bulk_ops 0.1.15 → 0.1.16

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: fea513373c0ae0267f9302311300b8f4ba03b9fa632db168aec201c2f8359182
-  data.tar.gz: baa0fe9b67bfbe7d2f8283ff7949cb8ec46e268c7e15ef17c7b73b9c3a80ef19
+  metadata.gz: e2951c792e8835750322d76ea69f9c81f4e5dd37190bfbc235c3b14f5f166de5
+  data.tar.gz: d2c26310a619ddd0476bb796d401fd7e1bb0f4659301ec69d8875eba230e75df
 SHA512:
-  metadata.gz: 33810a935cc44ee6de4448a12e37d4c0889b6a4c7d409011fc5dd9d0bddc18e1a53f0f18337c933ab3dd6903d4112b0a968579f20e7e204d4278220c0dbb0315
-  data.tar.gz: b7ff43aed578a7aba0cb59d0862af6d1ffe7f50eccce6715171063a09e1edf2670e3d23333b4e506ff3d473ff6dbed56f672ff66b2f43209e58e67950706072a
+  metadata.gz: 77e34cc192aedbd99549f7a1296019e164218e03e7c80356ccc30bcb62cb0756835b07b6ea43d5efb8edc71dcf8277e214c941e788277b4a32a4ffaf70c4bd3f
+  data.tar.gz: 8d4f3d182a8e298b51090c04fb470899d4c4075e207f75a61d1108eb22b78983e8d8baa48cf017b04e822c801f48131ef5efd08f068639284b6740e39ecead62

data/lib/bulk_ops/github_access.rb

@@ -212,16 +212,16 @@ class BulkOps::GithubAccess
     client.merge_pull_request(repo, pull_id, message)
   end
 
-  def get_metadata_row row_number
-    @current_metadata ||= load_metadata
-    @current_metadata[row_number - BulkOps::ROW_OFFSET]
-  end
-  
-  def get_past_metadata_row commit_sha, row_number
-    past_metadata = Base64.decode64( client.contents(repo, path: filename, ref: commit_sha) )
-    past_metadata[row_number - BulkOps::ROW_OFFSET]
-  end
-
+#  def get_metadata_row row_number
+#    @current_metadata ||= load_metadata
+#    @current_metadata[row_number - BulkOps::ROW_OFFSET]
+#  end
+#  
+#  def get_past_metadata_row commit_sha, row_number
+#    past_metadata = Base64.decode64( client.contents(repo, path: filename, ref: commit_sha) )
+#    past_metadata[row_number - BulkOps::ROW_OFFSET]
+#  end
+#
   def get_file filename
     client.contents(repo, path: filename, ref: name)
   end

data/lib/bulk_ops/operation.rb

@@ -81,15 +81,19 @@ module BulkOps
       #Destroy any existing work proxies (which should not exist for an ingest). Create new proxies from finalized spreadsheet only.
       work_proxies.each{|proxy| proxy.destroy!}
 
-      #create a work proxy for each row in the spreadsheet
+      #create a work proxy for each work in the spreadsheet, creating filesets where appropriate
       @metadata.each_with_index do |values,row_number|
         next if values.to_s.gsub(',','').blank?
+
+        next if BulkOps::Parser.is_file_set? @metadata, row_number
+
         work_proxies.create(status: "queued",
                             last_event: DateTime.now,
                             row_number: row_number,
                             visibility: options['visibility'],
                             message: "created during ingest initiated by #{user.name || user.email}")
       end
+      
       # 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

data/lib/bulk_ops/parser.rb

@@ -5,6 +5,18 @@ class BulkOps::Parser
 
   delegate :relationships, :operation, :row_number, :work_id, :visibility, :work_type, :reference_identifier, :order, to: :proxy
 
