bulk_ops 0.1.17 → 0.1.18

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 41f67b2f02e5495a78e4bd57457c6e8c4fc47b2d17579a5844ee57d950bd8cae
-  data.tar.gz: 119b59420890b6838dd26540a6209b32c04cd5e8e328510db88539d26c90715e
+  metadata.gz: a60889c5673952113a76b07c00b17b6ac5f99a4a2a3ac1099b3e2cb58ac04fd7
+  data.tar.gz: f7e2488b18b09092e90ef1fb3dad38e460d75bdfb355942065be98e28e743008
 SHA512:
-  metadata.gz: 68913835f6fbdca1773f446b3894278eefbb122feec5e327f9a44ca30aad2540f5791ac0379d2def5467509889000ad7ae95f4093309ad08e48dd3509d2550cb
-  data.tar.gz: 5a6dff8e04c77f3c0822509fe429331116e8a81d6cb4691a0bad89224acf189279862228fcde72f2d6e908e393642952bf6b7d712caa0fd4d4c719cfa434c3eb
+  metadata.gz: df3fc8ea76aae33515e73202d16a73ddbd86a3ae3c7a73245ec303c7599751b8c8e9961c1deee5971339cebd5bb9cd6ece60a962bebc6747c55244002a212326
+  data.tar.gz: cc1f2e6360da271a4203c5a6ccc6c67c793f7c5f5ce13b671ad79532c47a9044e76db9131af18090885b06b17aec9960a828db7bb2417a1d82a9afc6df90c6ba

data/lib/bulk_ops/error.rb

@@ -118,7 +118,7 @@ class BulkOps::Error
       end
 
     when :bad_object_reference 
-      message = "\n-- Error: bad object reference --\m" 
+      message = "\n-- Error: bad object reference --\n" 
       message += "We enountered #{errors.count} problems resolving object references.\n"
       if errors.count < max_error
          message += "The row numbers with problems were:\n"

data/lib/bulk_ops/operation.rb

@@ -97,7 +97,7 @@ module BulkOps
       # make sure the work proxies we just created are loaded in memory
       reload
       #loop through the work proxies to create a job for each work
-      @metadata.each_with_index do |values,row_number|
+      @metadata.dup.each_with_index do |values,row_number|
         proxy = work_proxies.find_by(row_number: row_number)
         proxy.update(message: "interpreted at #{DateTime.now.strftime("%d/%m/%Y %H:%M")} " + proxy.message)
         data = BulkOps::Parser.new(proxy, @metadata).interpret_data(raw_row: values)
@@ -240,7 +240,8 @@ module BulkOps
     end
 
     def get_spreadsheet return_headers: false
-      git.load_metadata return_headers: return_headers
+      branch = ((running? || complete?) ? "master" : nil)
+      git.load_metadata return_headers: return_headers, branch: branch
     end
 
     def spreadsheet_count
@@ -275,7 +276,7 @@ module BulkOps
     end
 
     def running?
-      return (stage == 'running')
+      return (['running','finishing'].include?(stage))
     end
 
     def complete?

data/lib/bulk_ops/parser.rb

@@ -9,9 +9,9 @@ class BulkOps::Parser
     return false unless metadata[row_number].present?
 #    If there are any valid fields other than relationship or file fields, it is a work
     metadata[row_number].each do |field, value|
-      next if is_file_field?(field)
-      next if ["parent", "order"].include(normalize_relationship_field_name(field))
-      next if ["title","label"].include(field.downcase.strip)
+      next if BulkOps::Verification.is_file_field?(field)
+      next if ["parent", "order"].include?(normalize_relationship_field_name(field))
+      next if ["title","label"].include?(field.downcase.strip)
       return false
     end
     return true
@@ -173,7 +173,7 @@ class BulkOps::Parser
       next if field == value
 
       # Check if this is a file field, and whether we are removing or adding a file
-      next unless (action = is_file_field?(field))
+      next unless (action = BulkOps::Verification.is_file_field?(field))
       
       # Move on if this field is the name of another property (e.g. masterFilename)
       next if find_field_name(field)
@@ -203,7 +203,7 @@ class BulkOps::Parser
         child_row.each do |field,value|
           next if value.blank?
           title = value if ["title","label"].include?(field.downcase.strip)
-          if is_file_field?(field)
+          if BulkOps::Verification.is_file_field?(field)
             operation.get_file_paths(value).each do |filepath|
               uploaded_file = Hyrax::UploadedFile.create(file:  File.open(filepath), user: operation.user)
             end
@@ -269,7 +269,7 @@ class BulkOps::Parser
         relationship_type = field
       end
 
-      relationship_type = normalize_relationship_field_name(relationship_type)
+      relationship_type = self.class.normalize_relationship_field_name(relationship_type)
       case relationship_type
       when "order"
         # If the field specifies the object's order among siblings 
@@ -312,7 +312,7 @@ class BulkOps::Parser
     @raw_row = row
   end
 
