
A quick redex question regarding debugging: Suppose I have a judgment rule that has two premises and I am using judgment-holds
to test if the judgment holds for some input and/or output.
Is there an easy way (I mean, easier than using current-traced-metafunctions
, or escaping to racket eprintf
) to tell which premise isn’t met when the judgment doesn’t hold?

I’ll refrain from making commits that only clean up leading or trailing whitespace. I did make that one commit for indentation in check-syntax.rkt
, because it’s harder to preserve unusual indentation when making non-trivial changes? And I only did that because the top-level indentation varied. There are Racket sources that got converted from module expressions to #lang
style, and everything is still indented a couple spaces — but consistently. I wouldn’t even touch those.

I don’t want to make a big deal of it. But the point is that the history teacking of some particular lines is easier without spurious changes. Preserving whitespace on lines you are modifying does not help with that goal.

So please don’t waste time trying to do that.

Please feel free to clean up whitespace if it helps you work with the file.

Does this make sense now?

Yes. I totally agree. I agree so much I was just trying to explain why I made that one exception. :smile:

Ah! Got it! Thanks.

I always use printf :/

Weirdness with downloading packages for Racket 6.2 and 6.2.1? I have monthly “cron” builds set on Travis CI for some packages like aws, frog, and rackjure. Even if nothing has changed, it builds/tests. Makes sure e.g. deps haven’t broken. The c. Feb 26 builds were all fine. But today, all three broke: - Only for Racket 6.2 or 6.2.1. - Downloading a pkg is a 302 error. - But not the same pkg among all three. Examples: https://travis-ci.org/github/greghendershott/frog/jobs/667319537#L238-L241 https://travis-ci.org/github/greghendershott/rackjure/jobs/667353768#L232-L235 https://travis-ci.org/github/greghendershott/aws/jobs/667319118#L232-L235
I can’t guess why or what’s going on. Maybe a GitHub problem? Has anyone else noticed anything like this? Anything changed in the last four weeks with pkg catalog?

Have you tried fetching those URLs with curl locally?

@ryanzidago has joined the channel