bulk_ops 0.1.9 → 0.1.10

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f55b62bd327612efa466184efc9249133480ace66e3069fc7d20705eedfddf0f
-  data.tar.gz: 109e4e3cf3567864eeb15e73d3c6603219ee9577c8333c6cfb012409cac1b957
+  metadata.gz: 5b52469608636bbd2fc39c85bdc717b1c1777b7a5fa468b9e40137217b9ee0fe
+  data.tar.gz: 3c8a65f05674b6d43fa3d7a06bdc7a2a0424d56378fdf762dfde917c9c234575
 SHA512:
-  metadata.gz: d1172dd5be0f063944505b86da8a6a8b28913c69f1447523de9685d85524a55cb8a148f48ad112dbd7c158e9b63cab9ca50fbab0d2ff49c3ddd759dbd5940b52
-  data.tar.gz: 119bb5c71f7be6eb8c8514307eead3dcebf31304f1d4221575790ce3d4c931c5f57e9f3eb1157592ce3cbe805dfa2a815141196904a3f38cb533ea804f60a9b8
+  metadata.gz: 7e3386a5f0a2f33edac1eeb47cf6cfb95fbc11cb01d9c35ae7a547004d86c41d7fb390b1f3e1d429af3302e95107793b46522091afbb673faadca12facf79d34
+  data.tar.gz: 332a0abf7e52d7ac298c8b82bb0f79432e20a3432f095f6df9829b6a6827aa1b4286c8ea972484133e0be35d8fae9325a0435211bd0dba7900a0f677782adb35

data/lib/bulk_ops/relationship.rb

@@ -25,15 +25,24 @@ class BulkOps::Relationship < ActiveRecord::Base
     when "title"
       #          TODO clean up solr query and add work type to it
       query = "{!field f=title_tesim}#{object_identifier}"
-      objects = ActiveFedora::SolrService.instance.conn.get(ActiveFedora::SolrService.select_path,params: { fq: query, rows: 100})["response"]["docs"].first
-      object = objects.first
-      object ||= Collection.create(title: [object_identifier]) if work_type == "Collection"
-      return object || false
+      objects = ActiveFedora::SolrService.instance.conn.get(ActiveFedora::SolrService.select_path,
+                                                            params: { fq: query, rows: 100})["response"]["docs"].first
+      if objects.present?
+        return ActiveFedora::Base.find(objects.first["id"])
+      elsif work_type == "Collection"
+        return Collection.create(title: [object_identifier])
+      else
+        return false
+      end
     when "identifier"
       query = "{!field f=identifier_tesim}#{object_identifier}"
       objects = ActiveFedora::SolrService.instance.conn.get(ActiveFedora::SolrService.select_path,params: { fq: query, rows: 100})["response"]["docs"]
       return false if objects.blank?
-      return objects.first
+      return ActiveFedora::Base.find(objects.first["id"])
+    when "row"
+      object_proxy = WorkProxy.find_by(operation_id: work_proxy.operation.id, 
+                                       row_number: object_identifier.to_i)
+      ActiveFedora::Base.find(object_proxy.work_id)
     end
   end
 

data/lib/bulk_ops/verification.rb

@@ -188,18 +188,19 @@ module BulkOps
     def verify_internal_references
       # TODO 
       # This is sketchy. Redo it.
-      get_spreadsheet.each do |row,row_num|
+      (metadata = get_spreadsheet).each do |row,row_num|
         ref_id = get_ref_id(row)
         BulkOps::Operation::RELATIONSHIP_COLUMNS.each do |relationship|
           next unless (obj_id = row[relationship])
-          if (split = obj_id.split(':')).count == 2
+          if (split = obj_id.split(':')).present? && split.count == 2
             ref_id = split[0].downcase
             obj_id = split[1]
           end
           