+  def self.is_file_set? metadata, row_number
+    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)
+      return false
+    end
+    return true
+  end
+
   def initialize prx, metadata_sheet=nil
     @proxy = prx
     @raw_data = (metadata_sheet || proxy.operation.metadata)
@@ -18,9 +30,11 @@ class BulkOps::Parser
     @proxy = proxy if proxy.present?
     @raw_data = raw_data if raw_data.present?
     setAdminSet
+    #The order here matters a little: interpreting the relationship fields specifies containing collections,
+    # which may have opinions about whether we should inherit metadata from parent works
+    interpret_relationship_fields
     setMetadataInheritance
     interpret_option_fields
-    interpret_relationship_fields
     disambiguate_columns
     interpret_file_fields
     interpret_controlled_fields
@@ -150,7 +164,6 @@ class BulkOps::Parser
     # some or all existing files, those replacement-related deletions are handled
     # by the BulkOps::Operation.
     #
-    # TODO: THIS DOES NOT YET MANAGE THE ORDER OF INGESTED FILESETS
 
     row = @raw_row.dup
     @raw_row.each do |field, value|
@@ -159,7 +172,6 @@ class BulkOps::Parser
       #If our CSV interpreter is feeding us the headers as a line, ignore it.
       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))
       
@@ -184,6 +196,22 @@ class BulkOps::Parser
           end
         end
       end
+
+      # Check if any of the upcoming rows are child filesets
+      i = 1
+      while self.class.is_file_set?(@metadata,row_number+i)
+        child_row.each do |field,value|
+          next if value.blank?
+          title = value if ["title","label"].include?(field.downcase.strip)
+          if 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
+          end
+        end
+        i+=1
+      end
+
     end
     @raw_row = row
   end
@@ -259,7 +287,7 @@ class BulkOps::Parser
         
         # correctly interpret the notation "id:a78C2d81"
         identifier_type, object_identifier = interpret_relationship_value(identifier_type, value)
-
+        
         relationship_parameters =  { work_proxy_id: @proxy.id,
                                      identifier_type: identifier_type,
                                      relationship_type: relationship_type,
@@ -302,14 +330,20 @@ class BulkOps::Parser
   def interpret_relationship_value id_type, value, field="parent"
     #Handle "id:20kj4259" syntax if it hasn't already been handled
     if (split = value.to_s.split(":")).count == 2
-      id_type = split.first
+      id_type, value = split.first
       value = split.last
     end
     #Handle special shorthand syntax for refering to relative row numbers
     if id_type == "row"
-      if value.to_i < 0
+      #if the value is an integer
+      if value =~ /\A[-+]?[0-9]+\z/
+        if value.to_i < 0
         # if given a negative integer, count backwards from the current row (remember that value.to_i is negative)
-        return [id_type,row_number + value.to_i]
+          return [id_type,row_number + value.to_i]
+        elsif value.to_i > 0
+          # if given a positive integer, remove the row offset
+          value = (value.to_i - BulkOps::ROW_OFFSET).to_s
+        end
       elsif value.to_s.downcase.include?("prev")
         # if given any variation of the word "previous", get the first preceding row with no parent of its own
         return [id_type,find_previous_parent(field)]

data/lib/bulk_ops/relationship.rb

@@ -41,7 +41,7 @@ class BulkOps::Relationship < ActiveRecord::Base
       return ActiveFedora::Base.find(objects.first["id"])
     when "row"
       object_proxy = BulkOps::WorkProxy.find_by(operation_id: work_proxy.operation_id, 
-                                                row_number: (object_identifier.to_i - 2))
+                                                row_number: (object_identifier.to_i))
       ActiveFedora::Base.find(object_proxy.work_id)
     when "proxy_id"
       return false unless (proxy = BulkOps::WorkProxy.find(proxy_id))

data/lib/bulk_ops/version.rb

@@ -1,3 +1,3 @@
 module BulkOps
-  VERSION = "0.1.15"
+  VERSION = "0.1.16"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bulk_ops
 version: !ruby/object:Gem::Version
-  version: 0.1.15
+  version: 0.1.16
 platform: ruby
 authors:
 - Ned Henry, UCSC Library Digital Initiatives
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-10-03 00:00:00.000000000 Z
+date: 2019-10-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rails