Clojurians Log v2

in case anyone noticed but didn't say anything, https://build.clojure.org's ssl cert expired a few days ago but is now fixed

but no one said anything, so I just assume no one noticed :)

@alexmiller While you're around... are there any further things that will go into 1.10.2? It's been sat at alpha 1 for over a month so I was wondering if something was blocking a "gold" release.

maybe

I've got a handful of jiras in the consideration bucket

we seemed to have unblocked the graalaverse so seems less urgent now :)

Ah, OK. And just to satisfy my curiosity, if I test against 1.11.0-master-SNAPSHOT, that is the same as 1.10.2-alpha1, yes?

I think it's actually older at this point?

yeah, 1.10.2-master-SNAPSHOT is the head right now

but that should be the same as 1.10.2-alpha1

1.11.0-master-SNAPSHOT I think should be same as 1.10.1 right now

Ah, of course. Silly me.

well, that's all admittedly confusing :)

we did look at a few tickets that were found .... wanting, so didn't make it into an alpha2

I'm just eager to start testing against new stuff as it drops on master 🙂

I would really like to get https://clojure.atlassian.net/browse/CLJ-2469 fixed

mostly b/c I broke it :)

and https://clojure.atlassian.net/browse/CLJ-2295 is tiny but one that's an old regression

and https://clojure.atlassian.net/browse/CLJ-2529 is something I've spent a bunch of time on and actually see every day because of how Cursive works and is super frustrating to me, but probably has too many possible ramifications to be worth considering in 1.10.x so will probably wait

and there's a few others that are maybe a little farther away but could be prioritized

Hmm, I think I've run into CLJ-2529 with Chlorine sometimes too.

Cursive wraps every execution so it happens on ALL exceptions reported in Cursive repl

it reports every error from inline execution as a syntax error

the history on the code for that one is long and features repeated misunderstandings and bandages on top of the actual problems that we addressed directly in 1.10 for errors. The patch is really about undoing some of that stuff and letting what we already did work. I could probably do a full talk about that ticket.

which like 3 people would appreciate

"unblock the graalaverse!" sounds like a line spoken by a villain in a B movie 🙂

(Come on, people, if we can get 30 reactions to that comment, maybe he will give the talk 🙂

I for one would love to hear it!

I think that that shouldn’t be a problem any more with Cursive if you’re working with a 1.10+ project though, right @alexmiller?

I think I've maxed out the audience here

That should be setting the metadata and not loading any more.

@cfleming I'm typically like 9 releases behind in my tools, so I may just not have the right bits yet

Is this the one time where @channel would have helped?

I'm on IDEA 2018.2.5 and Cursive 1.8.0-2018.2

"old school"

Yeah, I think that’s fixed in 1.9.0.

That is indeed the classic edition.

I fear change

At least you’re now motivated to fix the Clojure bug.

yeah

iirc one of the earlier related bugs was the source of a lot of CIDER angst, so I'm worried about "fixing" it without making sure we haven't violated some assumption there

Yeah, I have a link to a bunch of CIDER/nREPL issues about how that worked, and differences between how load worked in CLJ and CLJS

IIRC in CLJS it doesn’t return the value of the last form evaluated in the file, and in CLJ it does.

So that workaround doesn’t work for CLJS.

(so glad Chlorine doesn't rely on much beyond what's actually in Clojure itself!)