           if ref_id == "row" || (ref_id == "id/row" && obj_id.is_a?(Integer))
+            obj_id = obj_id.to_i
             # This is a row number reference. It should be an integer in the range of possible row numbers.
-            unless obj_id.is_a? Integer && obj_id > 0 && obj_id <= metadata.count
+            unless obj_id.present?(obj_id > 0) && (obj_id <= metadata.count)
               @verification_errors << BulkOps::Error.new({type: :bad_object_reference, object_id: obj_id, row_number: row_num + ROW_OFFSET})
             end  
           elsif ref_id == "id" || ref_id == "hyrax id" || (ref_id == "id/row" && (obj_id.is_a? Integer))
@@ -207,14 +208,9 @@ module BulkOps
             unless record_exists?(obj_id)
               @verification_errors << BulkOps::Error.new({type: :bad_object_reference, object_id: obj_id, row_number: row_num+ROW_OFFSET})
             end
-          else
-
-            # This must be based on some other presumably unique field in hyrax, or a dummy field in the spreadsheet. We haven't added this functionality yet. Ignore for now.
-
           end
         end      
       end
     end
-
   end
 end

data/lib/bulk_ops/version.rb

@@ -1,3 +1,3 @@
 module BulkOps
-  VERSION = "0.1.9"
+  VERSION = "0.1.10"
 end

data/lib/bulk_ops/work_job.rb

@@ -58,7 +58,7 @@ class BulkOps::WorkJob < ActiveJob::Base
       return
     end
 
-    return unless define_work
+    return unless define_work(workClass)
 
     user = User.find_by_email(user_email)
     update_status "running", "Started background task at #{DateTime.now.strftime("%d/%m/%Y %H:%M")}"

data/lib/bulk_ops/work_proxy.rb

@@ -167,7 +167,8 @@ class BulkOps::WorkProxy < ActiveRecord::Base
 #          error_message =  "cannot fetch remote label for url: #{value}"
 #          report_error( :cannot_retrieve_label , error_message, url: value, row_number: row_number) unless label
         else
-          # It's a label, so get the id
+          # It's a label, so unescape it and get the id
+          value = unescape_csv(value)
           id = get_remote_id(value, property: field_name_norm, authority: authority) || localAuthUrl(field_name_norm, value)
 #          label = value
           report_error(:cannot_retrieve_url, 
@@ -175,7 +176,7 @@ class BulkOps::WorkProxy < ActiveRecord::Base
                        url: value, 
                        row_number: row_number) unless id
         end
-        controlled_data[field_name_norm] << {id: id, label: label, remove: field_name.downcase.starts_with?("remove")}
+        controlled_data[field_name_norm] << {id: id, remove: field_name.downcase.starts_with?("remove")}
       end
     end
     
@@ -210,6 +211,7 @@ class BulkOps::WorkProxy < ActiveRecord::Base
       values.split(SEPARATOR).each do |value|
         next if value.blank?
         value = value.strip.encode('utf-8', :invalid => :replace, :undef => :replace, :replace => '_') unless value.blank?
+        value = unescape_csv(value)
         (metadata[field_name] ||= []) << value
       end
     end
@@ -295,6 +297,7 @@ class BulkOps::WorkProxy < ActiveRecord::Base
     raw_data.each do |field,value|
       next if value.blank?  or field.blank?
       field = field.to_s              
+      value = unescape_csv(value)
 
       next if value == field
 
@@ -347,6 +350,10 @@ class BulkOps::WorkProxy < ActiveRecord::Base
       return "row"
     end
   end
+
+  def unescape_csv(value)
+    value.gsub(/\\(['";,])/,'\1')
+  end
   
   def format_worktype(value)
     # format the value like a class name
@@ -384,9 +391,8 @@ class BulkOps::WorkProxy < ActiveRecord::Base
     return nil unless (entries = Qa::Authorities::Local.subauthority_for(auth).search(value))
     entries.each do |entry|
       #require exact match
-      next unless entry["label"].encode('UTF-8') == value.encode('UTF-8')
-      url = entry["url"]
-      url ||= entry["id"]
+      next unless entry["label"].force_encoding('UTF-8') == value.force_encoding('UTF-8')
+      url = entry["url"] || entry["id"]
 #      url = localIdToUrl(url,auth) unless url =~ URI::regexp
       return url
     end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bulk_ops
 version: !ruby/object:Gem::Version
-  version: 0.1.9
+  version: 0.1.10
 platform: ruby
 authors:
 - Ned Henry, UCSC Library Digital Initiatives
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-07-11 00:00:00.000000000 Z
+date: 2019-08-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rails