yasuri 3.3.0 → 3.3.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.github/workflows/ruby.yml +1 -1
- data/.rubocop.yml +49 -0
- data/.rubocop_todo.yml +0 -0
- data/README.md +4 -2
- data/Rakefile +1 -1
- data/examples/example.rb +0 -1
- data/lib/yasuri/version.rb +1 -1
- data/lib/yasuri/yasuri.rb +68 -61
- data/lib/yasuri/yasuri_cli.rb +49 -35
- data/lib/yasuri/yasuri_links_node.rb +3 -5
- data/lib/yasuri/yasuri_map_node.rb +2 -2
- data/lib/yasuri/yasuri_node.rb +5 -8
- data/lib/yasuri/yasuri_node_generator.rb +1 -3
- data/lib/yasuri/yasuri_paginate_node.rb +22 -18
- data/lib/yasuri/yasuri_struct_node.rb +1 -3
- data/lib/yasuri/yasuri_text_node.rb +4 -6
- data/spec/servers/httpserver.rb +0 -2
- data/spec/spec_helper.rb +0 -2
- data/spec/yasuri_cli_spec.rb +79 -61
- data/spec/yasuri_links_node_spec.rb +72 -62
- data/spec/yasuri_map_spec.rb +10 -14
- data/spec/yasuri_paginate_node_spec.rb +89 -90
- data/spec/yasuri_spec.rb +15 -24
- data/spec/yasuri_struct_node_spec.rb +120 -96
- data/spec/yasuri_text_node_spec.rb +22 -31
- data/yasuri.gemspec +29 -24
- metadata +67 -11
- data/spec/yasuri_node_spec.rb +0 -11
@@ -1,77 +1,68 @@
|
|
1
1
|
|
2
|
-
# Author:: TAC (tac@tac42.net)
|
3
2
|
|
4
3
|
require_relative 'spec_helper'
|
5
4
|
|
6
|
-
########
|
7
|
-
# Text #
|
8
|
-
########
|
9
5
|
describe 'Yasuri' do
|
10
6
|
include_context 'httpserver'
|
11
7
|
|
12
|
-
before do
|
13
|
-
@uri = uri
|
14
|
-
end
|
15
|
-
|
16
8
|
describe '::TextNode' do
|
17
|
-
before { @node = Yasuri::TextNode.new('/html/body/p[1]', "title") }
|
18
|
-
|
19
9
|
it 'scrape text text <p>Hello,Yasuri</p>' do
|
20
|
-
|
10
|
+
node = Yasuri::TextNode.new('/html/body/p[1]', "title")
|
11
|
+
actual = node.scrape(uri)
|
21
12
|
expect(actual).to eq "Hello,Yasuri"
|
22
13
|
end
|
23
14
|
|
24
15
|
it 'return empty text if no match node' do
|
25
16
|
no_match_node = Yasuri::TextNode.new('/html/body/no_match_node', "title")
|
26
|
-
actual = no_match_node.scrape(
|
17
|
+
actual = no_match_node.scrape(uri)
|
27
18
|
expect(actual).to be_empty
|
28
19
|
end
|
29
20
|
|
30
21
|
it 'fail with invalid xpath' do
|
31
22
|
invalid_xpath = '/html/body/no_match_node['
|
32
23
|
node = Yasuri::TextNode.new(invalid_xpath, "title")
|
33
|
-
expect { node.scrape(
|
24
|
+
expect { node.scrape(uri) }.to raise_error(Nokogiri::XML::XPath::SyntaxError)
|
34
25
|
end
|
35
26
|
|
36
27
|
it "can be defined by DSL, return single TextNode title" do
|
37
28
|
generated = Yasuri.text_title '/html/body/p[1]'
|
38
|
-
original
|
39
|
-
compare_generated_vs_original(generated, original,
|
29
|
+
original = Yasuri::TextNode.new('/html/body/p[1]', "title")
|
30
|
+
compare_generated_vs_original(generated, original, uri)
|
40
31
|
end
|
41
32
|
|
42
|
-
it "can
|
43
|
-
node
|
44
|
-
actual = node.scrape(
|
33
|
+
it "can truncate head by regexp" do
|
34
|
+
node = Yasuri.text_title '/html/body/p[1]', truncate: /^[^,]+/
|
35
|
+
actual = node.scrape(uri)
|
45
36
|
expect(actual).to eq "Hello"
|
46
37
|
end
|
47
38
|
|
48
|
-
it "
|
49
|
-
node
|
