Add collection_handles column to CSV

app/util/csv_exporter.rb

@@ -121,7 +121,17 @@ def export_unit(unit, elements: [])
   # @return [String]
   #
   def select_clause(elements)
-    columns = ["items.id", "handles.suffix"]
+    columns = ["items.id", "item_handles.suffix AS item_handle"]
+    # collection_handles column
+    columns << "array_to_string(
+       array(
+         SELECT CONCAT('#{Handle.prefix}/', h.suffix) AS suffix
+         FROM handles h
+         LEFT JOIN collection_item_memberships cim ON cim.item_id = items.id
+         WHERE cim.item_id = items.id
+         AND h.collection_id = cim.collection_id
+         ORDER BY cim.primary DESC NULLS LAST
+       ), '||') AS collection_handles\n"
     # files column
     columns << "array_to_string(
        array(
@@ -192,7 +202,7 @@ def from_clause
   end
 
   def join_clauses
-    "LEFT JOIN handles ON handles.item_id = items.id\n" +
+    "LEFT JOIN handles item_handles ON item_handles.item_id = items.id\n" +
     "LEFT JOIN collection_item_memberships cim ON cim.item_id = items.id "
   end
 
@@ -209,4 +219,4 @@ def to_csv(elements, results)
     end
   end
 
-end
+end

app/util/csv_importer.rb

@@ -29,14 +29,27 @@
 #
 # The first column, `id`, refers to an item's internal database ID. The second
 # column, `handle`, refers to an item's handle. (These columns cannot be
-# modified.) The next several columns refer to various system-level item
-# properties. The remaining columns correspond to the elements in the effective
+# modified.)
+#
+# The third column, `collection_handles`, refers to the handles of the
+# collections to which the item belongs. There may be multiple handles listed
+# in this column, separated by a double pipe (`||`), as an item may belong to
+# multiple collections. If this is the case, the first handle listed will
+# correspond to the item's primary collection. Whether or not there are
+# multiple handles, the first handle in this column must match that of the
+# collection associated with the import.
+#
+# The next several columns refer to various other item properties
+# These columns are collectively listed in {CsvImporter#REQUIRED_COLUMNS}.
+#
+# The remaining columns correspond to the elements in the effective
 # {MetadataProfile} of the collection that apply to the item.
 #
-# All columns are required except metadata columns. When a metadata column is
-# missing, the corresponding metadata elements of the items in the CSV file
-# will be left unchanged. For new items in the CSV file, a missing metadata
-# column may be an error depending on whether it is required or not.
+# All columns are required to exist (even if empty) except metadata columns.
+# When a metadata column is missing, the corresponding metadata elements of the
+# items in the CSV file will be left unchanged. For new items in the CSV file,
+# a missing metadata column may be an error depending on whether it is required
+# or not.
 #
 # ## Non-header rows
 #
@@ -48,6 +61,13 @@
 #
 # The next group of cells correspond to various system properties of the item:
 #
+# * `handle`                     The item's handle. This is for informational
+#                                purposes only--it cannot be updated by
+#                                changing the value.
+# * `collection_handle`          The handle of the item's owning {Collection}.
+#                                Changing the value to the handle of some other
+#                                collection will cause the item to be moved
+#                                into it.
 # * `files`                      Names of all files attached to an item. For
 #                                uploads, this may be a path relative to a
 #                                package root in which the CSV resides.
@@ -113,11 +133,12 @@
 #
 class CsvImporter
 
-  MULTI_VALUE_DELIMITER  = "||"
-  NEW_ITEM_INDICATOR     = "+"
-  REQUIRED_COLUMNS       = %w[id handle files file_descriptions embargo_types
-                              embargo_expirations embargo_exempt_user_groups
-                              embargo_reasons]
+  MULTI_VALUE_DELIMITER = "||"
+  NEW_ITEM_INDICATOR    = "+"
+  REQUIRED_COLUMNS      = %w[id handle collection_handles files
+                             file_descriptions embargo_types
+                             embargo_expirations embargo_exempt_user_groups
+                             embargo_reasons]
 
