addressable 2.8.1 → 2.8.6
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +76 -42
- data/Rakefile +11 -5
- data/addressable.gemspec +28 -0
- data/lib/addressable/idna/native.rb +8 -2
- data/lib/addressable/idna/pure.rb +10 -182
- data/lib/addressable/template.rb +9 -10
- data/lib/addressable/uri.rb +176 -134
- data/lib/addressable/version.rb +1 -1
- data/spec/addressable/idna_spec.rb +6 -5
- data/spec/addressable/template_spec.rb +60 -264
- data/spec/addressable/uri_spec.rb +95 -0
- data/tasks/gem.rake +1 -1
- metadata +5 -4
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 4023555b4bb6c374726bb9647a29d30943c637115619670a09632fe7a4e28765
|
4
|
+
data.tar.gz: 95ad31a9e1dd01f2bc9a7c097eb452f07ed52be16dcd2ad4946652648dd7808c
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: d7dd6eed173bc4a06a3abb74f6fdbf501ee12b3fcb2828c0f21db98d1779175bce4dc900999269db24cc8647c3148b06cf1d9c768c97c7591a48f1e43751bb7b
|
7
|
+
data.tar.gz: 3bbe976f23a4c3a6776a44d6ff2147b9ab87418162af57b2f278089528416eb3d564b114f96f165946d9126ee68d5da86c0ac01a9e2fa2792547bb312f40f046
|
data/CHANGELOG.md
CHANGED
@@ -1,20 +1,54 @@
|
|
1
|
-
# Addressable 2.8.
|
1
|
+
# Addressable 2.8.6 <a name="v2.8.6">
|
2
|
+
- Memoize regexps for common character classes ([#524])
|
3
|
+
|
4
|
+
[#524]: https://github.com/sporkmonger/addressable/pull/524
|
5
|
+
|
6
|
+
# Addressable 2.8.5 <a name="v2.8.5">
|
7
|
+
- Fix thread safety issue with encoding tables ([#515])
|
8
|
+
- Define URI::NONE as a module to avoid serialization issues ([#509])
|
9
|
+
- Fix YAML serialization ([#508])
|
10
|
+
|
11
|
+
[#508]: https://github.com/sporkmonger/addressable/pull/508
|
12
|
+
[#509]: https://github.com/sporkmonger/addressable/pull/509
|
13
|
+
[#515]: https://github.com/sporkmonger/addressable/pull/515
|
14
|
+
|
15
|
+
# Addressable 2.8.4 <a name="v2.8.4">
|
16
|
+
- Restore `Addressable::IDNA.unicode_normalize_kc` as a deprecated method ([#504])
|
17
|
+
|
18
|
+
[#504]: https://github.com/sporkmonger/addressable/pull/504
|
19
|
+
|
20
|
+
# Addressable 2.8.3 <a name="v2.8.3">
|
21
|
+
- Fix template expand level 2 hash support for non-string objects ([#499], [#498])
|
22
|
+
|
23
|
+
[#499]: https://github.com/sporkmonger/addressable/pull/499
|
24
|
+
[#498]: https://github.com/sporkmonger/addressable/pull/498
|
25
|
+
|
26
|
+
# Addressable 2.8.2 <a name="v2.8.2">
|
27
|
+
- Improve cache hits and JIT friendliness ([#486](https://github.com/sporkmonger/addressable/pull/486))
|
28
|
+
- Improve code style and test coverage ([#482](https://github.com/sporkmonger/addressable/pull/482))
|
29
|
+
- Ensure reset of deferred validation ([#481](https://github.com/sporkmonger/addressable/pull/481))
|
30
|
+
- Resolve normalization differences between `IDNA::Native` and `IDNA::Pure` ([#408](https://github.com/sporkmonger/addressable/issues/408), [#492])
|
31
|
+
- Remove redundant colon in `Addressable::URI::CharacterClasses::AUTHORITY` regex ([#438](https://github.com/sporkmonger/addressable/pull/438)) (accidentally reverted by [#449] merge but [added back](https://github.com/sporkmonger/addressable/pull/492#discussion_r1105125280) in [#492])
|
32
|
+
|
33
|
+
[#492]: https://github.com/sporkmonger/addressable/pull/492
|
34
|
+
|
35
|
+
# Addressable 2.8.1 <a name="v2.8.1">
|
2
36
|
- refactor `Addressable::URI.normalize_path` to address linter offenses ([#430](https://github.com/sporkmonger/addressable/pull/430))
|
3
|
-
- remove redundant colon in `Addressable::URI::CharacterClasses::AUTHORITY` regex ([#438](https://github.com/sporkmonger/addressable/pull/438))
|
4
37
|
- update gemspec to reflect supported Ruby versions ([#466], [#464], [#463])
|
5
38
|
- compatibility w/ public_suffix 5.x ([#466], [#465], [#460])
|
6
39
|
- fixes "invalid byte sequence in UTF-8" exception when unencoding URLs containing non UTF-8 characters ([#459](https://github.com/sporkmonger/addressable/pull/459))
|
7
|
-
- `Ractor` compatibility ([#449]
|
40
