Copyright

Other than any prior works of which it is a derivative, the copyright in this work is owned by La Trobe University.

Licenses

Rights of use and distribution are granted under the terms of the GNU Affero General Public License version 3 (AGPL-3.0). You should find a copy of this license in the root of the repository.

Acknowledgements

La Trobe University Library is grateful to all who have contributed to this project.

Ruth Lewis provided the instructions for export and import of Alma data.

Contact

The maintainer of this repository is Hugh Rundle, who can be contacted at h.rundle@latrobe.edu.au

Description

This is a set of basic Python scripts for scraping book covers for Open Educational Resources for import into Ex Libris Alma.

The script takes an input file which should be an export from Alma converted to CSV format.

Run the script against a file exported from Alma with your OER URLs. It will produce an xlsx file with all the cover images it can find for the books in your list. You can then import this file back into Alma to update your local records.

Installation

This is a Python script. An easy way to run it is to use the uv package and project manager. Install uv according to their instructions.

Preparing your source file

The script operates on a report exported from Alma Analytics. To recreate it in Alma Analytics:

1. In Subject area E-Inventory:

   Select

   • Electronic Collection - Electronic Collection Public Name
   • Bibliographic Details - Title
   • Bibliographic Details - MMS Id
   • Portfolio URL Information - Portfolio Static URL
   • Portfolio URL Information - Portfolio Static URL (override)
   • Portfolio URL Information - Portfolio URL Type
   • Portfolio URL Information - Portfolio Parser Parameters
   • Electronic Collection URL Information - Electronic Collection Level URL

   Filters:

   Filter to your requirements, e.g.

   • Electronic Collection Public Name is equal to / is in CAUL OER Collective; La Trobe University eBureau; Milne Open Textbooks (Open SUNY Textbooks); OpenStax College; OAPEN
   • AND Availability is equal to / is in Available
   • AND Lifecycle is equal to / is in In Repository

or if you prefer SQL:

SELECT
0 s_0,
"E-Inventory"."Bibliographic Details"."MMS Id" s_1,
"E-Inventory"."Bibliographic Details"."Title" s_2,
"E-Inventory"."Electronic Collection URL Information"."Electronic Collection Level URL" s_3,
"E-Inventory"."Electronic Collection"."Electronic Collection Public Name" s_4,
"E-Inventory"."Portfolio URL Information"."Portfolio Parser Parameters" s_5,
"E-Inventory"."Portfolio URL Information"."Portfolio Static URL (override)" s_6,
"E-Inventory"."Portfolio URL Information"."Portfolio Static URL" s_7,
"E-Inventory"."Portfolio URL Information"."Portfolio URL Type" s_8
FROM "E-Inventory"
WHERE
(("Electronic Collection"."Electronic Collection Public Name" IN ('CAUL OER Collective', 'La Trobe University eBureau', 'Milne Open Textbooks (Open SUNY Textbooks)', 'OpenStax College', 'OAPEN')) AND ("Portfolio"."Availability" = 'Available') AND ("Portfolio"."Lifecycle" = 'In Repository'))
ORDER BY 5 ASC NULLS FIRST, 3 ASC NULLS FIRST, 2 ASC NULLS FIRST, 8 ASC NULLS FIRST, 7 ASC NULLS FIRST, 9 ASC NULLS FIRST, 4 ASC NULLS FIRST, 6 ASC NULLS FIRST
FETCH FIRST 10000001 ROWS ONLY

Running the script

Then run the script from a terminal/PowerShell:

uv run oer_covers.py inputfilename outputfilename

inputfilename should be your file exported from Alma and converted to csv. outputfilename should be a new .xlsx filename that will contain the output.

e.g.

uv run oer_covers.py alma_export.csv output_file.xlsx

You should end up with an Excel file with two tabs:

• covers_for_upload is for importing back into Alma (see below)
• errors will contain a list of any resources where some kind of HTTP or connection error occured. This may indicate that the URL you have in Alma is out of date, broken, or otherwise needs attention.

Importing the output file

Once you have your output file, you need to do something with it!

In Alma:

1. Click on Resources > Import.
2. Locate the Load cover images import profile.
3. Click on the ellipsis button and select Run.
4. Select the spreadsheet as the file to load.
5. Click Submit.

The job will add the URL for the cover image to the 956 field in matching bibliographic records. This may take a few hours to show up in your discovery interface.

Logging

The script will spit out any additional errors. You might like to send them to a log file rather than your terminal:

uv run oer_covers.py source_file.csv output_file.xlsx >> logfile.txt

Retrying errors

You may get errors for some links. These will be saved in the second errors tab in your output file.

You may wish to attempt these again, especially if you got a lot of 429 "Too many requests" errors. To do this, you can use the retry_errors.py file. Save the errors tab as a new csv file, then run retry_errors.py with this new CSV file as your input:

uv run retry_errors.py errors.csv output_file.xlsx >> logfile.txt

This script deliberately runs slower than oer_covers.py (to avoid 429 errors, which may be what you got originally) and expects your csv file to be structured like the errors tab in an output file. To upload the output from this run, follow the same procedure outlined in Importing the output file, above.

Image sources

Currently this will only work for books from:

• latrobe.edu.au (La Trobe eBureau)
• library.oapen.org (OAPEN)
• milneopentextbooks.org & milnepublishing.geneseo.edu (Milne Library Open Publishing, SUNY)
• oercollective.caul.edu.au (CAUL OER Collective)
• open.umn.edu (Open Textbook Library)
• jcu.pressbooks.pub (James Cook University)

You can add your own or log an Issue with a request for a new source. Note that Rice University inexplicably reserves all rights on OpenStax book covers, so they cannot be used, even though the rest of the book is CC licensed.