Problem: Archivematica defaults to application/octet-stream when individual files are requested but it would be good to return the mimetype of contents of AIPs/DIPs (ArchivesDirect)

[kimpham54]

Please describe the problem you'd like to be solved.

We have an instance of ArchivesDirect and are accessing the DIPs that are being stored in Duracloud via our digital collections website. Having discussed this issue with @sallain, "Archivematica is setting the MIME type as application/octet-stream in the API store request itself. No attempt is made to generate the MIME type for a particular file."

Describe the solution you'd like to see implemented.

Archivematica passes on the mimetype for content in AIPs and DIPs so that it can be accessed in the Duracloud API header

Describe alternatives you've considered.

We are currently getting content from Duracloud and identifying mimetypes using a tool built into linux

Additional context

---

For Artefactual use: Please make sure these steps are taken before moving this issue from Review to Verified in Waffle:

• All PRs related to this issue are properly linked 👍
• All PRs related to this issue have been merged 👍
• Test plan for this issue has been implemented and passed 👍
• Documentation regarding this issue has been written and it has been added to the release notes, if needed 👍

[kimpham54]

Let me know how can I help! If you want to set up the development environment, this is is the right place to start: https://github.com/artefactual-labs/am/tree/master/compose. We also have a guide for contributors: https://github.com/artefactual/archivematica/blob/stable/1.8.x/CONTRIBUTING.md. My colleague Ross wrote this new walk-through that you may find useful: https://gist.github.com/ross-spencer/f7df0ad9e327555691485648002872e6.

The upload that you're looking at happens in Storage Service. In particular, duracloud.py which is a Space. When the AIP is compressed, I guess you'd want to identify the MIME type of the archive. IIRC, I think that would be easy since we know in advance the compression algorithm used and it's stored. When uncompressed, your best chance is to look up the MIME types of individual files inside the METS file of the AIP. My understanding is that you are interested on the latter use case. We use mets-reader-writer (https://github.com/artefactual-labs/mets-reader-writer) to read the METS, it makes very easy to extract the metadata of the files including all the things related to PREMIS.

[ross-spencer]

Hi @kimpham54, cc. @sallain and @sromkey in case it's useful for reference, I've done a little work to teach the storage service how to use file in this PR here to help identify AIPs without pointer files.

Specifically file is used in these lines.

The new module is in the common set of modules belonging to the storage service so that might provide some guidance where you could add new capability.

The way we extract compression types from pointer files is also moved in this PR to the new module.

So it might be a good start to get some of the information you require. You'd still need to get the information into the response header but that might not be too much work.

Did I read correctly that you wanted to get information about the other objects in the packages too?? That might be another fascinating problem!! But interested to see how a solution evolves, and echoing the response above, let me know if I can help at all.