Boston College ArchivesSpace Batch DAO Process

Object

To automatically generate description in ArchivesSpace for digitized archival materials, and to re-use that description to ingest files into a Digital Libraries repository.

Prerequisites

Install the following project dependencies:

pip install requests
pip install python-dotenv
pip install progressbar2

Copy sample.env to .env and include your ArchivesSpace credentials.

Using virtual env

It is recommended to use a python virtual environment like venv.

Here is a sample use case:

pip install virtualenv
python3 -m venv env
source env/bin/activate

Steps:

Select for digitization

When material is selected for digitization, it is identified in ArchivesSpace by the addition of a Component Unique Identifier (CUI) at the level of digitization (usually but not always the File level). The format of the Component Unique Identifier is as follows: the collection identifier reformatted as LLYYYY_NNN_ followed by the unique database ID number of the object in question. This is present in the ASpace URL when you are viewing the object as "tree::archival_object_NNNN" - we are only interested in the number at the end of the URL, and it may consist of any number of digits. As an example, a properly formatted Component Unique Identifier may look like this: MS2013_043_54063.

Export EAD

Once all objects selected for digitization within a given collection have been marked with CUIs, export the EAD for the entire collection with only the "include tags" option checked.

Transform EAD

Transform the collection EAD into two different tab delimited files, using aspace_ead_to_tab.xsl and scanners_tsv.xsl.

• The output from aspace_ead_to_tab.xsl will be used to run aspace_batch_dao.py once digitization is complete. The output of this transofrmation can be named tab_file.tsv.
• The output from scanners_tsv.xsl should be sent to the digitization staff as a tracking worksheet for digitization. The output of this transformation can be named scanner-worsksheet.tsv.

Generate FITS and transform into a JSON file

When the digitization is complete, run FITS over the image files. Run the fits-to-json.xsl XSL over the FITS xml file, to create a JSON file of technical metadata. The output of this transformation can be named fits-file.json.

Run batch DAO script

Run aspace_batch_dao.py from the command line with the following usage:

python aspace_batch_dao.py {DEV|STAGE|PROD} ead_to_tab.tsv fits-file.json

• select one of {DEV|STAGE|PROD} for which ASpace server to update
• ead_to_tab.tsv is the output of aspace_ead_to_tab.xsl
• fits-file.json is the output of the FITS transformation

Optionally, you can include the --dryrun flag to run the script without editing or creating any new records.

This script will call on the ArchivesSpace API to create a Digital Object, and one or more Digital Object Components for each object and the image files that represent it. If there are errors, the object metadata in ArchivesSpace or various aspects of the python script may need editing.

Batch script METS output

After running aspace_batch_dao.py, there will now be a folder in the directory the script was run from named METS, which contains METS metadata files for each DAO created. See the BC wiki for documentation on how to use these files in repository ingest. METS output has been deactivated in the aspace_batch_dao.py script.