Ah, just discovered that @standard.site is recommending to validate with site-validator.fly.dev instead. And there the rkeys aren't flagged 🥳
But! there is a warning about the trailing slash in pub.url and no leading slash in doc.path But shouldn't those be what the website is serving?
(3/4)
Soon we can drop ts-blank-space from Mastro 🥳
Deno now also supports `import { stripTypeScriptTypes } from 'node:module'` – only Bun left that doesn't have this built-in.
Mastro uses it to strip types from *.client.ts files for the browser: mastrojs.github.io/docs/install...
Got something working! github.com/mastrojs/atp...
While it works in practice (I do get the shiny button on Bluesky"), I just discovered that the schema validation says the document rkeys should be TIDs, although it doesn't seem to mention that on @standard.site? 🧐 I hope the schema is wrong…
(1/2)
A friendly tool to check if a post has valid standard.site records
Like: mastrojs.github.io/blog redirects to mastrojs.github.io/blog/, which is the canonical for my blog. So I want that in publication.url, right?!
(4/4)
Because otherwise that would throw a wrench into the approach of deriving the rkey from the URL path, which would have allowed us to not having to update the blog post markdown files.
pdsls.dev/at://did:plc...
Just discovered that curl has a --json flag. Instead of:
curl-X POST -H 'Content-Type: application/json' -d '{...}' ...
you can write
curl --json '{...}' ...
Working on packaging up the standard.site functionality in a little script. Similar in spirit to e.g. github.com/mastrojs/feed
I think I've figured out a nice API based on deriving the rkey from URL slugs. Now need to validate it actually works 😅 Collaborators welcome!