50
|
-
actual = node.scrape(
|
51
|
-
expect(actual).to eq "
|
39
|
+
it "can truncate tail by regexp" do
|
40
|
+
node = Yasuri.text_title '/html/body/p[1]', truncate: /[^,]+$/
|
41
|
+
actual = node.scrape(uri)
|
42
|
+
expect(actual).to eq "Yasuri"
|
52
43
|
end
|
53
44
|
|
54
|
-
it "
|
55
|
-
node = Yasuri.text_title '/html/body/p[1]', truncate
|
56
|
-
actual = node.scrape(
|
57
|
-
expect(actual).to eq "
|
45
|
+
it "return first captured if matched given capture pattern" do
|
46
|
+
node = Yasuri.text_title '/html/body/p[1]', truncate: /H(.+)i/
|
47
|
+
actual = node.scrape(uri)
|
48
|
+
expect(actual).to eq "ello,Yasur"
|
58
49
|
end
|
59
50
|
|
60
51
|
it "return empty string if truncated with no match to regexp" do
|
61
|
-
node = Yasuri.text_title '/html/body/p[1]', truncate
|
62
|
-
actual = node.scrape(
|
52
|
+
node = Yasuri.text_title '/html/body/p[1]', truncate: /^hoge/
|
53
|
+
actual = node.scrape(uri)
|
63
54
|
expect(actual).to be_empty
|
64
55
|
end
|
65
56
|
|
66
57
|
it "return symbol method applied string" do
|
67
58
|
node = Yasuri.text_title '/html/body/p[1]', proc: :upcase
|
68
|
-
actual = node.scrape(
|
59
|
+
actual = node.scrape(uri)
|
69
60
|
expect(actual).to eq "HELLO,YASURI"
|
70
61
|
end
|
71
62
|
|
72
63
|
it "return apply multi arguments" do
|
73
|
-
node = Yasuri.text_title '/html/body/p[1]', proc: :upcase, truncate
|
74
|
-
actual = node.scrape(
|
64
|
+
node = Yasuri.text_title '/html/body/p[1]', proc: :upcase, truncate: /H(.+)i/
|
65
|
+
actual = node.scrape(uri)
|
75
66
|
expect(actual).to eq "ELLO,YASUR"
|
76
67
|
end
|
77
68
|
end
|
data/yasuri.gemspec
CHANGED
@@ -1,33 +1,38 @@
|
|
1
|
-
|
1
|
+
|
2
2
|
lib = File.expand_path('../lib', __FILE__)
|
3
3
|
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
4
|
require 'yasuri/version'
|
5
5
|
|
6
6
|
Gem::Specification.new do |spec|
|
7
|
-
spec.name
|
8
|
-
spec.version
|
9
|
-
spec.authors
|
10
|
-
spec.email
|
11
|
-
spec.summary
|
12
|
-
spec.description
|
13
|
-
spec.homepage
|
14
|
-
spec.license
|
7
|
+
spec.name = 'yasuri'
|
8
|
+
spec.version = Yasuri::VERSION
|
9
|
+
spec.authors = ['TAC']
|
10
|
+
spec.email = ['tac@tac42.net']
|
11
|
+
spec.summary = %q{Yasuri is easy scraping library.}
|
12
|
+
spec.description = %q{Yasuri is an easy web-scraping library for supporting 'Mechanize'.}
|
13
|
+
spec.homepage = 'https://github.com/tac0x2a/yasuri'
|
14
|
+
spec.license = 'MIT'
|
15
15
|
|
16
|
-
spec.files
|
17
|
-
spec.executables
|
18
|
-
spec.bindir
|
19
|
-
spec.test_files
|
20
|
-
spec.require_paths
|
16
|
+
spec.files = `git ls-files -z`.split("\x0")
|
17
|
+
spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
18
|
+
spec.bindir = 'exe'
|
19
|
+
spec.test_files = spec.files.grep(%r{^(test|spec|features)/})
|
20
|
+
spec.require_paths = ['lib']
|
21
|
+
spec.required_ruby_version = '>= 2.7.0'
|
21
22
|
|
22
|
-
spec.add_development_dependency
|
23
|
-
spec.add_development_dependency
|
24
|
-
spec.add_development_dependency
|
25
|
-
spec.add_development_dependency
|
26
|
-
