Skip to content

How To Generate (Machine Readable) Glossaries

The generation of machine readable glossaries (MRGs) consists of finding the specifications of the terminologies for which such MRGs are to be created, and then generate them.

This results in the generated MRGs to become available in the mrgs-directory (as specified by the SAF). In our case, that would be

  • mrg.mve.mve-terms.yaml, which holds the MRG that contains all terms that are defined in our MVE, and
  • mrg.mve.yaml, which holds the MRG that contains all terms in the default terminology of our MVE-scope.

It so happens that these files have the same contents, but that need not always be the case.

# use this command in a GitHub Action or for local testing:
mrgt -c "tev2-config.yaml"
Note: Make sure you execute this command in the directory that contains the file `saf.yaml`. In our case, that would be `/docs`.

Notes:

  1. We can use multiple configuration files. Each configuration file has a particular purpose. Using those makes it easer to generate stuff in different contexts, such as the local (testing) context of the GitHub (deployment) context.

  2. You can override the parameters in the configuration file by providing them yourself on the command-line. The list of parameters is documented in the HRGT documentation.

  3. The directory in which the MRGs are created is the one as specified in the SAF. In our case, this is /docs/mrgs. This directory will be created if it doesn't yet exist. If this directory already contains an MRG with the same name(s), they will be overwritten with the newly generated contents.

  4. Additional documentation on the MRGT can be found in its specifications.