|
+
- `Ractor` compatibility ([#449])
|
8
41
|
- use the whole string instead of a single line for template match ([#431](https://github.com/sporkmonger/addressable/pull/431))
|
9
42
|
- force UTF-8 encoding only if needed ([#341](https://github.com/sporkmonger/addressable/pull/341))
|
10
43
|
|
44
|
+
[#449]: https://github.com/sporkmonger/addressable/pull/449
|
11
45
|
[#460]: https://github.com/sporkmonger/addressable/pull/460
|
12
46
|
[#463]: https://github.com/sporkmonger/addressable/pull/463
|
13
47
|
[#464]: https://github.com/sporkmonger/addressable/pull/464
|
14
48
|
[#465]: https://github.com/sporkmonger/addressable/pull/465
|
15
49
|
[#466]: https://github.com/sporkmonger/addressable/pull/466
|
16
50
|
|
17
|
-
# Addressable 2.8.0
|
51
|
+
# Addressable 2.8.0 <a name="v2.8.0">
|
18
52
|
- fixes ReDoS vulnerability in Addressable::Template#match
|
19
53
|
- no longer replaces `+` with spaces in queries for non-http(s) schemes
|
20
54
|
- fixed encoding ipv6 literals
|
@@ -26,14 +60,14 @@
|
|
26
60
|
- performance improvements
|
27
61
|
- switch CI/CD to GitHub Actions
|
28
62
|
|
29
|
-
# Addressable 2.7.0
|
63
|
+
# Addressable 2.7.0 <a name="v2.7.0">
|
30
64
|
- added `:compacted` flag to `normalized_query`
|
31
65
|
- `heuristic_parse` handles `mailto:` more intuitively
|
32
66
|
- dropped explicit support for JRuby 9.0.5.0
|
33
67
|
- compatibility w/ public_suffix 4.x
|
34
68
|
- performance improvements
|
35
69
|
|
36
|
-
# Addressable 2.6.0
|
70
|
+
# Addressable 2.6.0 <a name="v2.6.0">
|
37
71
|
- added `tld=` method to allow assignment to the public suffix
|
38
72
|
- most `heuristic_parse` patterns are now case-insensitive
|
39
73
|
- `heuristic_parse` handles more `file://` URI variations
|
@@ -44,17 +78,17 @@
|
|
44
78
|
- minor performance improvements in regexps
|
45
79
|
- fixes to eliminate warnings
|
46
80
|
|
47
|
-
# Addressable 2.5.2
|
81
|
+
# Addressable 2.5.2 <a name="v2.5.2">
|
48
82
|
- better support for frozen string literals
|
49
83
|
- fixed bug w/ uppercase characters in scheme
|
50
84
|
- IDNA errors w/ emoji URLs
|
51
85
|
- compatibility w/ public_suffix 3.x
|
52
86
|
|
53
|
-
# Addressable 2.5.1
|
87
|
+
# Addressable 2.5.1 <a name="v2.5.1">
|
54
88
|
- allow unicode normalization to be disabled for URI Template expansion
|
55
89
|
- removed duplicate test
|
56
90
|
|
57
|
-
# Addressable 2.5.0
|
91
|
+
# Addressable 2.5.0 <a name="v2.5.0">
|
58
92
|
- dropping support for Ruby 1.9
|
59
93
|
- adding support for Ruby 2.4 preview
|
60
94
|
- add support for public suffixes and tld; first runtime dependency
|
@@ -68,7 +102,7 @@
|
|
68
102
|
- host parts longer than 63 bytes will be ignored and not passed to libidn
|
69
103
|
- normalized values always encoded as UTF-8
|
70
104
|
|
71
|
-
# Addressable 2.4.0
|
105
|
+
# Addressable 2.4.0 <a name="v2.4.0">
|
72
106
|
- support for 1.8.x dropped
|
73
107
|
- double quotes in a host now raises an error
|
74
108
|
- newlines in host will no longer get unescaped during normalization
|
@@ -80,17 +114,17 @@
|
|
80
114
|
- fixed minor bug where an exception would be thrown for a missing ACE suffix
|
81
115
|
- better partial expansion of URI templates
|
82
116
|
|
83
|
-
# Addressable 2.3.8
|
117
|
+
# Addressable 2.3.8 <a name="v2.3.8">
|
84
118
|
- fix warnings
|
85
119
|
- update dependency gems
|
86
120
|
- support for 1.8.x officially deprecated
|
87
121
|
|
88
|
-
# Addressable 2.3.7
|
122
|
+
# Addressable 2.3.7 <a name="v2.3.7">
|
89
123
|
- fix scenario in which invalid URIs don't get an exception until inspected
|
90
124
|
- handle hostnames with two adjacent periods correctly
|
91
125
|
- upgrade of RSpec
|
92
126
|
|
93
|
-
# Addressable 2.3.6
|
127
|
+
# Addressable 2.3.6 <a name="v2.3.6">
|
94
128
|
- normalization drops empty query string
|
95
129
|
- better handling in template extract for missing values
|
96
130
|
- template modifier for `'?'` now treated as optional
|
@@ -99,19 +133,19 @@
