJumpError, "No block given." unless block_given?
|
2354
2418
|
@validation_deferred = true
|
2355
|
-
|
2419
|
+
yield
|
2356
2420
|
@validation_deferred = false
|
2357
2421
|
validate
|
2358
2422
|
return nil
|