tomosia_amanaplus_crawl 0.1.4 → 0.1.9
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/Gemfile +0 -4
- data/Gemfile.lock +52 -0
- data/README.md +2 -5
- data/Rakefile +6 -0
- data/exe/tomosia_amanaplus_crawl +5 -0
- data/lib/tomosia_amanaplus_crawl.rb +18 -8
- data/lib/tomosia_amanaplus_crawl/cli.rb +14 -0
- data/lib/tomosia_amanaplus_crawl/version.rb +1 -1
- data/spec/spec_helper.rb +14 -0
- data/spec/tomosia_amanaplus_crawl_spec.rb +9 -0
- data/tomosia_amanaplus_crawl.gemspec +8 -4
- metadata +67 -5
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: a3d62cab8ce5e9076b05594851cde7da03d627b8c386dfcbc57e64d561e7569a
|
4
|
+
data.tar.gz: 7ce113e8fde42d24405c662998c07cc6992a7d7e7ce94816590ef0a3f634ce0f
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: c4841b8f6948aa3a2afc7456380d26b374157ce5a453d9eb0bab6141ff5fb71b88cd3e6f51e0556cfb43421cef12c572a17a4a05f5d4d00404d19b8af452d079
|
7
|
+
data.tar.gz: 407ea07476e5fcbc4f356f540d975a4d15d36aab8ae02103e906575dc1153c257b2ada9f1d9309f39d9557e9109bfe464eae94680715143259af5b41ec6c4d74
|
data/Gemfile
CHANGED
data/Gemfile.lock
ADDED
@@ -0,0 +1,52 @@
|
|
1
|
+
PATH
|
2
|
+
remote: .
|
3
|
+
specs:
|
4
|
+
tomosia_amanaplus_crawl (0.1.8)
|
5
|
+
httparty (= 0.18.1)
|
6
|
+
nokogiri (= 1.10.10)
|
7
|
+
spreadsheet (= 1.2.6)
|
8
|
+
thor
|
9
|
+
|
10
|
+
GEM
|
11
|
+
remote: https://rubygems.org/
|
12
|
+
specs:
|
13
|
+
diff-lcs (1.4.4)
|
14
|
+
httparty (0.18.1)
|
15
|
+
mime-types (~> 3.0)
|
16
|
+
multi_xml (>= 0.5.2)
|
17
|
+
mime-types (3.3.1)
|
18
|
+
mime-types-data (~> 3.2015)
|
19
|
+
mime-types-data (3.2020.0512)
|
20
|
+
mini_portile2 (2.4.0)
|
21
|
+
multi_xml (0.6.0)
|
22
|
+
nokogiri (1.10.10)
|
23
|
+
mini_portile2 (~> 2.4.0)
|
24
|
+
rake (12.3.3)
|
25
|
+
rspec (3.9.0)
|
26
|
+
rspec-core (~> 3.9.0)
|
27
|
+
rspec-expectations (~> 3.9.0)
|
28
|
+
rspec-mocks (~> 3.9.0)
|
29
|
+
rspec-core (3.9.2)
|
30
|
+
rspec-support (~> 3.9.3)
|
31
|
+
rspec-expectations (3.9.2)
|
32
|
+
diff-lcs (>= 1.2.0, < 2.0)
|
33
|
+
rspec-support (~> 3.9.0)
|
34
|
+
rspec-mocks (3.9.1)
|
35
|
+
diff-lcs (>= 1.2.0, < 2.0)
|
36
|
+
rspec-support (~> 3.9.0)
|
37
|
+
rspec-support (3.9.3)
|
38
|
+
ruby-ole (1.2.12.2)
|
39
|
+
spreadsheet (1.2.6)
|
40
|
+
ruby-ole (>= 1.0)
|
41
|
+
thor (1.0.1)
|
42
|
+
|
43
|
+
PLATFORMS
|
44
|
+
ruby
|
45
|
+
|
46
|
+
DEPENDENCIES
|
47
|
+
rake (~> 12.0)
|
48
|
+
rspec (~> 3.0)
|
49
|
+
tomosia_amanaplus_crawl!