|
|
99
133
|
- added `:sorted` option to normalization of query strings
|
100
134
|
- fixed issue with normalization of hosts given in `'example.com.'` form
|
101
135
|
|
102
|
-
# Addressable 2.3.5
|
136
|
+
# Addressable 2.3.5 <a name="v2.3.5">
|
103
137
|
- added Addressable::URI#empty? method
|
104
138
|
- Addressable::URI#hostname methods now strip square brackets from IPv6 hosts
|
105
139
|
- compatibility with Net::HTTP in Ruby 2.0.0
|
106
140
|
- Addressable::URI#route_from should always give relative URIs
|
107
141
|
|
108
|
-
# Addressable 2.3.4
|
142
|
+
# Addressable 2.3.4 <a name="v2.3.4">
|
109
143
|
- fixed issue with encoding altering its inputs
|
110
144
|
- query string normalization now leaves ';' characters alone
|
111
145
|
- FakeFS is detected before attempting to load unicode tables
|
112
146
|
- additional testing to ensure frozen objects don't cause problems
|
113
147
|
|
114
|
-
# Addressable 2.3.3
|
148
|
+
# Addressable 2.3.3 <a name="v2.3.3">
|
115
149
|
- fixed issue with converting common primitives during template expansion
|
116
150
|
- fixed port encoding issue
|
117
151
|
- removed a few warnings
|
@@ -120,59 +154,59 @@
|
|
120
154
|
- no template match should now result in nil instead of an empty MatchData
|
121
155
|
- added license information to gemspec
|
122
156
|
|
123
|
-
# Addressable 2.3.2
|
157
|
+
# Addressable 2.3.2 <a name="v2.3.2">
|
124
158
|
- added Addressable::URI#default_port method
|
125
159
|
- fixed issue with Marshalling Unicode data on Windows
|
126
160
|
- improved heuristic parsing to better handle IPv4 addresses
|
127
161
|
|
128
|
-
# Addressable 2.3.1
|
162
|
+
# Addressable 2.3.1 <a name="v2.3.1">
|
129
163
|
- fixed missing unicode data file
|
130
164
|
|
131
|
-
# Addressable 2.3.0
|
165
|
+
# Addressable 2.3.0 <a name="v2.3.0">
|
132
166
|
- updated Addressable::Template to use RFC 6570, level 4
|
133
167
|
- fixed compatibility problems with some versions of Ruby
|
134
168
|
- moved unicode tables into a data file for performance reasons
|
135
169
|
- removing support for multiple query value notations
|
136
170
|
|
137
|
-
# Addressable 2.2.8
|
171
|
+
# Addressable 2.2.8 <a name="v2.2.8">
|
138
172
|
- fixed issues with dot segment removal code
|
139
173
|
- form encoding can now handle multiple values per key
|
140
174
|
- updated development environment
|
141
175
|
|
142
|
-
# Addressable 2.2.7
|
176
|
+
# Addressable 2.2.7 <a name="v2.2.7">
|
143
177
|
- fixed issues related to Addressable::URI#query_values=
|
144
178
|
- the Addressable::URI.parse method is now polymorphic
|
145
179
|
|
146
|
-
# Addressable 2.2.6
|
180
|
+
# Addressable 2.2.6 <a name="v2.2.6">
|
147
181
|
- changed the way ambiguous paths are handled
|
148
182
|
- fixed bug with frozen URIs
|
149
183
|
- https supported in heuristic parsing
|
150
184
|
|
151
|
-
# Addressable 2.2.5
|
185
|
+
# Addressable 2.2.5 <a name="v2.2.5">
|
152
186
|
- 'parsing' a pre-parsed URI object is now a dup operation
|
153
187
|
- introduced conditional support for libidn
|
154
188
|
- fixed normalization issue on ampersands in query strings
|
155
189
|
- added additional tests around handling of query strings
|
156
190
|
|
157
|
-
# Addressable 2.2.4
|
191
|
+
# Addressable 2.2.4 <a name="v2.2.4">
|
158
192
|
- added origin support from draft-ietf-websec-origin-00
|
159
193
|
- resolved issue with attempting to navigate below root
|
160
194
|
- fixed bug with string splitting in query strings
|
161
195
|
|
162
|
-
# Addressable 2.2.3
|
196
|
+
# Addressable 2.2.3 <a name="v2.2.3">
|
163
197
|
- added :flat_array notation for query strings
|
164
198
|
|
165
|
-
# Addressable 2.2.2
|
199
|
+
# Addressable 2.2.2 <a name="v2.2.2">
|
166
200
|
- fixed issue with percent escaping of '+' character in query strings
|
167
201
|
|
168
|
-
# Addressable 2.2.1
|
202
|
+
# Addressable 2.2.1 <a name="v2.2.1">
|
169
203
|
- added support for application/x-www-form-urlencoded.
