Brian Beck

It should include it verbatim, that part seems OK: in general the strategy is to leave the schema Markdown alone and let the Markdown renderer (whatever that ends up being)...

So the current Prettier settings in the `eslintrc` are supposed to be applied, but I assume what's happening is that the autoformat plugin uses the Prettier defaults instead, since it...

> Maybe, but that's news to me. I assume prettier v1 and v2 wants a .prettierrc file I meant that we don't run `prettier` itself currently at all. Only ESLint...

To my knowledge, this would break the rendering in Github Flavored Markdown, but I haven't tested it yet. Historically at least, GFM requires the newlines in order to treat the...

> To my knowledge, this would break the rendering in Github Flavored Markdown, but I haven't tested it yet. Looks like this is still the case – see this example:...

> Note that Markdown formatting syntax is not processed within block-level HTML tags. E.g., you can't use Markdown-style `*emphasis*` inside an HTML block. What should happen to the field &...

Maybe the most flexible would be to have a hook where they could supply a function to process the Markdown content however they'd like? Then the default could be what...

Here's a simple test I just tried out that makes me think this behavior doesn't make sense: * In a test, make an HTTP request and use `t.snapshot` to snapshot...

Here's a repro: https://github.com/exogen/test-builder-log It requires that the script logs a message of >8192 bytes (that's 2^13 if you're curious), then calls `process.exit()` instead of letting node exit normally.

I wonder if revisiting #20 and copying npm's task running mechanism would fix this as a more long-term solution?