I hope to build Vectara into a docs-as-code documentation website (docs.qumulo.com). Two+ years ago, I began to build this website by taking the template from my former colleague Tom Johnson and customizing it in dozens of different ways.
Currently, the codebase lives in a private GitHub repo. Only the docs admin (me) has the permission to merge to
mainline, whereupon GitHub Actions mirrors the entire repo (only
mainline, to avoid leaking features) to a public repo. I hope to use the
.md files in the public repo.
The problem is, I’m a little overwhelmed by all that Vectara has to offer and I’m not quite sure as to where and how to begin. If my understanding is correct, I need to find a way to feed docs updates to the Vectara API as a corpus as part of the same GitHub Actions process that pushes the private repo to the public repo. Then, I need to configure how Vectara handles the data. Finally, I need to use the Vectara API to enable search from the site.
Is my understanding correct? I’d appreciate any pointers for a beginner with this tool.