update-collection-handling #451
Conversation
| @@ -41,6 +41,7 @@ def build_metadata | |||
| self.parsed_metadata = {} | |||
| add_identifier | |||
| add_ingested_metadata | |||
| add_collections | |||
There was a problem hiding this comment.
@sephirothkod I don't think this is a good choice for bulkrax because this line seems to have been removed intentionally for some reason. however, without it self.collection_ids is always empty so collections_created? will return false and throw a CollectionCreatedError. even when the collection is in fact created.
There was a problem hiding this comment.
@sephirothkod @bkiahstroud actually...I now needed to bring this code into LV and BL through overrides. may it be better to put this code into bulkrax proper, but with a note to reevaluate removing add_collections from here in the future once the updated logic is worked out?
There was a problem hiding this comment.
sure lets leave it for now and then I'll have a look when i have time as to how to remove the collections stuff entirely and only reference collections via the new parents code.
| parent_field_mapping = self.class.parent_field(parser) | ||
| return [] unless parent_field_mapping.present? && record[parent_field_mapping].present? | ||
|
|
||
| identifiers = [] | ||
| split_titles = record[collection_field_mapping].split(/\s*[;|]\s*/) | ||
| split_titles.each do |c_title| | ||
| matching_collection_entries = importerexporter.entries.select { |e| e.raw_metadata['title'] == c_title } | ||
| split_references = record[parent_field_mapping].split(/\s*[;|]\s*/) | ||
| split_references.each do |c_reference| |
There was a problem hiding this comment.
more semantic names.
app/models/bulkrax/csv_entry.rb
Outdated
| matching_collection_entries = importerexporter.entries.select { |e| e.raw_metadata['title'] == c_title } | ||
| split_references = record[parent_field_mapping].split(/\s*[;|]\s*/) | ||
| split_references.each do |c_reference| | ||
| matching_collection_entries = importerexporter.entries.select { |e| e.raw_metadata['source_identifier'] == c_reference && e.is_a?(CsvCollectionEntry) } |
There was a problem hiding this comment.
we're no longer using titles to track collections. we also are making sure that we only return collection parents and not work parents so that collections_created? isn't incorrectly throwing false.
alishaevn
changed the title
TODO
discuss with braydon
expected behavior