|
50
|
+
|
51
|
+
BUNDLED WITH
|
52
|
+
2.1.4
|
data/README.md
CHANGED
@@ -10,9 +10,6 @@ Add this line to your application's Gemfile:
|
|
10
10
|
|
11
11
|
```ruby
|
12
12
|
gem 'tomosia_amanaplus_crawl'
|
13
|
-
gem 'httparty'
|
14
|
-
gem 'nokogiri'
|
15
|
-
gem 'spreadsheet'
|
16
13
|
```
|
17
14
|
|
18
15
|
And then execute:
|
@@ -26,9 +23,9 @@ Or install it yourself as:
|
|
26
23
|
## Usage
|
27
24
|
|
28
25
|
```ruby
|
29
|
-
|
30
|
-
TomosiaAmanaplusCrawl::Crawler.new.run(keyword, path, max)
|
26
|
+
tomosia_amanaplus_crawl crawl "keyword" --destination "./lib" --max=123
|
31
27
|
```
|
28
|
+
Example: tomosia_amanaplus_crawl crawl "hoian" --destination "./" --max=123
|
32
29
|
keyword: hoian, danang, ...
|
33
30
|
path: './', '/desktop/', ...
|
34
31
|
max: số lượng ảnh muốn lấy về. Nếu max lớn hơn tổng số ảnh các page thì vẫn lấy hết tất cả ảnh
|
data/Rakefile
CHANGED
@@ -8,7 +8,7 @@ module TomosiaAmanaplusCrawl
|
|
8
8
|
require 'spreadsheet'
|
9
9
|
|
10
10
|
class Crawler
|
11
|
-
URL = "https://plus.amanaimages.com/items/search
|
11
|
+
URL = "https://plus.amanaimages.com/items/search"
|
12
12
|
|
13
13
|
def run(keyword, destination, max)
|
14
14
|
unparsed_page = HTTParty.get("#{URL}/#{keyword}")
|
@@ -19,7 +19,7 @@ module TomosiaAmanaplusCrawl
|
|
19
19
|
|
20
20
|
# lấy tổng số image
|
21
21
|
total = parsed_page.css("h1.p-search-result__ttl").text.split(' ').first
|
22
|
-
total = total[
|
22
|
+
total = total[(6 + keyword.length)..(total.length - 1)].chop.chop.chop.gsub(',', '').to_i
|
23
23
|
if max > total # nếu max lớn hơn total thì max = total => vẫn lấy hết
|
24
24
|
max = total
|
25
25
|
end
|
@@ -70,13 +70,23 @@ module TomosiaAmanaplusCrawl
|
|
70
70
|
print "\nDownloading"
|
71
71
|
images.each do |curr_image|
|
72
72
|
threads << Thread.new(curr_image) {
|
73
|
-
|
74
|
-
|
75
|
-
|
76
|
-
curr_image[:
|
77
|
-
|
73
|
+
timeout = 0
|
74
|
+
begin
|
75
|
+
open(curr_image[:url]) do |image|
|
76
|
+
File.open("#{path}/#{curr_image[:url].split('/').last}", "a+") do |file|
|
77
|
+
file.write(image.read) # lưu hình ảnh
|
78
|
+
curr_image[:size] = image.size # cập nhật lại size trong mảng images
|
79
|
+
print "."
|
80
|
+
end
|
81
|
+
end # end open
|
82
|
+
rescue => exception
|
83
|
+
if timeout < 3
|
84
|
+
timeout += 1
|
85
|
+
retry
|
86
|
+
else
|
87
|
+
next
|
78
88
|
end
|
79
|
-
end
|
89
|
+
end
|
80
90
|
}
|
81
91
|
end
|
82
92
|
threads.each { |t| t.join }
|
@@ -0,0 +1,14 @@
|
|
1
|
+
require 'thor'
|
2
|
+
require_relative '../tomosia_amanaplus_crawl'
|
3
|
+
|
4
|
+
module TomosiaAmanaplusCrawl
|
5
|
+
class Cli < Thor
|
6
|
+
|
7
|
+
desc "crawl KEYWORD", "enter KEYWORD to search"
|
8
|
+
option :destination
|
9
|
+
option :max
|
10
|
+
def crawl(keyword)
|
11
|
+
TomosiaAmanaplusCrawl::Crawler.new.run(keyword, options[:destination], options[:max].to_i)
|
12
|
+
end
|
13
|
+
end
|
14
|
+
end
|
data/spec/spec_helper.rb
ADDED
@@ -0,0 +1,14 @@
|
|
1
|
+
require "bundler/setup"
|
2
|
+
require "tomosia_amanaplus_crawl"
|
3
|
+
|
4
|
+
RSpec.configure do |config|
|
5
|
+
# Enable flags like --only-failures and --next-failure
|
6
|
+
config.example_status_persistence_file_path = ".rspec_status"
|
7
|
+
|
8
|
+
# Disable RSpec exposing methods globally on `Module` and `main`
|
9
|
+
config.disable_monkey_patching!