|
170
204
|
|
171
|
-
# Addressable 2.2.0
|
205
|
+
# Addressable 2.2.0 <a name="v2.2.0">
|
172
206
|
- added site methods
|
173
207
|
- improved documentation
|
174
208
|
|
175
|
-
# Addressable 2.1.2
|
209
|
+
# Addressable 2.1.2 <a name="v2.1.2">
|
176
210
|
- added HTTP request URI methods
|
177
211
|
- better handling of Windows file paths
|
178
212
|
- validation_deferred boolean replaced with defer_validation block
|
@@ -180,14 +214,14 @@
|
|
180
214
|
- fixed issue with constructing URIs with relative paths
|
181
215
|
- fixed warnings
|
182
216
|
|
183
|
-
# Addressable 2.1.1
|
217
|
+
# Addressable 2.1.1 <a name="v2.1.1">
|
184
218
|
- more type checking changes
|
185
219
|
- fixed issue with unicode normalization
|
186
220
|
- added method to find template defaults
|
187
221
|
- symbolic keys are now allowed in template mappings
|
188
222
|
- numeric values and symbolic values are now allowed in template mappings
|
189
223
|
|
190
|
-
# Addressable 2.1.0
|
224
|
+
# Addressable 2.1.0 <a name="v2.1.0">
|
191
225
|
- refactored URI template support out into its own class
|
192
226
|
- removed extract method due to being useless and unreliable
|
193
227
|
- removed Addressable::URI.expand_template
|
@@ -201,15 +235,15 @@
|
|
201
235
|
- worked around issue with freezing URIs
|
202
236
|
- improved specs
|
203
237
|
|
204
|
-
# Addressable 2.0.2
|
238
|
+
# Addressable 2.0.2 <a name="v2.0.2">
|
205
239
|
- fixed issue with URI template expansion
|
206
240
|
- fixed issue with percent escaping characters 0-15
|
207
241
|
|
208
|
-
# Addressable 2.0.1
|
242
|
+
# Addressable 2.0.1 <a name="v2.0.1">
|
209
243
|
- fixed issue with query string assignment
|
210
244
|
- fixed issue with improperly encoded components
|
211
245
|
|
212
|
-
# Addressable 2.0.0
|
246
|
+
# Addressable 2.0.0 <a name="v2.0.0">
|
213
247
|
- the initialize method now takes an options hash as its only parameter
|
214
248
|
- added query_values method to URI class
|
215
249
|
- completely replaced IDNA implementation with pure Ruby
|
@@ -224,20 +258,20 @@
|
|
224
258
|
- updated URI Template code to match v 03 of the draft spec
|
225
259
|
- added a bunch of new specifications
|
226
260
|
|
227
|
-
# Addressable 1.0.4
|
261
|
+
# Addressable 1.0.4 <a name="v1.0.4">
|
228
262
|
- switched to using RSpec's pending system for specs that rely on IDN
|
229
263
|
- fixed issue with creating URIs with paths that are not prefixed with '/'
|
230
264
|
|
231
|
-
# Addressable 1.0.3
|
265
|
+
# Addressable 1.0.3 <a name="v1.0.3">
|
232
266
|
- implemented a hash method
|
233
267
|
|
234
|
-
# Addressable 1.0.2
|
268
|
+
# Addressable 1.0.2 <a name="v1.0.2">
|
235
269
|
- fixed minor bug with the extract_mapping method
|
236
270
|
|
237
|
-
# Addressable 1.0.1
|
271
|
+
# Addressable 1.0.1 <a name="v1.0.1">
|
238
272
|
- fixed minor bug with the extract_mapping method
|
239
273
|
|
240
|
-
# Addressable 1.0.0
|
274
|
+
# Addressable 1.0.0 <a name="v1.0.0">
|
241
275
|
- heuristic parse method added
|
242
276
|
- parsing is slightly more strict
|
243
277
|
- replaced to_h with to_hash
|
@@ -246,16 +280,16 @@
|
|
246
280
|
- improved heckle rake task
|
247
281
|
- no surviving heckle mutations
|
248
282
|
|
249
|
-
# Addressable 0.1.2
|
283
|
+
# Addressable 0.1.2 <a name="v0.1.2">
|
250
284
|
- improved normalization
|
251
285
|
- fixed bug in joining algorithm
|
252
286
|
- updated specifications
|
253
287
|
|
254
|
-
# Addressable 0.1.1
|
288
|
+
# Addressable 0.1.1 <a name="v0.1.1">
|
255
289
|
- updated documentation
|
256
290
|
- added URI Template variable extraction
|
257
291
|
|
258
|
-
# Addressable 0.1.0
|
292
|
+
# Addressable 0.1.0 <a name="v0.1.0">
|
259
293
|
- initial release
|
260
294
|
- implementation based on RFC 3986, 3987
|
261
295
|
- support for IRIs via libidn
|
data/Rakefile
CHANGED
@@ -20,11 +20,17 @@ additionally provides extensive support for IRIs and URI templates.
