repocrawler 0.1.16 → 0.1.17
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
 - data/lib/repocrawler/crawler.rb +49 -26
 - data/lib/repocrawler/version.rb +1 -1
 - metadata +1 -1
 
    
        checksums.yaml
    CHANGED
    
    | 
         @@ -1,7 +1,7 @@ 
     | 
|
| 
       1 
1 
     | 
    
         
             
            ---
         
     | 
| 
       2 
2 
     | 
    
         
             
            SHA1:
         
     | 
| 
       3 
     | 
    
         
            -
              metadata.gz:  
     | 
| 
       4 
     | 
    
         
            -
              data.tar.gz:  
     | 
| 
      
 3 
     | 
    
         
            +
              metadata.gz: c26e53deca5a1fdc7695619ad8c2b0e7e12c9b53
         
     | 
| 
      
 4 
     | 
    
         
            +
              data.tar.gz: 99baff26279fcf60051973161f8f0bdda2bd32cf
         
     | 
| 
       5 
5 
     | 
    
         
             
            SHA512:
         
     | 
| 
       6 
     | 
    
         
            -
              metadata.gz:  
     | 
| 
       7 
     | 
    
         
            -
              data.tar.gz:  
     | 
| 
      
 6 
     | 
    
         
            +
              metadata.gz: 557ddd7bc81605a3ad8bfd32de1b83cc275cd0f98f1809fcc2066cd41ca38cf94d51d5bb5181aaad6e439cc372acbd6f8f402a736934174948b097f3ebae4b30
         
     | 
| 
      
 7 
     | 
    
         
            +
              data.tar.gz: 9d9f93e0e171091f0af2a26ffdd24650e646a64626de2d4d42cbaf2547aff77d3ca766c84b59b0ef989240d89a36eb75b2cc05652a0e5c8ac39c9a1e59323d1a
         
     | 
    
        data/lib/repocrawler/crawler.rb
    CHANGED
    
    | 
         @@ -37,11 +37,16 @@ module Repos 
     | 
|
| 
       37 
37 
     | 
    
         
             
                def get_contributors
         
     | 
| 
       38 
38 
     | 
    
         
             
                  contributors = HTTParty.get(@GITHUB_API_BASE_URL + "/contributors?access_token=#{@access_token}", headers: {
         
     | 
| 
       39 
39 
     | 
    
         
             
                    "User-Agent" => @user_agent
         
     | 
| 
       40 
     | 
    
         
            -
                  }) 
     | 
| 
       41 
     | 
    
         
            -
             
     | 
| 
       42 
     | 
    
         
            -
             
     | 
| 
       43 
     | 
    
         
            -
             
     | 
| 
       44 
     | 
    
         
            -
                     
     | 
| 
      
 40 
     | 
    
         
            +
                  })
         
     | 
| 
      
 41 
     | 
    
         
            +
                  if contributors['message'] === 'Not Found'
         
     | 
| 
      
 42 
     | 
    
         
            +
                    contributor = nil
         
     | 
| 
      
 43 
     | 
    
         
            +
                  else
         
     | 
| 
      
 44 
     | 
    
         
            +
                    contributors.map do |contributor|
         
     | 
| 
      
 45 
     | 
    
         
            +
                      {
         
     | 
| 
      
 46 
     | 
    
         
            +
                        'name' => contributor['login'],
         
     | 
| 
      
 47 
     | 
    
         
            +
                        'contributions' => contributor['contributions']
         
     | 
| 
      
 48 
     | 
    
         
            +
                      }
         
     | 
| 
      
 49 
     | 
    
         
            +
                    end
         
     | 
| 
       45 
50 
     | 
    
         
             
                  end
         
     | 
| 
       46 
51 
     | 
    
         | 
| 
       47 
52 
     | 
    
         
             
                  contributors
         
     | 
| 
         @@ -50,8 +55,13 @@ module Repos 
     | 
|
| 
       50 
55 
     | 
    
         
             