   ##
   # Imports items from a CSV string.
@@ -126,8 +147,12 @@ class CsvImporter
   # @param file_paths [Enumerable<String>] Absolute paths of files that are
   #                                        referenced in the CSV's `files`
   #                                        column.
-  # @param submitter [User]
-  # @param primary_collection [Collection] Collection to import new items into.
+  # @param submitter [User]                The user performing the import.
+  # @param primary_collection [Collection] Collection associated with the
+  #                                        import. This may be the collection
+  #                                        that items are added to if it is
+  #                                        not overridden by the
+  #                                        `collection_handles` column.
   # @param imported_items [Array<Hash>]    For each imported item, whether
   #                                        created or updated, a hash
   #                                        containing `:item_id` and `:handle`
@@ -144,6 +169,9 @@ def import(pathname:,
              imported_items:     [],
              print_progress:     false,
              task:               nil)
+    raise ArgumentError, "Pathname is nil" unless pathname
+    raise ArgumentError, "File does not exist: #{pathname}" unless File.exist?(pathname)
+    raise ArgumentError, "Nil submitter argument" unless submitter
     num_rows = 0
     File.foreach(pathname) do
       num_rows += 1
@@ -174,18 +202,19 @@ def import(pathname:,
                                file_paths:         file_paths)
           else
             item = Item.find(item_id)
-            item = update_item(item:          item,
-                               submitter:     submitter,
-                               element_names: header_row[REQUIRED_COLUMNS.length..],
-                               row:           row,
-                               file_paths:    file_paths)
+            item = update_item(item:               item,
+                               submitter:          submitter,
+                               primary_collection: primary_collection,
+                               element_names:      header_row[REQUIRED_COLUMNS.length..],
+                               row:                row,
+                               file_paths:         file_paths)
           end
           imported_items << {
             item_id: item.id,
             handle:  item.handle&.handle
           }
           status_text = "Importing #{num_rows} items from CSV"
-          row_index += 1
+          row_index  += 1
           progress&.report(row_index, status_text)
           task&.progress(row_index / (num_rows - 1).to_f,
                          status_text: status_text)
@@ -213,6 +242,10 @@ def create_item(submitter:,
                                  stage:              Item::Stages::APPROVED,
                                  event_description:  "Item imported from CSV.").execute
     item.assign_handle
+    move_into_collections(item:               item,
+                          primary_collection: primary_collection,
+                          collection_handles: row[2],
+                          submitter:          submitter)
     associate_bitstreams(item:       item,
                          row:        row,
                          file_paths: file_paths) if file_paths
@@ -227,12 +260,17 @@ def create_item(submitter:,
 
   def update_item(item:,
                   submitter:,
+                  primary_collection:,
                   element_names:,
                   row:,
-                  file_paths:    nil)
+                  file_paths:         nil)
     UpdateItemCommand.new(item:        item,
                           user:        submitter,
                           description: "Updated via CSV").execute do
+      move_into_collections(item:               item,
+                            primary_collection: primary_collection,
+                            collection_handles: row[2],
+                            submitter:          submitter)
       associate_bitstreams(item:       item,
                            row:        row,
                            file_paths: file_paths)
@@ -350,6 +388,45 @@ def associate_embargoes(item:, row:)
     end
   end
 
+  ##
+  # @param item [Item]                     Item to move.
+  # @param primary_collection [Collection] Collection associated with the
+  #                                        import.
+  # @param collection_handles [String]     Value of the `collection_handles`
+  #                                        column.
+  # @param submitter [User]                User performing the import.
+  #
+  def move_into_collections(item:,
+                            primary_collection:,
+                            collection_handles:,
+                            submitter:)
+    return if collection_handles.blank?
+    collection_handles = collection_handles.split(MULTI_VALUE_DELIMITER)
+    return if collection_handles.empty?
+    item.collection_item_memberships.destroy_all
+    collection_handles.each_with_index do |handle, index|
+      suffix = handle.split("/").last.strip
+      handle = Handle.find_by_suffix(suffix)
+      unless handle
+        raise ArgumentError, "Collection with handle #{handle} does not exist"
+      end
+      collection = handle.collection
+      if index == 0 && collection != primary_collection
+        raise ArgumentError, "The first handle in the collection_handles "\
+                             "column must be that of the collection "\
+                             "associated with the import"
+      end
+      unless submitter.effective_collection_admin?(collection)
+        raise ArgumentError, "User #{submitter} does not have permission to "\
+                             "add items to #{collection.title}"
+      end
+      unless item.collections.include?(collection)
+        item.collection_item_memberships.build(collection: handle.collection,
+                                               primary:    index == 0).save!
+      end
+    end
+  end
+
   ##
   # @param row [Hash<String>]
   # @param submission_profile [SubmissionProfile]
@@ -364,7 +441,7 @@ def validate_header(row:, submission_profile:)
     end
     REQUIRED_COLUMNS.each_with_index do |column, index|
       if row[index] != column
-        raise ArgumentError, "Missing #{column} column"
+        raise ArgumentError, "Missing #{column} column (expected at position #{index + 1})"
       end
     end
   end

test/fixtures/files/csv/illegal_element.csv

@@ -1,2 +1,2 @@
-id,handle,files,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,bogus
-+,"","","","","","","",Value of bogus element
+id,handle,collection_handles,files,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,bogus
++,"","","","","","","","",Value of bogus element

test/fixtures/files/csv/new.csv

@@ -1,2 +1,2 @@
-id,handle,files,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,dc:title
-+,"",hello.txt,"","","","","",New CSV Item
+id,handle,collection_handles,files,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,dc:title
++,"","",hello.txt,"","","","","",New CSV Item