|
|
20
20
|
TEXT
|
21
21
|
|
22
22
|
PKG_FILES = FileList[
|
23
|
-
"
|
24
|
-
"
|
25
|
-
"
|
26
|
-
|
27
|
-
|
23
|
+
"data/**/*",
|
24
|
+
"lib/**/*.rb",
|
25
|
+
"spec/**/*.rb",
|
26
|
+
"tasks/**/*.rake",
|
27
|
+
"addressable.gemspec",
|
28
|
+
"CHANGELOG.md",
|
29
|
+
"Gemfile",
|
30
|
+
"LICENSE.txt",
|
31
|
+
"README.md",
|
32
|
+
"Rakefile",
|
33
|
+
]
|
28
34
|
|
29
35
|
task :default => "spec"
|
30
36
|
|
data/addressable.gemspec
ADDED
@@ -0,0 +1,28 @@
|
|
1
|
+
# -*- encoding: utf-8 -*-
|
2
|
+
# stub: addressable 2.8.6 ruby lib
|
3
|
+
|
4
|
+
Gem::Specification.new do |s|
|
5
|
+
s.name = "addressable".freeze
|
6
|
+
s.version = "2.8.6".freeze
|
7
|
+
|
8
|
+
s.required_rubygems_version = Gem::Requirement.new(">= 0".freeze) if s.respond_to? :required_rubygems_version=
|
9
|
+
s.metadata = { "changelog_uri" => "https://github.com/sporkmonger/addressable/blob/main/CHANGELOG.md#v2.8.6" } if s.respond_to? :metadata=
|
10
|
+
s.require_paths = ["lib".freeze]
|
11
|
+
s.authors = ["Bob Aman".freeze]
|
12
|
+
s.date = "2023-12-09"
|
13
|
+
s.description = "Addressable is an alternative implementation to the URI implementation that is\npart of Ruby's standard library. It is flexible, offers heuristic parsing, and\nadditionally provides extensive support for IRIs and URI templates.\n".freeze
|
14
|
+
s.email = "bob@sporkmonger.com".freeze
|
15
|
+
s.extra_rdoc_files = ["README.md".freeze]
|
16
|
+
s.files = ["CHANGELOG.md".freeze, "Gemfile".freeze, "LICENSE.txt".freeze, "README.md".freeze, "Rakefile".freeze, "addressable.gemspec".freeze, "data/unicode.data".freeze, "lib/addressable.rb".freeze, "lib/addressable/idna.rb".freeze, "lib/addressable/idna/native.rb".freeze, "lib/addressable/idna/pure.rb".freeze, "lib/addressable/template.rb".freeze, "lib/addressable/uri.rb".freeze, "lib/addressable/version.rb".freeze, "spec/addressable/idna_spec.rb".freeze, "spec/addressable/net_http_compat_spec.rb".freeze, "spec/addressable/security_spec.rb".freeze, "spec/addressable/template_spec.rb".freeze, "spec/addressable/uri_spec.rb".freeze, "spec/spec_helper.rb".freeze, "tasks/clobber.rake".freeze, "tasks/gem.rake".freeze, "tasks/git.rake".freeze, "tasks/metrics.rake".freeze, "tasks/profile.rake".freeze, "tasks/rspec.rake".freeze, "tasks/yard.rake".freeze]
|
17
|
+
s.homepage = "https://github.com/sporkmonger/addressable".freeze
|
18
|
+
s.licenses = ["Apache-2.0".freeze]
|
19
|
+
s.rdoc_options = ["--main".freeze, "README.md".freeze]
|
20
|
+
s.required_ruby_version = Gem::Requirement.new(">= 2.2".freeze)
|
21
|
+
s.rubygems_version = "3.4.22".freeze
|
22
|
+
s.summary = "URI Implementation".freeze
|
23
|
+
|
24
|
+
s.specification_version = 4
|
25
|
+
|
26
|
+
s.add_runtime_dependency(%q<public_suffix>.freeze, [">= 2.0.2".freeze, "< 6.0".freeze])
|
27
|
+
s.add_development_dependency(%q<bundler>.freeze, [">= 1.0".freeze, "< 3.0".freeze])
|
28
|
+
end
|
@@ -29,8 +29,14 @@ module Addressable
|
|
29
29
|
IDN::Punycode.decode(value.to_s)
|
30
30
|
end
|
31
31
|
|
32
|
-
|
33
|
-
|
32
|
+
class << self
|
33
|
+
# @deprecated Use {String#unicode_normalize(:nfkc)} instead
|
34
|
+
def unicode_normalize_kc(value)
|
35
|
+
value.to_s.unicode_normalize(:nfkc)
|
36
|
+
end
|
37
|
+
|
38
|
+
extend Gem::Deprecate
|
39
|
+
deprecate :unicode_normalize_kc, "String#unicode_normalize(:nfkc)", 2023, 4
|
34
40
|
end
|
35
41
|
|
36
42
|
def self.to_ascii(value)