                # get the total commits
         
     | 
| 
       51 
56 
     | 
    
         
             
                def get_total_commits
         
     | 
| 
       52 
57 
     | 
    
         
             
                  contributors = get_contributors
         
     | 
| 
       53 
     | 
    
         
            -
             
     | 
| 
       54 
     | 
    
         
            -
             
     | 
| 
      
 58 
     | 
    
         
            +
             
     | 
| 
      
 59 
     | 
    
         
            +
                  if contributors.nil?
         
     | 
| 
      
 60 
     | 
    
         
            +
                    commits = nil
         
     | 
| 
      
 61 
     | 
    
         
            +
                  else
         
     | 
| 
      
 62 
     | 
    
         
            +
                    commits = contributors.reduce(0) do |sum, num|
         
     | 
| 
      
 63 
     | 
    
         
            +
                      sum + num['contributions']
         
     | 
| 
      
 64 
     | 
    
         
            +
                    end
         
     | 
| 
       55 
65 
     | 
    
         
             
                  end
         
     | 
| 
       56 
66 
     | 
    
         | 
| 
       57 
67 
     | 
    
         
             
                  commits
         
     | 
| 
         @@ -110,6 +120,11 @@ module Repos 
     | 
|
| 
       110 
120 
     | 
    
         
             
                    commits_fetch = HTTParty.get(@GITHUB_API_BASE_URL + "/commits?page=#{page}&access_token=#{@access_token}", headers: {
         
     | 
| 
       111 
121 
     | 
    
         
             
                      "User-Agent" => @user_agent
         
     | 
| 
       112 
122 
     | 
    
         
             
                    })
         
     | 
| 
      
 123 
     | 
    
         
            +
             
     | 
| 
      
 124 
     | 
    
         
            +
                    if commits_fetch['message'] === 'Not Found'
         
     | 
| 
      
 125 
     | 
    
         
            +
                      break
         
     | 
| 
      
 126 
     | 
    
         
            +
                    end
         
     | 
| 
      
 127 
     | 
    
         
            +
             
     | 
| 
       113 
128 
     | 
    
         
             
                    if commits_fetch.count === 0
         
     | 
| 
       114 
129 
     | 
    
         
             
                      stop = true
         
     | 
| 
       115 
130 
     | 
    
         
             
                    end
         
     | 
| 
         @@ -137,6 +152,11 @@ module Repos 
     | 
|
| 
       137 
152 
     | 
    
         
             
                    issue_fetch = HTTParty.get(@GITHUB_API_BASE_URL + "/issues?state=closed&page=#{page}&access_token=#{@access_token}", headers: {
         
     | 
| 
       138 
153 
     | 
    
         
             
                      "User-Agent" => @user_agent
         
     | 
| 
       139 
154 
     | 
    
         
             
                    })
         
     | 
| 
      
 155 
     | 
    
         
            +
             
     | 
| 
      
 156 
     | 
    
         
            +
                    if issue_fetch['message'] === 'Not Found'
         
     | 
| 
      
 157 
     | 
    
         
            +
                      break
         
     | 
| 
      
 158 
     | 
    
         
            +
                    end
         
     | 
| 
      
 159 
     | 
    
         
            +
             
     | 
| 
       140 
160 
     | 
    
         
             
                    if issue_fetch.count === 0
         
     | 
| 
       141 
161 
     | 
    
         
             
                      stop = true
         
     | 
| 
       142 
162 
     | 
    
         
             
                    end
         
     | 
| 
         @@ -179,30 +199,33 @@ module Repos 
     | 
|
| 
       179 
199 
     | 
    
         
             
                    "User-Agent" => @user_agent
         
     | 
| 
       180 
200 
     | 
    
         
             
