tomosia_amanaplus_crawl 0.1.5 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7e09b5361e7b84f880485d64480eca862a19dcbd32c0cf4a0bab3160c3b5ab11
-  data.tar.gz: fc69ac8dcb7e3ba4c74412bb4760ef2cd996831d572eb63f184624330d8789a8
+  metadata.gz: c0f30081ca8543573b8f1f16c2aaf77384d899152d4f5f424f5008dfe1c7c0fc
+  data.tar.gz: a819309522dd41b9346a0c43fe538bbd1b380b7524e0a3cac2dc00cdf4193d48
 SHA512:
-  metadata.gz: 8ef2cbc7fa812b63ff04e136dcc5ff479fede20b7489485b83ce5dc84c5642fe3eba846bfc8a14c335b7c9271acf4bbcf2a6e0d4e2e557302488359306e820f4
-  data.tar.gz: 414a5491c0227c9f43830a11384e90b55305b0cdcb14d1b963c60c36aa278b748631296ad4bee837ed05184c490df8f747a0cc239025c6bd12d831ddb3ecb704
+  metadata.gz: 23afb7fba2307a0107b549a9e060fe4959de8f89dfad79cd63c0b0c4f2d7627ace2ef447b830ad7ad22c42ed9fb9a9b16b37a73d102386b01489a9046263d64d
+  data.tar.gz: a2d1201db75b05b6d80fa1a17fa4a0594283d92ad928c2dc22c214305877447ce2aa05685aa57e4bf6a06155416bda62e0d70a3b77dda9d02a8ace6714f09b74

data/Gemfile.lock

@@ -1,10 +1,11 @@
 PATH
   remote: .
   specs:
-    tomosia_amanaplus_crawl (0.1.5)
+    tomosia_amanaplus_crawl (0.1.8)
       httparty (= 0.18.1)
       nokogiri (= 1.10.10)
       spreadsheet (= 1.2.6)
+      thor
 
 GEM
   remote: https://rubygems.org/
@@ -37,6 +38,7 @@ GEM
     ruby-ole (1.2.12.2)
     spreadsheet (1.2.6)
       ruby-ole (>= 1.0)
+    thor (1.0.1)
 
 PLATFORMS
   ruby

data/README.md

@@ -10,9 +10,6 @@ Add this line to your application's Gemfile:
 
 ```ruby
 gem 'tomosia_amanaplus_crawl'
-gem 'httparty'
-gem 'nokogiri'
-gem 'spreadsheet'
 ```
 
 And then execute:
@@ -26,9 +23,9 @@ Or install it yourself as:
 ## Usage
 
 ```ruby
-    require 'tomosia_amanaplus_crawl'
-    TomosiaAmanaplusCrawl::Crawler.new.run(keyword, path, max)
+    tomosia_amanaplus_crawl crawl "keyword" --destination "/home/usr/Documents" --max=123
 ```
+    Example: tomosia_amanaplus_crawl crawl "hoian" --destination "./" --max=123
     keyword: hoian, danang, ...
     path: './', '/desktop/', ...
     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

@@ -4,3 +4,9 @@ require "rspec/core/rake_task"
 RSpec::Core::RakeTask.new(:spec)
 
 task :default => :spec
+
+namespace :gem do
+  task :build do
+    system "rake build && rake install"
+  end
+end

data/exe/tomosia_amanaplus_crawl

@@ -0,0 +1,5 @@
+#!/usr/bin/env ruby
+
+require 'tomosia_amanaplus_crawl/cli'
+
+TomosiaAmanaplusCrawl::Cli.start

data/lib/tomosia_amanaplus_crawl.rb

@@ -8,10 +8,10 @@ module TomosiaAmanaplusCrawl
   require 'spreadsheet'
 
   class Crawler
-    URL = "https://plus.amanaimages.com/items/search/"
+    URL = "https://plus.amanaimages.com/items/search"
 
     def run(keyword, destination, max)
-      unparsed_page = HTTParty.get("#{URL}/#{keyword}")
+      unparsed_page = open("#{URL}/#{keyword}").read
       parsed_page = Nokogiri::HTML(unparsed_page)
 
       pages = parsed_page.css("div.c-paginate__nums").css('a').last.text.to_i # tổng số page
@@ -19,7 +19,7 @@ module TomosiaAmanaplusCrawl
     
       # lấy tổng số image
       total = parsed_page.css("h1.p-search-result__ttl").text.split(' ').first
-      total = total[11..(total.length - 1)].chop.chop.chop.sub(',', '').to_i
+      total = total[(6 + keyword.length)..(total.length - 1)].chop.chop.chop.gsub(',', '').to_i
       if max > total # nếu max lớn hơn total thì max = total => vẫn lấy hết
         max = total
       end
@@ -36,7 +36,7 @@ module TomosiaAmanaplusCrawl
       while curr_page <= pages
         puts "Crawling page #{curr_page}..........."
     