|
@@ -66,7 +66,7 @@ module Addressable
|
|
66
66
|
# domain name as described in RFC 3490.
|
67
67
|
def self.to_ascii(input)
|
68
68
|
input = input.to_s unless input.is_a?(String)
|
69
|
-
input = input.dup
|
69
|
+
input = input.dup.force_encoding(Encoding::UTF_8).unicode_normalize(:nfkc)
|
70
70
|
if input.respond_to?(:force_encoding)
|
71
71
|
input.force_encoding(Encoding::ASCII_8BIT)
|
72
72
|
end
|
@@ -77,7 +77,7 @@ module Addressable
|
|
77
77
|
part.force_encoding(Encoding::ASCII_8BIT)
|
78
78
|
end
|
79
79
|
if part =~ UTF8_REGEX && part =~ UTF8_REGEX_MULTIBYTE
|
80
|
-
ACE_PREFIX + punycode_encode(
|
80
|
+
ACE_PREFIX + punycode_encode(part)
|
81
81
|
else
|
82
82
|
part
|
83
83
|
end
|
@@ -112,13 +112,14 @@ module Addressable
|
|
112
112
|
output
|
113
113
|
end
|
114
114
|
|
115
|
-
|
116
|
-
|
117
|
-
|
118
|
-
|
119
|
-
|
120
|
-
|
121
|
-
|
115
|
+
class << self
|
116
|
+
# @deprecated Use {String#unicode_normalize(:nfkc)} instead
|
117
|
+
def unicode_normalize_kc(value)
|
118
|
+
value.to_s.unicode_normalize(:nfkc)
|
119
|
+
end
|
120
|
+
|
121
|
+
extend Gem::Deprecate
|
122
|
+
deprecate :unicode_normalize_kc, "String#unicode_normalize(:nfkc)", 2023, 4
|
122
123
|
end
|
123
124
|
|
124
125
|
##
|
@@ -136,164 +137,6 @@ module Addressable
|
|
136
137
|
end
|
137
138
|
private_class_method :unicode_downcase
|
138
139
|
|
139
|
-
def self.unicode_compose(unpacked)
|
140
|
-
unpacked_result = []
|
141
|
-
length = unpacked.length
|
142
|
-
|
143
|
-
return unpacked if length == 0
|
144
|
-
|
145
|
-
starter = unpacked[0]
|
146
|
-
starter_cc = lookup_unicode_combining_class(starter)
|
147
|
-
starter_cc = 256 if starter_cc != 0
|
148
|
-
for i in 1...length
|
149
|
-
ch = unpacked[i]
|
150
|
-
|
151
|
-
if (starter_cc == 0 &&
|
152
|
-
(composite = unicode_compose_pair(starter, ch)) != nil)
|
153
|
-
starter = composite
|
154
|
-
else
|
155
|
-
unpacked_result << starter
|
156
|
-
starter = ch
|
157
|
-
end
|
158
|
-
end
|
159
|
-
unpacked_result << starter
|
160
|
-
return unpacked_result
|
161
|
-
end
|
162
|
-
private_class_method :unicode_compose
|
163
|
-
|
164
|
-
def self.unicode_compose_pair(ch_one, ch_two)
|
165
|
-
if ch_one >= HANGUL_LBASE && ch_one < HANGUL_LBASE + HANGUL_LCOUNT &&
|
166
|
-
ch_two >= HANGUL_VBASE && ch_two < HANGUL_VBASE + HANGUL_VCOUNT
|
167
|
-
# Hangul L + V
|
168
|
-
return HANGUL_SBASE + (
|
169
|
-
(ch_one - HANGUL_LBASE) * HANGUL_VCOUNT + (ch_two - HANGUL_VBASE)
|
170
|
-
) * HANGUL_TCOUNT
|
171
|
-
elsif ch_one >= HANGUL_SBASE &&
|
172
|
-
ch_one < HANGUL_SBASE + HANGUL_SCOUNT &&
|
173
|
-
(ch_one - HANGUL_SBASE) % HANGUL_TCOUNT == 0 &&
|
174
|
-
ch_two >= HANGUL_TBASE && ch_two < HANGUL_TBASE + HANGUL_TCOUNT
|
175
|
-
# Hangul LV + T
|
176
|
-
return ch_one + (ch_two - HANGUL_TBASE)
|
177
|
-
end
|
178
|
-
|
179
|
-
p = []
|
180
|
-
|
181
|
-
ucs4_to_utf8(ch_one, p)
|
182
|
-
ucs4_to_utf8(ch_two, p)
|
183
|
-
|
184
|
-
return lookup_unicode_composition(p)
|
185
|
-
end
|
186
|
-
private_class_method :unicode_compose_pair
|
187
|
-
|
188
|
-
def self.ucs4_to_utf8(char, buffer)
|
189
|
-
if char < 128
|
190
|
-
buffer << char
|
191
