An extension of the marc-json-schema project by @thisismattmiller. This project is an XQuery port of the original Python code.

In addition to the MARC Bibliographic Format, the Authority and Holdings Formats are now represented as well.

Schema files are generated in Avram schema format:

• marc21_authority_schema.json
• marc21_bibliographic_schema.json
• marc21_holdings_schema.json

Contributions and bug reports are welcome!

• BaseX XML database and query processor
• jq command-line tool for JSON normalization
• Optional make for automatic execution via Unix command line

• Download and install BaseX either GUI or command line (see the BaseX wiki for detailed information).
• Install jq using a package manager such as apt or homebrew (depending on your environment).

To install BaseX into subdirectory basex from the command line:

wget -N https://files.basex.org/releases/BaseX.zip
unzip BaseX.zip

The GUI is more convenient for manual execution:

1. Launch the BaseX GUI
2. run-scraper.xq will fetch the HTML pages for the MARC standards documentation and save them in a BaseX database.
3. run-parser.xq will generate the JSON file and write it to a local directory.

Given BaseX is installed in subdirectory basex, the queries can be run in a single command:

./basex/bin/basex -c "RUN run-scraper.xq; RUN run-parser.xq"

Optionally, the value of a directory path can be passed on the command line to the ms:DIR variable:

./basex/bin/basex -Q run-scraper.xq -b ms:DIR="/Users/Abc/Desktop/" -Q run-parser.xq

Installation, scraping and parsing can also be run automatically by calling make:

make -B

• Tim Thompson (port and extension in XQuery)
• Matt Miller (original Python code)
• Jakob Voß (adjustments to comply with Avram specification)