Bazel Catalog

See the live site: https://bazel-contrib.github.io/bazel-catalog

Developing

This is written in Bash and JQ.

To work in this repository, it helps to understand the structure and dataflows.

1. Ingest data from sources into this repo to produce a "sync" PR.

   • From BCR using scripts/merge_bcr_metadata.sh. This is intended to be non-destructive.
   • From GitHub API using scripts/collect_github_stats.sh

2. git data produced by collect_gitstats.sh is slow to collect and has a giant output, which goes into the gitstats folder in the gh-pages branch to avoid bloating the main branch.

GitHub Actions will then take over automatically, running the render.yml workflow which:

1. Runs scripts/render_ghpages.sh to produce the rest of the content in the gh-pages branch.
   • This script runs filters/*_data.jq to get some datapoints which we render in markdown
2. Pushes the gh-pages branch
3. A separate GHA workflow updates the live GitHub pages site after the gh-pages branch changes.

Still needed:

• Cron-style automation for data ingest.