|
-
elsif char < 2048
|
192
|
-
buffer << (char >> 6 | 192)
|
193
|
-
buffer << (char & 63 | 128)
|
194
|
-
elsif char < 0x10000
|
195
|
-
buffer << (char >> 12 | 224)
|
196
|
-
buffer << (char >> 6 & 63 | 128)
|
197
|
-
buffer << (char & 63 | 128)
|
198
|
-
elsif char < 0x200000
|
199
|
-
buffer << (char >> 18 | 240)
|
200
|
-
buffer << (char >> 12 & 63 | 128)
|
201
|
-
buffer << (char >> 6 & 63 | 128)
|
202
|
-
buffer << (char & 63 | 128)
|
203
|
-
elsif char < 0x4000000
|
204
|
-
buffer << (char >> 24 | 248)
|
205
|
-
buffer << (char >> 18 & 63 | 128)
|
206
|
-
buffer << (char >> 12 & 63 | 128)
|
207
|
-
buffer << (char >> 6 & 63 | 128)
|
208
|
-
buffer << (char & 63 | 128)
|
209
|
-
elsif char < 0x80000000
|
210
|
-
buffer << (char >> 30 | 252)
|
211
|
-
buffer << (char >> 24 & 63 | 128)
|
212
|
-
buffer << (char >> 18 & 63 | 128)
|
213
|
-
buffer << (char >> 12 & 63 | 128)
|
214
|
-
buffer << (char >> 6 & 63 | 128)
|
215
|
-
buffer << (char & 63 | 128)
|
216
|
-
end
|
217
|
-
end
|
218
|
-
private_class_method :ucs4_to_utf8
|
219
|
-
|
220
|
-
def self.unicode_sort_canonical(unpacked)
|
221
|
-
unpacked = unpacked.dup
|
222
|
-
i = 1
|
223
|
-
length = unpacked.length
|
224
|
-
|
225
|
-
return unpacked if length < 2
|
226
|
-
|
227
|
-
while i < length
|
228
|
-
last = unpacked[i-1]
|
229
|
-
ch = unpacked[i]
|
230
|
-
last_cc = lookup_unicode_combining_class(last)
|
231
|
-
cc = lookup_unicode_combining_class(ch)
|
232
|
-
if cc != 0 && last_cc != 0 && last_cc > cc
|
233
|
-
unpacked[i] = last
|
234
|
-
unpacked[i-1] = ch
|
235
|
-
i -= 1 if i > 1
|
236
|
-
else
|
237
|
-
i += 1
|
238
|
-
end
|
239
|
-
end
|
240
|
-
return unpacked
|
241
|
-
end
|
242
|
-
private_class_method :unicode_sort_canonical
|
243
|
-
|
244
|
-
def self.unicode_decompose(unpacked)
|
245
|
-
unpacked_result = []
|
246
|
-
for cp in unpacked
|
247
|
-
if cp >= HANGUL_SBASE && cp < HANGUL_SBASE + HANGUL_SCOUNT
|
248
|
-
l, v, t = unicode_decompose_hangul(cp)
|
249
|
-
unpacked_result << l
|
250
|
-
unpacked_result << v if v
|
251
|
-
unpacked_result << t if t
|
252
|
-
else
|
253
|
-
dc = lookup_unicode_compatibility(cp)
|
254
|
-
unless dc
|
255
|
-
unpacked_result << cp
|
256
|
-
else
|
257
|
-
unpacked_result.concat(unicode_decompose(dc.unpack("U*")))
|
258
|
-
end
|
259
|
-
end
|
260
|
-
end
|
261
|
-
return unpacked_result
|
262
|
-
end
|
263
|
-
private_class_method :unicode_decompose
|
264
|
-
|
265
|
-
def self.unicode_decompose_hangul(codepoint)
|
266
|
-
sindex = codepoint - HANGUL_SBASE;
|
267
|
-
if sindex < 0 || sindex >= HANGUL_SCOUNT
|
268
|
-
l = codepoint
|
269
|
-
v = t = nil
|
270
|
-
return l, v, t
|
271
|
-
end
|
272
|
-
l = HANGUL_LBASE + sindex / HANGUL_NCOUNT
|
273
|
-
v = HANGUL_VBASE + (sindex % HANGUL_NCOUNT) / HANGUL_TCOUNT
|
274
|
-
t = HANGUL_TBASE + sindex % HANGUL_TCOUNT
|
275
|
-
if t == HANGUL_TBASE
|
276
|
-
t = nil
|
277
|
-
end
|
278
|
-
return l, v, t
|
279
|
-
end
|
280
|
-
private_class_method :unicode_decompose_hangul
|
281
|
-
|
282
|
-
def self.lookup_unicode_combining_class(codepoint)
|
283
|
-
codepoint_data = UNICODE_DATA[codepoint]
|
284
|
-
(codepoint_data ?