                  })
         
     | 
| 
       181 
201 
     | 
    
         | 
| 
      
 202 
     | 
    
         
            +
                  if github_contents['message'] === 'Not Found'
         
     | 
| 
      
 203 
     | 
    
         
            +
                    return nil
         
     | 
| 
      
 204 
     | 
    
         
            +
                  else
         
     | 
| 
      
 205 
     | 
    
         
            +
                    readme_file = ''
         
     | 
| 
      
 206 
     | 
    
         
            +
                    github_contents.each do |content|
         
     | 
| 
      
 207 
     | 
    
         
            +
                      readme_file = content['name'] if content['name'] =~ /^README/
         
     | 
| 
      
 208 
     | 
    
         
            +
                    end
         
     | 
| 
       182 
209 
     | 
    
         | 
| 
       183 
     | 
    
         
            -
             
     | 
| 
       184 
     | 
    
         
            -
             
     | 
| 
       185 
     | 
    
         
            -
             
     | 
| 
       186 
     | 
    
         
            -
             
     | 
| 
       187 
     | 
    
         
            -
             
     | 
| 
       188 
     | 
    
         
            -
                  stop_words = []
         
     | 
| 
       189 
     | 
    
         
            -
                  File.open(File.expand_path("../../public/stop_words.txt",  File.dirname(__FILE__)), "r") do |f|
         
     | 
| 
       190 
     | 
    
         
            -
                    f.each_line do |line|
         
     | 
| 
       191 
     | 
    
         
            -
                      stop_words << line.gsub(/\n/,"")
         
     | 
| 
      
 210 
     | 
    
         
            +
                    stop_words = []
         
     | 
| 
      
 211 
     | 
    
         
            +
                    File.open(File.expand_path("../../public/stop_words.txt",  File.dirname(__FILE__)), "r") do |f|
         
     | 
| 
      
 212 
     | 
    
         
            +
                      f.each_line do |line|
         
     | 
| 
      
 213 
     | 
    
         
            +
                        stop_words << line.gsub(/\n/,"")
         
     | 
| 
      
 214 
     | 
    
         
            +
                      end
         
     | 
| 
       192 
215 
     | 
    
         
             
                    end
         
     | 
| 
       193 
     | 
    
         
            -
                  end
         
     | 
| 
       194 
216 
     | 
    
         | 
| 
       195 
     | 
    
         
            -
             
     | 
| 
       196 
     | 
    
         
            -
             
     | 
| 
       197 
     | 
    
         
            -
             
     | 
| 
       198 
     | 
    
         
            -
             
     | 
| 
       199 
     | 
    
         
            -
             
     | 
| 
       200 
     | 
    
         
            -
             
     | 
| 
      
 217 
     | 
    
         
            +
                    readme = HTTParty.get(@GITHUB_README_URL + "/#{readme_file}")
         
     | 
| 
      
 218 
     | 
    
         
            +
                    words = readme.split(' ')
         
     | 
| 
      
 219 
     | 
    
         
            +
                    freqs = Hash.new(0)
         
     | 
| 
      
 220 
     | 
    
         
            +
                    words.each do |word|
         
     | 
| 
      
 221 
     | 
    
         
            +
                      if word =~ /^\w+$/ && !stop_words.include?(word.downcase)
         
     | 
| 
      
 222 
     | 
    
         
            +
                        freqs[word] += 1
         
     | 
| 
      
 223 
     | 
    
         
            +
                      end
         
     | 
| 
       201 
224 
     | 
    
         
             
                    end
         
     | 
| 
       202 
     | 
    
         
            -
             
     | 
| 
       203 
     | 
    
         
            -
                  freqs = freqs.sort_by { |word, freq| freq }.reverse!
         
     | 
| 
      
 225 
     | 
    
         
            +
                    freqs = freqs.sort_by { |word, freq| freq }.reverse!
         
     | 
| 
       204 
226 
     | 
    
         | 
| 
       205 
     | 
    
         
            -
             
     | 
| 
      
 227 
     | 
    
         
            +
                    freqs
         
     | 
| 
      
 228 
     | 
    
         
            +
                  end
         
     | 
| 
       206 
229 
     | 
    
         
             
                end
         
     | 
| 
       207 
230 
     | 
    
         
             
              end
         
     | 
| 
       208 
231 
     | 
    
         | 
    
        data/lib/repocrawler/version.rb
    CHANGED