test/fixtures/files/packages/csv/missing_file/package.csv

@@ -1,2 +1,2 @@
-id,handle,filenames,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,dc:title,dc:subject,dc:creator,dc:contributor,dc:description:abstract,dc:date:issued,dc:identifier:uri,dc:type
-+,"",missing.txt,Hello world,"","","","",New Item 1,Subject 1||Subject 2,,,,,,
+id,handle,collection_handles,filenames,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,dc:title,dc:subject,dc:creator,dc:contributor,dc:description:abstract,dc:date:issued,dc:identifier:uri,dc:type
++,"","",missing.txt,Hello world,"","","","",New Item 1,Subject 1||Subject 2,,,,,,

test/fixtures/files/packages/csv/valid_items/package.csv

@@ -1,3 +1,3 @@
-id,handle,files,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,dc:title,dc:subject,dc:creator,dc:contributor,dc:description:abstract,dc:date:issued,dc:identifier:uri,dc:type
-+,,item1/hello.txt||item1/license.txt,Hello world||License file,,,,,New Item 1,Subject 1||Subject 2,,,,,,
-+,,item2/hello.txt||item2/license.txt,Hello world||License file,0||1,2045-02-05||2055-03-10,sysadmin||,Reason 1||Reason 2,New Item 2,Subject 1||Subject 2,,,,,,
+id,handle,collection_handles,files,file_descriptions,embargo_types,embargo_expirations,embargo_exempt_user_groups,embargo_reasons,dc:title,dc:subject,dc:creator,dc:contributor,dc:description:abstract,dc:date:issued,dc:identifier:uri,dc:type
++,,,item1/hello.txt||item1/license.txt,Hello world||License file,,,,,New Item 1,Subject 1||Subject 2,,,,,,
++,,,item2/hello.txt||item2/license.txt,Hello world||License file,0||1,2045-02-05||2055-03-10,sysadmin||,Reason 1||Reason 2,New Item 2,Subject 1||Subject 2,,,,,,

test/fixtures/handles.yml

@@ -2,66 +2,74 @@
 
 southwest_unit1_collection1:
   collection: southwest_unit1_collection1
-  suffix:     5014
+  suffix:     5000
 
 southwest_unit1_collection1_item1:
   item:   southwest_unit1_collection1_item1
-  suffix: 5000
+  suffix: 5001
 
 ################################### southeast ####################################
 
 # collection handles
 
 southeast_collection1:
   collection: southeast_collection1
-  suffix:     5001
+  suffix:     5002
+
+southeast_collection2:
+  collection: southeast_collection2
+  suffix:     5003
+
+southeast_empty:
+  collection: southeast_empty
+  suffix:     5004
 
 # item handles
 
 southeast_item1:
   item:   southeast_item1
-  suffix: 5002
+  suffix: 5005
 
 southeast_approved:
   item:   southeast_approved
-  suffix: 5003
+  suffix: 5006
 
 southeast_awaiting_ingest_into_medusa:
   item:   southeast_awaiting_ingest_into_medusa
-  suffix: 5004
+  suffix: 5007
 
 southeast_in_medusa:
   item:   southeast_in_medusa
-  suffix: 5005
+  suffix: 5008
 
 southeast_described:
   item:   southeast_described
-  suffix: 5006
+  suffix: 5009
 
 southeast_embargoed:
   item:   southeast_embargoed
-  suffix: 5007
+  suffix: 5010
 
 southeast_undescribed:
   item:   southeast_undescribed
-  suffix: 5008
+  suffix: 5011
 
 southeast_submitting:
   item:   southeast_submitting
-  suffix: 5009
+  suffix: 5012
 
 southeast_withdrawn:
   item:   southeast_withdrawn
-  suffix: 5010
+  suffix: 5013
 
 southeast_buried:
   item:   southeast_buried
-  suffix: 5011
+  suffix: 5014
 
 southeast_multiple_bitstreams:
   item:   southeast_multiple_bitstreams
-  suffix: 5012
+  suffix: 5015
 
 southeast_collection1_collection1_item1:
   item:   southeast_collection1_collection1_item1
-  suffix: 5013
+  suffix: 5016

test/jobs/import_job_test.rb

@@ -58,7 +58,8 @@ class ImportJobTest < ActiveSupport::TestCase
     ObjectStore.instance.put_object(key:  import.file_key,
                                     path: fixture)
 
-    format = ImportJob.perform_now(import: import)
+    format = ImportJob.perform_now(import: import,
+                                   user:   users(:southeast_admin))
     assert_equal Import::Format::CSV_FILE, format
   end
 
@@ -72,7 +73,8 @@ class ImportJobTest < ActiveSupport::TestCase
     ObjectStore.instance.put_object(key:  import.file_key,
                                     path: csv_package)
 
-    format = ImportJob.perform_now(import: import)
+    format = ImportJob.perform_now(import: import,
+                                   user:   users(:southeast_admin))
     assert_equal Import::Format::CSV_PACKAGE, format
   end