data_collector 0.56.0 → 0.57.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0f53de79d6c24688e67edb48195a2f1ce4adaf69eaa616033cf1886cccd1e696
-  data.tar.gz: f5d13e7d3d03a9218c66c15907501f78b3e565ae8c69c012d6139ac28156a493
+  metadata.gz: d5f35afbd231bcacdb1d225cded55d4d7a0879ef71a9521241bb5e32c0ebacc7
+  data.tar.gz: 9c93c8e26d33921fd678668cf0ce2a152cbcd35d7d7fe928f03f328b25bd0f00
 SHA512:
-  metadata.gz: 426fe700363732ab23659a865d31d9cafe1eea16d7ad3e9deaddeb614d831e723f3e27d7cbd4ced72141e15928272eed11c1af4c7cb049aa38ad5aaf0088a48b
-  data.tar.gz: b0c57cad550f8f04c004fcae3c3b2b4879c8c01030aef5e29506752d8da397db1efce01a65c3df3cede620ef66acb64ee0e5cc2b3e926ac732b399b61e8742d4
+  metadata.gz: 3cd04dd04c330c3232ffe61810ec3b1e250c820872495bdbb6019590a68d603db596cb419e080bd01e858bd50ec50f630101254cb3787bacf042c8cf3e4c3991
+  data.tar.gz: d77da04c3d513adaedf983b83215a5c1427d0e1599df9c8d3cf6d702461b5eb96f52ef3d86a29f0471dc3ae314d3a9ac8d74f9772a7433d3b24df8b42e6e9742

data/lib/data_collector/input.rb

@@ -104,8 +104,6 @@ module DataCollector
 
       http = HTTP
 
-      # http.use(logging: {logger: @logger})
-
       if options.key?(:user) && options.key?(:password)
         @logger.debug "Set Basic_auth"
         user = options[:user]
@@ -157,13 +155,9 @@ module DataCollector
       when 200..299
         @raw = data = http_response.body.to_s
 
-        # File.open("#{rand(1000)}.xml", 'wb') do |f|
-        #   f.puts data
-        # end
-
         file_type = options.with_indifferent_access.has_key?(:content_type) ? options.with_indifferent_access[:content_type] : file_type_from(http_response.headers)
 
-        unless options.with_indifferent_access.has_key?(:raw) && options.with_indifferent_access[:raw] == true
+        unless options.with_indifferent_access.has_key?(:raw) && options.with_indifferent_access[:raw] == true && !['text/turtle', 'application/rdf+xml'].include?(file_type)
           case file_type
           when 'application/ld+json'
             data = JSON.parse(data)
@@ -181,6 +175,8 @@ module DataCollector
             data = html_to_hash(data, options)
           when 'text/turtle'
             data = rdf_to_hash(data, options)
+          when 'application/rdf+xml'
+            data = rdf_to_hash(data, options)
           when /^image/
             options['file_type'] = file_type
             data = image_to_data(data, options)
@@ -206,9 +202,11 @@ module DataCollector
     end
 
     def rdf_to_hash(data, options = {})
-      graph = RDF::Graph.new do |graph|
-        RDF::Turtle::Reader.new(data) { |reader| graph << reader }
-      end
+      format = options[:file_type] || options[:content_type] || 'text/turtle'
+      graph = RDF::Graph.new{ |graph| RDF::Format.content_types[format].first.reader.new(data){|reader| graph << reader}}
+
+      return graph if options.with_indifferent_access[:raw]
+
       data = JSON.parse(graph.dump(:jsonld, validate: false, standard_prefixes: true))
     end
 
@@ -223,7 +221,7 @@ module DataCollector
         sio.rewind
         file.write(sio.read)
         file.close
-        from_file(URI("file://#{file.path}"))
+        from_file(URI("file://#{file.path}"), options)
       ensure
         file.unlink
       end
@@ -234,7 +232,7 @@ module DataCollector
       uri = normalize_uri(uri)
       absolute_path = File.absolute_path(uri)
       file_type = MIME::Types.type_for(uri).first.to_s
-      options['file_type'] = file_type
+      options['file_type'] = options[:content_type] || file_type
       raise DataCollector::Error, "#{uri.to_s} not found" unless File.exist?("#{absolute_path}")
       unless options.has_key?('raw') && options['raw'] == true
         @raw = data = File.read("#{absolute_path}")
@@ -277,6 +275,8 @@ module DataCollector
           data = image_to_data(data, options)
         when '.ttl'
           data = rdf_to_hash(data, options)
+        when '.rdf'
+          data = rdf_to_hash(data, options)
         else
           raise "Do not know how to process #{uri.to_s}"
         end

data/lib/data_collector/version.rb

@@ -1,4 +1,4 @@
 # encoding: utf-8
 module DataCollector
-  VERSION = "0.56.0"
+  VERSION = "0.57.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: data_collector
 version: !ruby/object:Gem::Version
-  version: 0.56.0
+  version: 0.57.0
 platform: ruby
 authors:
 - Mehmet Celik
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2025-02-09 00:00:00.000000000 Z
+date: 2025-02-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport