Max Loeb

Just curious--if you're really worried about folks snooping your schema, why not just close all arbitrary-query access to the endpoint entirely and use persisted queries?

I'm having some trouble negotiating with your CI build. I'm getting this error: ``` Time: 00:03.819, Memory: 80.00 MB There was 1 PHPUnit test runner deprecation: 1) Your XML configuration...

> Slevomat CS is a heavy user of phpdoc-parser so we're testing every phpdoc-parser commit against Slevomat CS, to see that we didn't break anything. Oh, that's clever, so you...

An, yes, I had the tests commented out while I was experimenting with the node-based comment approach. Restored, thank you.

> You should probably replicate what the workflow does locally, which will allow you to debug what's going wrong in the test. Great idea, thanks.

Heh, well hurray for debugging. It turns out that the comment-parsing I was doing was getting thrown off by this kind of thing: ``` * @see https://www.slevomat.cz ``` I've modified...

Fair point, but it was actually fairly easy to get working as all the hard work happens in `Comment::getReformattedText`, which I was able to lift unchanged from nickic. We'll see,...

With my latest changes, multiline comments seem to work with format-preserving printer: ```php /** * @return array{ * c: string, * /* This is a * multiline * comment *...

Oh, I also reverted the trailing comment style: ```php /** @return array{ a: int // comment } ``` Nikic parser doesn't seem to support it, and I suppose it could...

Took a shot at implementing delayed add. At this point I think I'm ready for a round of reviews. I'm sure there are going to be a lot of use...