BRS spec fix
If you look at https://unicode-org.github.io/cldr/ldml/tr35.html and others, the formatting is completely mucked up, showing raw html. See the end of this ticket.
Jan figured out that Jekyll, which is powering GitHub pages, uses a different markdown engine (kramdown) than GitHub, but it can be changed to the GitHub one by changing or adding a _config.yml file in the docs directory:
Since we don't have one to modify, we need to add the minimal file.
(We can change the name later.)
Looks like that does the trick.
I just approved and merged the 3rd try, to see if that works.
Tom, the problem appears to be that the yml file is in the wrong place. See my notes on the PR.
Steven, as far as I can tell, there is no good reason to have a different markdown engine for https://unicode-org.github.io/cldr/ldml/tr35.html than we do for .
Any differences will mean that we can’t predict from what https://unicode-org.github.io/cldr/ldml/tr35.html will look like.
So if Jan’s solution works (and it should, given the instructions on https://docs.github.com/en/github/working-with-github-pages/setting-a-markdown-processor-for-your-github-pages-site-using-jekyll), then we should use the same markdown engine for both. The instructions should do that without creating a workflow and having special scripts. We can do that later if we find a need, but right now we need to have a workable alpha.
OK, a more specifc link : https://unicode-org.github.io/cldr/ldml/tr35.html#Unicode_language_identifier
It’s not completely mucked up, only where we have raw HTML tables. This confused me when I saw this PR go by.
So by the way, we don’t have to use the ‘automatic’ GH processing. We can create a workflow (see for example <https://github.blog/2020-09-25-github-action-hero-james-ives-and-github-pages-deploy/>) and have changes to master automatically run whatever-scripts-we-want-to-run and deploy the resultant HTML to gh-pages. GitHub’s ‘automatic’ processing basically does this, w/ Jekyll.
What a minute. Huh? This has been working for me since - did something regress?