-        pagination_unparsed_page = HTTParty.get("https://plus.amanaimages.com/items/search/#{keyword}?page=#{curr_page}")
+        pagination_unparsed_page = open("#{URL}/#{keyword}?page=#{curr_page}").read
         pagination_parsed_page = Nokogiri::HTML(pagination_unparsed_page)
         pagination_images_listings = pagination_parsed_page.css("div.p-item-thumb")
     
@@ -70,13 +70,23 @@ module TomosiaAmanaplusCrawl
       print "\nDownloading"
       images.each do |curr_image|
         threads << Thread.new(curr_image) {
-          open(curr_image[:url]) do |image|
-            File.open("#{path}/#{curr_image[:url].split('/').last}", "a+") do |file|
-              file.write(image.read) # lưu hình ảnh
-              curr_image[:size] = image.size # cập nhật lại size trong mảng images
-              print "."
+          timeout = 0
+          begin
+            URI.open(curr_image[:url]) do |image|
+              File.open("#{path}/#{curr_image[:url].split('/').last}", "a+") do |file|
+                file.write(image.read) # lưu hình ảnh
+                curr_image[:size] = image.size # cập nhật lại size trong mảng images
+                print "."
+              end
+            end # end open
+          rescue => exception
+            if timeout < 3
+              timeout += 1
+              retry
+            else
+              next
             end
-          end # end open
+          end
         }
       end
       threads.each { |t| t.join }

data/lib/tomosia_amanaplus_crawl/cli.rb

@@ -0,0 +1,14 @@
+require 'thor'
+require_relative '../tomosia_amanaplus_crawl'
+
+module TomosiaAmanaplusCrawl
+  class Cli < Thor
+
+    desc "crawl KEYWORD", "enter KEYWORD to search"
+    option :destination
+    option :max
+    def crawl(keyword)
+      TomosiaAmanaplusCrawl::Crawler.new.run(keyword, options[:destination], options[:max].to_i)
+    end
+  end
+end

data/lib/tomosia_amanaplus_crawl/version.rb

@@ -1,3 +1,3 @@
 module TomosiaAmanaplusCrawl
-  VERSION = "0.1.5"
+  VERSION = "0.2.0"
 end

data/spec/spec_helper.rb

@@ -0,0 +1,14 @@
+require "bundler/setup"
+require "tomosia_amanaplus_crawl"
+
+RSpec.configure do |config|
+  # Enable flags like --only-failures and --next-failure
+  config.example_status_persistence_file_path = ".rspec_status"
+
+  # Disable RSpec exposing methods globally on `Module` and `main`
+  config.disable_monkey_patching!
+
+  config.expect_with :rspec do |c|
+    c.syntax = :expect
+  end
+end

data/spec/tomosia_amanaplus_crawl_spec.rb

@@ -0,0 +1,9 @@
+RSpec.describe TomosiaAmanaplusCrawl do
+  it "has a version number" do
+    expect(TomosiaAmanaplusCrawl::VERSION).not_to be nil
+  end
+
+  it "does something useful" do
+    expect(false).to eq(true)
+  end
+end

data/tomosia_amanaplus_crawl.gemspec

@@ -9,13 +9,13 @@ Gem::Specification.new do |spec|
   spec.homepage      = "https://github.com/tthuydang/tomosia_amanaplus_crawl"
   spec.required_ruby_version = Gem::Requirement.new(">= 2.3.0")
 
-  spec.files         = Dir.chdir(File.expand_path('..', __FILE__)) do
-    `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
-  end
-  spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
+  spec.files         = `git ls-files`.split("\n")
+  spec.bindir        = "exe"
+  spec.executables   = 'tomosia_amanaplus_crawl'
   spec.require_paths = ["lib"]
 
   spec.add_runtime_dependency('httparty', '0.18.1')
   spec.add_runtime_dependency('nokogiri', '1.10.10')
   spec.add_runtime_dependency('spreadsheet', '1.2.6')
+  spec.add_runtime_dependency('thor')
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: tomosia_amanaplus_crawl
 version: !ruby/object:Gem::Version
-  version: 0.1.5
+  version: 0.2.0
 platform: ruby
 authors:
 - Nhat Huy
 autorequire: 
-bindir: bin
+bindir: exe
 cert_chain: []
-date: 2020-08-07 00:00:00.000000000 Z
+date: 2020-08-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: httparty
@@ -52,9 +52,24 @@ dependencies:
     - - '='
       - !ruby/object:Gem::Version
         version: 1.2.6
+- !ruby/object:Gem::Dependency
+  name: thor
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 description: 
 email: 
-executables: []
+executables:
+- tomosia_amanaplus_crawl
 extensions: []
 extra_rdoc_files: []
 files:
@@ -69,8 +84,12 @@ files:
 - Rakefile
 - bin/console
 - bin/setup
+- exe/tomosia_amanaplus_crawl
 - lib/tomosia_amanaplus_crawl.rb
+- lib/tomosia_amanaplus_crawl/cli.rb
 - lib/tomosia_amanaplus_crawl/version.rb
+- spec/spec_helper.rb
+- spec/tomosia_amanaplus_crawl_spec.rb
 - tomosia_amanaplus_crawl.gemspec
 homepage: https://github.com/tthuydang/tomosia_amanaplus_crawl
 licenses: []