|
285
|
-
(codepoint_data[UNICODE_DATA_COMBINING_CLASS] || 0) :
|
286
|
-
0)
|
287
|
-
end
|
288
|
-
private_class_method :lookup_unicode_combining_class
|
289
|
-
|
290
|
-
def self.lookup_unicode_compatibility(codepoint)
|
291
|
-
codepoint_data = UNICODE_DATA[codepoint]
|
292
|
-
(codepoint_data ?
|
293
|
-
codepoint_data[UNICODE_DATA_COMPATIBILITY] : nil)
|
294
|
-
end
|
295
|
-
private_class_method :lookup_unicode_compatibility
|
296
|
-
|
297
140
|
def self.lookup_unicode_lowercase(codepoint)
|
298
141
|
codepoint_data = UNICODE_DATA[codepoint]
|
299
142
|
(codepoint_data ?
|
@@ -302,21 +145,6 @@ module Addressable
|
|
302
145
|
end
|
303
146
|
private_class_method :lookup_unicode_lowercase
|
304
147
|
|
305
|
-
def self.lookup_unicode_composition(unpacked)
|
306
|
-
return COMPOSITION_TABLE[unpacked]
|
307
|
-
end
|
308
|
-
private_class_method :lookup_unicode_composition
|
309
|
-
|
310
|
-
HANGUL_SBASE = 0xac00
|
311
|
-
HANGUL_LBASE = 0x1100
|
312
|
-
HANGUL_LCOUNT = 19
|
313
|
-
HANGUL_VBASE = 0x1161
|
314
|
-
HANGUL_VCOUNT = 21
|
315
|
-
HANGUL_TBASE = 0x11a7
|
316
|
-
HANGUL_TCOUNT = 28
|
317
|
-
HANGUL_NCOUNT = HANGUL_VCOUNT * HANGUL_TCOUNT # 588
|
318
|
-
HANGUL_SCOUNT = HANGUL_LCOUNT * HANGUL_NCOUNT # 11172
|
319
|
-
|
320
148
|
UNICODE_DATA_COMBINING_CLASS = 0
|
321
149
|
UNICODE_DATA_EXCLUSION = 1
|
322
150
|
UNICODE_DATA_CANONICAL = 2
|
data/lib/addressable/template.rb
CHANGED
@@ -892,25 +892,24 @@ module Addressable
|
|
892
892
|
# operator.
|
893
893
|
#
|
894
894
|
# @param [Hash, Array, String] value
|
895
|
-
# Normalizes keys and values with
|
895
|
+
# Normalizes unicode keys and values with String#unicode_normalize (NFC)
|
896
896
|
#
|
897
897
|
# @return [Hash, Array, String] The normalized values
|
898
898
|
def normalize_value(value)
|
899
|
-
unless value.is_a?(Hash)
|
900
|
-
value = value.respond_to?(:to_ary) ? value.to_ary : value.to_str
|
901
|
-
end
|
902
|
-
|
903
899
|
# Handle unicode normalization
|
904
|
-
if value.
|
905
|
-
value.map! { |val|
|
900
|
+
if value.respond_to?(:to_ary)
|
901
|
+
value.to_ary.map! { |val| normalize_value(val) }
|
906
902
|
elsif value.kind_of?(Hash)
|
907
903
|
value = value.inject({}) { |acc, (k, v)|
|
908
|
-
acc[
|
909
|
-
Addressable::IDNA.unicode_normalize_kc(v)
|
904
|
+
acc[normalize_value(k)] = normalize_value(v)
|
910
905
|
acc
|
911
906
|
}
|
912
907
|
else
|
913
|
-
value =
|
908
|
+
value = value.to_s if !value.kind_of?(String)
|
909
|
+
if value.encoding != Encoding::UTF_8
|
910
|
+
value = value.dup.force_encoding(Encoding::UTF_8)
|
911
|
+
end
|
912
|
+
value = value.unicode_normalize(:nfc)
|
914
913
|
end
|
915
914
|
value
|
916
915
|
end
|