KislakCenter / VisColl

Modeling and visualizing physical manuscript collation
https://viscoll.org/
MIT License
50 stars 9 forks source link

Image mapping not available after IIIF manifest import #10

Open Ennego opened 1 year ago

Ennego commented 1 year ago

Hello! I'm having an issue mapping images from a IIIF manifest. After importing the manifest, going to the "map images" screen, and selecting "uploaded images," I do not see the manifest images as an option in the backlog box. Any suggestions?

Here is the IIIF manifest: https://figgy.princeton.edu/concern/scanned_resources/ac79d75c-1d0f-4605-b006-8b14785adb84/manifest

leoba commented 1 year ago

Confirming that this is an issue. I created a test project and uploaded both this manifest and one from Penn's Colenda system. Both of them show up as being imported, but only the Colenda manifest is available for mapping in the drop-down (see screnshots)

Screen Shot 2023-09-20 at 9 35 37 AM Screen Shot 2023-09-20 at 9 36 19 AM
Ennego commented 1 year ago

The IIIF Presentation API validator does give some warnings for this Princeton manifest (below), although it is valid presentation 2.1. Could any of these warnings be the cause of the issue for VCEditor?

tpendragon commented 1 year ago

@Ennego @leoba Hi! Developer from Princeton here. It looks like your code doesn't support multiple manifest labels. I'm seeing this error in the console:

EditManifest.js:74 Uncaught TypeError: r.state.name.trim is not a function

When I put a break point there I get this:

r.state.name
['Missal, use of Seville.']

It's valid for a manifest to have multiple labels - it looks like your code should probably wrap label to an array and pick the first one?

Presentation API:

A Manifest must have the label property with at least one entry.

Thanks!

leoba commented 1 year ago

Thank you @tpendragon !

@demery is this something we can take care of now or should I add it to my wish list for the grant proposal?

Ennego commented 1 year ago

@demery just checking back on @leoba's question. I would like to use the Princeton manifest to map the images to a collation model the fast &easy way if possible!