-
Notifications
You must be signed in to change notification settings - Fork 0
Ingesting Image Collections
Upload files and MODS to the AWS S3 input bucket, 704869648062-input, via the AWS S3 Management Console.
Image folder should be named like the prefix of the digital IDs, e.g., PC-RM-Stereos; PC-008
MODS folder should be named similarly, but with ".MODS" appended to the end, e.g., PC-RM-Stereos.MODS; PC-008.MODS
Once files are uploaded to the input bucket, bags are automatically created in the output bucket, 704869648062-output.
Review files in the output bucket.
SSH into the Islandora server us-west-2.pem, navigate to /data/Collections.BATCH. Run archive script to transfer collection to archive bucket, 704869648062-archive.
cd /data/Collections.BATCH
./archive-collection.sh -c <collection_name.bags>$ ./archive-collection.sh -c <collection_name.bags>
In Islandora, create a new collection (usually at the top-level of the repository): http://digitallibrary.californiahistoricalsociety.org/islandora
Go to Manage, then click + Add an object to this Collection
Complete Collection PID as namespace:collection, where namespace = chs. For example, for the California Agriculture collection, enter chs:agriculture.
Uncheck "Inherit collection policy?" and select the Allowable content models you which to set for this collection. Usually it will be "islandora:sp_large_image_cmodel ~ Islandora Large Image Content Model"
Click Next, then Next again.
Enter metadata for collection and save. Make a note of the PID of this new collection. You'll need that when creating an ingest batch.
We're using Islandora Context to assign metadata display profiles based on collection type. For example, our Maps collection uses a different display profile than do our photograph collections.
Go to Structure > Context, then edit the Display Photos context.
Under Conditions, click on Islandora Collection Membership and select the new collection from the list. Save.
SSH into the Islandora server us-west-2.pem, navigate to /data/Collections.BATCH.
Run script to get collection from archive bucket to Collections.BATCH directory.
cd /data/Collections.BATCH
./get-collection.sh -c <collection_name.bags>Run the ./create-batch.sh script
$ ./get-collection.sh -c <collection_name.bags>
$ ./create-batch.sh -i <collection_name.bags> -c chs:<collection_pid> -m islandora:sp_large_image_cmodel
In Islandora go to Reports > Islandora Batch Ingest Sets (http://digitallibrary.californiahistoricalsociety.org/admin/reports/islandora_batch_sets)
Identify your new batch set and click "View items in set" at right
Click + Process Set, then Start Batch Processing
Once assets have been ingested, find a good representative thumbnail image for the collection by going to an individual image's datastreams and downloading the JPG. Resize the JPG to 293px wide. Go to the collection, click Manage > Datastreams, and replace the TN datastream with the newly resized image. Remember to acquire the lock.
Go to new collection, Manage > Collection > Regenerate child DC metadata > Regenerate DC metadata for all collection objects
Once this process is complete, delete the collections from the input and output buckets in AWS S3. Likewise, delete collections bags and batches from the Collections.BATCH directory on the Islandora server.