pal-museum-metadata

Metadata validation and packaging tools for Merritt Ingest.

This code will be run from a Cloud9 environment into which the following resources have been loaded.

File structure

• an inventory listing of existing tif files residing in S3
  • /mrt/inventory/inventory.txt
• mods files describing the tif files
  • /mrt/mods
• temp dir for pulling tif file samples
  • /mrt/files
• code directory
  • /home/ec2-user/environment/code/pal-museum-metadata

Saving Changes to GitHub

This Cloud9 environment is shared by members of the UC3 team.

Therefore, it will be important to not save your github credentials into this working environement.

All of our code will live in a public repository, so it will be easy to pull code into this environment.

git fetch origin main

When you want to save changes back to GitHub, you have a few options

• push the changes from Cloud9 to GitHub
  • you will need to provide github credentials each time you push
  • because you have 2FA enabled (a good thing!), you will need to use a GitHub Personal Access Token to save your work
    • Create a Personal Access Token
    • Name it something like "Leading Fellows Token"
    • Set the expiration date for the end of the fellowship
    • Enable only "public_repo" for this token
    • Save the generated token in a safe place that will be easy to copy/paste
  • When you are prompted for a username and password
    • use your github username for username
    • use your personal access token as a password
• make the changes through the github website
• clone the repository to your PC and push the changes from there

git push origin main

Running the code

cd ~/environment
python code/pal-museum-metadata/src/scan.py 

Goals

• What becomes a Merritt Object
• What identifier(s) will be used
  • This will be used for any metadata updates
  • What if we get access to the database
• What metadata will be stored with the images
• What percent objects have / do not have images and metadata
• Create Merritt ingest manifest file(s) for each object
  • Has identifier(s)
  • Has erc descriptive metadata
  • Has full file list
    • Url to the mods files
      • Terry will build a web service to make these accessible to the ingest service (done)
    • Url to the image files
      • Terry will build a web service to make these accessible to the ingest process (done)

Tasks

Weeks 1-3 (does this include travel time?) - starting Aug 9

• Analyze match between files in the inventory vs identifiers in mods
  • List of matching image and metadata
  • List images missing metadata
  • List metadata missing images
• Recommend local identifier(s) to utilize likely some form of: 0001.02.0001
• Map mods fields to Merritt erc
• Hand generate a manifest file for a single Pal Museum object (urls depend on where mods and images are served)
  • Create ingest manifest for an object with one or more files; supply metadata through Merritt UI
• Load hand generated manifest to Merritt stage

Next steps

• Generate list of files per object identifier
• Create ingest manifests for objects with one or more files; create a manifest of manifests to supply corresponding metadata

Questions to answer

• Metadata
  • What is the format of our LocalId?
  • What mods metadata do we want to publish?
• What objects to publish?
  • Mods + Image - Yes
  • Mods only - ?
  • Image only - ? (if a valid id can be created)
  • Who consults on this decision?
• Additional data
  • What is in the new batch of data?
    • Images only?
    • Images and mods?
    • Replacement files?
  • Copy to S3
  • Modify program to pull new resources
• What is in the database?
  • Is there unique data in the database that is not already in mods?
  • Can we associate this data with an identifier?
• Manifest Generation (technical questions)
  • Where should the web server run for the image files?
    • images are in S3, served by docker01
    • mods are in S3, served by docker01
  • How will ERC metadata be associated with objects?
    • manifest of manifests?
    • erc files?
    • Terry will discuss this with Mark
  • How will a batch of individual manifests be published to a url for the ingest process?
    • presume we will copy these into S3. Additional rights will be needed to push to S3
    • Terry will discuss options