|
10
|
+
|
11
|
+
config.expect_with :rspec do |c|
|
12
|
+
c.syntax = :expect
|
13
|
+
end
|
14
|
+
end
|
@@ -9,9 +9,13 @@ Gem::Specification.new do |spec|
|
|
9
9
|
spec.homepage = "https://github.com/tthuydang/tomosia_amanaplus_crawl"
|
10
10
|
spec.required_ruby_version = Gem::Requirement.new(">= 2.3.0")
|
11
11
|
|
12
|
-
spec.files =
|
13
|
-
|
14
|
-
|
15
|
-
spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
12
|
+
spec.files = `git ls-files`.split("\n")
|
13
|
+
spec.bindir = "exe"
|
14
|
+
spec.executables = 'tomosia_amanaplus_crawl'
|
16
15
|
spec.require_paths = ["lib"]
|
16
|
+
|
17
|
+
spec.add_runtime_dependency('httparty', '0.18.1')
|
18
|
+
spec.add_runtime_dependency('nokogiri', '1.10.10')
|
19
|
+
spec.add_runtime_dependency('spreadsheet', '1.2.6')
|
20
|
+
spec.add_runtime_dependency('thor')
|
17
21
|
end
|
metadata
CHANGED
@@ -1,18 +1,75 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: tomosia_amanaplus_crawl
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.9
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Nhat Huy
|
8
8
|
autorequire:
|
9
|
-
bindir:
|
9
|
+
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2020-08-
|
12
|
-
dependencies:
|
11
|
+
date: 2020-08-08 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: httparty
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - '='
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: 0.18.1
|
20
|
+
type: :runtime
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - '='
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: 0.18.1
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: nokogiri
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - '='
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: 1.10.10
|
34
|
+
type: :runtime
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - '='
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: 1.10.10
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: spreadsheet
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - '='
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: 1.2.6
|
48
|
+
type: :runtime
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - '='
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: 1.2.6
|
55
|
+
- !ruby/object:Gem::Dependency
|
56
|
+
name: thor
|
57
|
+
requirement: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - ">="
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '0'
|
62
|
+
type: :runtime
|
63
|
+
prerelease: false
|
64
|
+
version_requirements: !ruby/object:Gem::Requirement
|
65
|
+
requirements:
|
66
|
+
- - ">="
|
67
|
+
- !ruby/object:Gem::Version
|
68
|
+
version: '0'
|
13
69
|
description:
|
14
70
|
email:
|
15
|
-
executables:
|
71
|
+
executables:
|
72
|
+
- tomosia_amanaplus_crawl
|
16
73
|
extensions: []
|
17
74
|
extra_rdoc_files: []
|
18
75
|
files:
|
@@ -21,13 +78,18 @@ files:
|
|
21
78
|
- ".travis.yml"
|
22
79
|
- CODE_OF_CONDUCT.md
|
23
80
|
- Gemfile
|
81
|
+
- Gemfile.lock
|
24
82
|
- LICENSE.txt
|
25
83
|
- README.md
|
26
84
|
- Rakefile
|
27
85
|
- bin/console
|
28
86
|
- bin/setup
|
87
|
+
- exe/tomosia_amanaplus_crawl
|
29
88
|
- lib/tomosia_amanaplus_crawl.rb
|
89
|
+
- lib/tomosia_amanaplus_crawl/cli.rb
|
30
90
|
- lib/tomosia_amanaplus_crawl/version.rb
|
91
|
+
- spec/spec_helper.rb
|
92
|
+
- spec/tomosia_amanaplus_crawl_spec.rb
|
31
93
|
- tomosia_amanaplus_crawl.gemspec
|
32
94
|
homepage: https://github.com/tthuydang/tomosia_amanaplus_crawl
|
33
95
|
licenses: []
|