Adding new objects to Credo

This page covers the process of adding new objects to Credo, starting from the point when scanning and metadata is complete. Before we outline the process, here are some general tips:

1. Add objects to Credo in manageable batches: keep the number of objects you batch ingest to a reasonable amount (100s rather than 1000s). This makes uploading and troubleshooting much easier. For example, if you are digitizing a multi-box collection, add new items per box as each box is completed.
2. Check error report regularly: Credo produces an error report every morning at 6:00am. Take a look in the morning after you've added files to Credo to make sure they all went in ok. The error report can be found here: http://credo.library.umass.edu/errorreport.html

There are two different tools that can be used to prepare metadata for Credo, a command line interface (CLI) and a desktop application. The desktop application is the simplest and recommended method. The CLI is considerably more powerful and can accomplish a bunch of other metadata maintenance tasks. The full documentation for the CLI lives at its own page.

To set it up:

Windows

1) Go to r:/credo/metadata/utilities/mods_UI/Windows and copy the entire folder named 'MODS' to somewhere in your profile, e.g. your Documents folder.

2) Enter your local copy, right-click on the MODS.exe file, select “Send to” and select Desktop. This should create a shortcut to your local version that you can access on your Desktop.

3) Launch the app by double-clicking the shortcut.

Mac

1) Go to r:/credo/metadata/utilities/mods_UI/Mac and copy MODS.app to your local Applications folder (or anywhere locally on your Mac).

2) Launch the application by double-clicking MODS.app.

To use the app, press the “Open” button to browse to a folder you want to check. Once the folder is selected (the full path should appear in the bar at the top), press “Run” to check the files. If the files have no errors, the app will automatically add IDs, so no need to do an extra step. If there are errors, the app will list the files and errors. Next to each error is an open button which will open that file in your default application so you can fix. Once fixed, select “Rerun” and hope for the best!

The QA application must be run on every MODS file before it is ingested into Credo. It's also probably wise to run again if items have been edited.

The CLI program is stored in r:/credo/metadata/utilities/mods/mods and called mods.exe. It is easiest if you put a ticket into LTS to have this script moved to a folder in your PATH. They will know what this means. Unfortunately, this can only be done by someone with admin privileges to your computer.

This script can do a whole bunch of mildly useful things but for the purpose of preparing metadata, it:

1. Checks for a number of critical errors in the metadata, including malformed dates, double-quotes in titles, or validity issues.
2. Adds IDs and URLs to records

The script must be run on every item before it is ingested into Credo. It's also probably wise to run again if items have been edited.

To prepare a batch of files for ingest you will need to run two separate commands:

$ mods -c [folder]

This will check all files in a specified folder for errors. If the script returns to the prompt without displaying any information, the files in the folder are ok.

$ mods -Iwu [folder]

This will add the IDs and URLs to the MODS records. Once both of these commands have been run, the files are ready for Credo.

Once the files are prepared, they need to be uploaded to Credo.

1. Log into the Credo server via SFTP. The server name is credo.library.umass.edu, the username is fedora, and Aaron R. will give you the password.
2. Upload the metadata and object files to the appropriate folder. If you are adding to an existing collection, add the files to the pre-existing folder. If you are creating a new collection, add a top-level collection folder, upload the files there, and follow the steps to add a new collection to Credo.
3. Add a copy of all the metadata files for the objects you would like to ingest into the 'ingest' folder. Ingest runs once an hour and will only add 100 objects per hour. You will know an object has been ingested if the file disappears from the ingest folder.

It's generally easiest to drag actual metadata files into the ingest folder but it's worth knowing that the ingest process doesn't care, as long as there is a file with the object PID and the extension '.xml'. The file can be empty or something completely different. Credo trivia!

1. Generate a .vtt file of the captions for the video you wish to add captions for.
2. Name the file with the same PID as the corresponding video: [PID-reference.vtt, eg. mums1093-s10c-i005.reference.vtt]
3. Upload the .vtt file along with the access and archival files and the xml. So you should have a total of 4 files for each video:

• mums1093-s10c0-i005-001.access.mp4
• mums1093-s10c0-i005-001.archival.mov
• mums1093-s10c0-i005.reference.vtt
• mums1093-s10c0-i005.xml

1. Upload the PDF that corresponds with the item
2. Name the file thusly: [PID.reference.pdf, eg. mums1156-s01-i001.reference.pdf]
3. Upload PDFs to appropriate folder
4. Ingest (or re-ingest) the corresponding XML records