Create CSVs with existing checksum values for QC purposes

[kmichaelis]

Format existing checksums in a way that will make post-migration comparison straightforward.

If possible, create a CSV that lists the hashes for all disk images and supplemental files in a container. Upload that CSV to the container the validation function can be run against during post-migration QC.

[rotated8]

I think this issue is resolved? We have the content_md5 column in the disk_images_all_fields_report_04_15_26 CSV. Is there other work necessary?

[kmichaelis]

@rotated8 I believe that @katfish2 wanted a file with the hashes for each object's files to be stored in the object's container. Otherwise the comparison process will be complicated. @katfish2 please confirm/correct that.

[katfish2]

@kmichaelis is correct—we need a separate list of files with md5s for each collection in order to validate the hashes within LIBSAFE Go. Is it possible to create another version of the script that's parsing descriptive metadata fields by collection to make the collection-level import CSV? Maybe this one could follow the same logic to split disk_images_all_fields_report_04_15_26 into separate CSVs by collection_source_id but retain only the pid, title, and content_md5 fields for each row.

[rotated8]

Can you tell me more about the workflow here? My impression from the meeting with Cari was that this was a manual process.

[kmichaelis]

As long as the md5 column exists in the collection-level CSV, @katfish2 can make it work. Need one list of all md5s. Can include md5 in descriptive MD.