spec.add_development_dependency
|
27
|
-
spec.add_development_dependency
|
28
|
-
spec.add_development_dependency
|
29
|
-
spec.add_development_dependency
|
23
|
+
spec.add_development_dependency 'bundler'
|
24
|
+
spec.add_development_dependency 'codeclimate-test-reporter'
|
25
|
+
spec.add_development_dependency 'coveralls'
|
26
|
+
spec.add_development_dependency 'fuubar'
|
27
|
+
spec.add_development_dependency 'glint'
|
28
|
+
spec.add_development_dependency 'rake'
|
29
|
+
spec.add_development_dependency 'rspec'
|
30
|
+
spec.add_development_dependency 'rubocop'
|
31
|
+
spec.add_development_dependency 'rubocop-performance'
|
32
|
+
spec.add_development_dependency 'rubocop-rspec'
|
33
|
+
spec.add_development_dependency 'rubocop-rubycw'
|
34
|
+
spec.add_development_dependency 'simplecov'
|
30
35
|
|
31
|
-
spec.add_dependency
|
32
|
-
spec.add_dependency
|
36
|
+
spec.add_dependency 'mechanize'
|
37
|
+
spec.add_dependency 'thor'
|
33
38
|
end
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: yasuri
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 3.3.
|
4
|
+
version: 3.3.1
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- TAC
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2021-03-
|
11
|
+
date: 2021-03-31 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|
@@ -25,7 +25,7 @@ dependencies:
|
|
25
25
|
- !ruby/object:Gem::Version
|
26
26
|
version: '0'
|
27
27
|
- !ruby/object:Gem::Dependency
|
28
|
-
name:
|
28
|
+
name: codeclimate-test-reporter
|
29
29
|
requirement: !ruby/object:Gem::Requirement
|
30
30
|
requirements:
|
31
31
|
- - ">="
|
@@ -39,7 +39,7 @@ dependencies:
|
|
39
39
|
- !ruby/object:Gem::Version
|
40
40
|
version: '0'
|
41
41
|
- !ruby/object:Gem::Dependency
|
42
|
-
name:
|
42
|
+
name: coveralls
|
43
43
|
requirement: !ruby/object:Gem::Requirement
|
44
44
|
requirements:
|
45
45
|
- - ">="
|
@@ -81,7 +81,7 @@ dependencies:
|
|
81
81
|
- !ruby/object:Gem::Version
|
82
82
|
version: '0'
|
83
83
|
- !ruby/object:Gem::Dependency
|
84
|
-
name:
|
84
|
+
name: rake
|
85
85
|
requirement: !ruby/object:Gem::Requirement
|
86
86
|
requirements:
|
87
87
|
- - ">="
|
@@ -95,7 +95,7 @@ dependencies:
|
|
95
95
|
- !ruby/object:Gem::Version
|
96
96
|
version: '0'
|
97
97
|
- !ruby/object:Gem::Dependency
|
98
|
-
name:
|
98
|
+
name: rspec
|
99
99
|
requirement: !ruby/object:Gem::Requirement
|
100
100
|
requirements:
|
101
101
|
- - ">="
|
@@ -109,7 +109,63 @@ dependencies:
|
|
109
109
|
- !ruby/object:Gem::Version
|
110
110
|
version: '0'
|
111
111
|
- !ruby/object:Gem::Dependency
|
112
|
-
name:
|
112
|
+
name: rubocop
|
113
|
+
requirement: !ruby/object:Gem::Requirement
|
114
|
+
requirements:
|
115
|
+
- - ">="
|
116
|
+
- !ruby/object:Gem::Version
|
117
|
+
version: '0'
|
118
|
+
type: :development
|
119
|
+
prerelease: false
|
120
|
+
version_requirements: !ruby/object:Gem::Requirement
|
121
|
+
requirements:
|
122
|
+
- - ">="
|
123
|
+
- !ruby/object:Gem::Version
|
124
|
+
version: '0'
|
125
|
+
- !ruby/object:Gem::Dependency
|
126
|
+
name: rubocop-performance