-  def normalize_relationship_field_name field
+  def self.normalize_relationship_field_name field
     normfield = field.downcase.parameterize.gsub(/[_\s-]/,'')
     BulkOps::RELATIONSHIP_FIELDS.find{|rel_field| normfield == rel_field }
   end
@@ -459,11 +459,6 @@ class BulkOps::Parser
     BulkOps::DeleteFileSetJob.perform_later(fileset_id, operation.user.email )
   end
  
-  
-  def is_file_field? field
-    operation.is_file_field? field
-  end
-
   def record_exists? id
     operation.record_exists? id
   end

data/lib/bulk_ops/verification.rb

@@ -31,9 +31,9 @@ module BulkOps
       end
     end
 
-    def is_file_field?(fieldname)
+    def self.is_file_field?(fieldname)
       return false if fieldname.blank?
-      return false if schema.get_field(fieldname)
+      return false if ScoobySnacks::METADATA_SCHEMA.get_field(fieldname)
       field_parts = fieldname.underscore.humanize.downcase.gsub(/[-_]/,' ').split(" ")
       return false unless field_parts.any?{ |field_type| BulkOps::FILE_FIELDS.include?(field_type) }
       return "remove" if field_parts.any?{ |field_type| ['remove','delete'].include?(field_type) }
@@ -46,7 +46,7 @@ module BulkOps
       name.gsub!(/[_\s-]?[lL]abel$/,'')
       name.gsub!(/^[rR]emove[_\s-]?/,'')
       name.gsub!(/^[dD]elete[_\s-]?/,'')
-      possible_fields = (Work.attribute_names + schema.all_field_names).uniq
+      possible_fields = ((Work.attribute_names || []) + schema.all_field_names).uniq
       matching_fields = possible_fields.select{|pfield| pfield.gsub(/[_\s-]/,'').parameterize == name.gsub(/[_\s-]/,'').parameterize }
       return false if matching_fields.blank?
       #      raise Exception "Ambiguous metadata fields!" if matching_fields.uniq.count > 1
@@ -72,7 +72,7 @@ module BulkOps
     def verify_files 
       file_errors = []
       get_spreadsheet.each_with_index do |row, row_num|
-        file_fields = row.select { |field, value| is_file_field?(field) }
+        file_fields = row.select { |field, value| BulkOps::Verification.is_file_field?(field) }
         file_fields.each do |column_name, filestring|
           next if filestring.blank? or column_name == filestring
           get_file_paths(filestring).each do |filepath|
@@ -103,7 +103,8 @@ module BulkOps
     end
 
     def downcase_first_letter(str)
-      str[0].downcase + str[1..-1]
+      return nil unless str.present?
+      (str[0].downcase || "") + str[1..-1]
     end
 
     # Make sure the headers in the spreadsheet are matching to properties
@@ -188,7 +189,7 @@ module BulkOps
     def verify_internal_references
       # TODO 
       # This is sketchy. Redo it.
-      (metadata = get_spreadsheet).each do |row,row_num|
+      (metadata = get_spreadsheet).each_with_index do |row,row_num|
         ref_id = get_ref_id(row)
         BulkOps::RELATIONSHIP_COLUMNS.each do |relationship|
           next unless (obj_id = row[relationship])
@@ -197,16 +198,15 @@ module BulkOps
             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.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})
+          if ref_id == "row" || (ref_id.include?("row") && obj_id.is_a?(Integer))
+            # This is a row number reference. It should be an integer or a string including "prev"
+            unless obj_id.present?  && ((obj_id =~ /\A[-+]?[0-9]+\z/) or (obj_id.strip.downcase.include?("prev") ))
+              @verification_errors << BulkOps::Error.new({type: :bad_object_reference, object_id: obj_id, row_number: row_num + BulkOps::ROW_OFFSET})
             end  
-          elsif ref_id == "id" || ref_id == "hyrax id" || (ref_id == "id/row" && (obj_id.is_a? Integer))
+          elsif ref_id.include?("id")
             # This is a hydra id reference. It should correspond to an object already in the repo
             unless record_exists?(obj_id)
-              @verification_errors << BulkOps::Error.new({type: :bad_object_reference, object_id: obj_id, row_number: row_num+ROW_OFFSET})
+              @verification_errors << BulkOps::Error.new({type: :bad_object_reference, object_id: obj_id, row_number: row_num + BulkOps:: ROW_OFFSET})
             end
           end
         end      

data/lib/bulk_ops/version.rb

@@ -1,3 +1,3 @@
 module BulkOps
-  VERSION = "0.1.17"
+  VERSION = "0.1.18"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: bulk_ops
 version: !ruby/object:Gem::Version
-  version: 0.1.17
+  version: 0.1.18
 platform: ruby
 authors:
 - Ned Henry, UCSC Library Digital Initiatives