Possibility to pull canonical files in other repos

[niklasmtj]

Yesterday @scottrigby had the idea of keeping our website in sync with changes in other repos in the GitHub org with canonical files. I wanted to create this issue to start a discussion about the possible implementation of such a feature.

From #64:

Ideally, the website would pull from canonical files in other repos in this GitHub org. For now, let's please just make sure any updates to this kind of content on the website has corresponding updates in the canonical source files.

That canonical source for the content from that page is currently here https://github.com/cncf/tag-app-delivery/tree/main/gitops-wg#community.

[christianh814]

+1

The website should pull the information from the other repos. Example, when v1.0.1 of the principles come out, the website "automagically" gets updated.

[niklasmtj]

Quick update: I started to check out possibilities to pull raw files from GitHub. As a PoC I will try to pull the manifest on build time which already works but I still have to check how I can inject it to the page and transform it into HTML.

We have to define the pages or files we would like to fetch this way. I still did not think about how to listen for changes in these repositories.

We can talk about this tomorrow (June 1st) in the meeting

[niklasmtj]

So I might found a way to pull in external sites. I built a small Gatsby plugin which will pull in the defined sources (in gatsby-config.js) in the raw markdown format and puts them in a defined place in the repository (e.g. content/glossary/GLOSSARY.mdx). Since we are using the gatsby-plugin-mdx these files are transformed into HTML.

What we have to discuss now is which pages we want to pull in this way since we probably need to adjust some of these files. For example when we want to pull in the get-involved page (as it is discussed in #62 ). Another question that my plugin right now does not include is the possibility to just pull in parts of a file like just a heading and it's content.

So my questions to you are now:

• Which pages do we want to pull from elsewhere and what is the URL?
• Do we need the possibility to pull in just parts of a file?