|
127
|
+
requirement: !ruby/object:Gem::Requirement
|
128
|
+
requirements:
|
129
|
+
- - ">="
|
130
|
+
- !ruby/object:Gem::Version
|
131
|
+
version: '0'
|
132
|
+
type: :development
|
133
|
+
prerelease: false
|
134
|
+
version_requirements: !ruby/object:Gem::Requirement
|
135
|
+
requirements:
|
136
|
+
- - ">="
|
137
|
+
- !ruby/object:Gem::Version
|
138
|
+
version: '0'
|
139
|
+
- !ruby/object:Gem::Dependency
|
140
|
+
name: rubocop-rspec
|
141
|
+
requirement: !ruby/object:Gem::Requirement
|
142
|
+
requirements:
|
143
|
+
- - ">="
|
144
|
+
- !ruby/object:Gem::Version
|
145
|
+
version: '0'
|
146
|
+
type: :development
|
147
|
+
prerelease: false
|
148
|
+
version_requirements: !ruby/object:Gem::Requirement
|
149
|
+
requirements:
|
150
|
+
- - ">="
|
151
|
+
- !ruby/object:Gem::Version
|
152
|
+
version: '0'
|
153
|
+
- !ruby/object:Gem::Dependency
|
154
|
+
name: rubocop-rubycw
|
155
|
+
requirement: !ruby/object:Gem::Requirement
|
156
|
+
requirements:
|
157
|
+
- - ">="
|
158
|
+
- !ruby/object:Gem::Version
|
159
|
+
version: '0'
|
160
|
+
type: :development
|
161
|
+
prerelease: false
|
162
|
+
version_requirements: !ruby/object:Gem::Requirement
|
163
|
+
requirements:
|
164
|
+
- - ">="
|
165
|
+
- !ruby/object:Gem::Version
|
166
|
+
version: '0'
|
167
|
+
- !ruby/object:Gem::Dependency
|
168
|
+
name: simplecov
|
113
169
|
requirement: !ruby/object:Gem::Requirement
|
114
170
|
requirements:
|
115
171
|
- - ">="
|
@@ -150,7 +206,7 @@ dependencies:
|
|
150
206
|
- - ">="
|
151
207
|
- !ruby/object:Gem::Version
|
152
208
|
version: '0'
|
153
|
-
description: Yasuri is an easy web-scraping library for supporting
|
209
|
+
description: Yasuri is an easy web-scraping library for supporting 'Mechanize'.
|
154
210
|
email:
|
155
211
|
- tac@tac42.net
|
156
212
|
executables:
|
@@ -162,6 +218,8 @@ files:
|
|
162
218
|
- ".github/workflows/ruby.yml"
|
163
219
|
- ".gitignore"
|
164
220
|
- ".rspec"
|
221
|
+
- ".rubocop.yml"
|
222
|
+
- ".rubocop_todo.yml"
|
165
223
|
- ".ruby-version"
|
166
224
|
- ".travis.yml"
|
167
225
|
- Gemfile
|
@@ -208,7 +266,6 @@ files:
|
|
208
266
|
- spec/yasuri_cli_spec.rb
|
209
267
|
- spec/yasuri_links_node_spec.rb
|
210
268
|
- spec/yasuri_map_spec.rb
|
211
|
-
- spec/yasuri_node_spec.rb
|
212
269
|
- spec/yasuri_paginate_node_spec.rb
|
213
270
|
- spec/yasuri_spec.rb
|
214
271
|
- spec/yasuri_struct_node_spec.rb
|
@@ -226,7 +283,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
|
|
226
283
|
requirements:
|
227
284
|
- - ">="
|
228
285
|
- !ruby/object:Gem::Version
|
229
|
-
version:
|
286
|
+
version: 2.7.0
|
230
287
|
required_rubygems_version: !ruby/object:Gem::Requirement
|
231
288
|
requirements:
|
232
289
|
- - ">="
|
@@ -260,7 +317,6 @@ test_files:
|
|
260
317
|
- spec/yasuri_cli_spec.rb
|
261
318
|
- spec/yasuri_links_node_spec.rb
|
262
319
|
- spec/yasuri_map_spec.rb
|
263
|
-
- spec/yasuri_node_spec.rb
|
264
320
|
- spec/yasuri_paginate_node_spec.rb
|
265
321
|
- spec/yasuri_spec.rb
|
266
322
|
- spec/yasuri_struct_node_spec.rb
|