Clojurians
#cljs-dev
<
2017-01-25
>

This page is not created by, affiliated with, or supported by Slack Technologies, Inc.

richiardiandrea18:01:26

I added a comment on http://dev.clojure.org/jira/browse/CLJS-1792, it looks like the patch would be trivial, but it surely has something that I am overseeing :smile:

danielcompton20:01:52

Is there any interest in http://dev.clojure.org/jira/browse/CLJS-1841? It would improve the onboarding flow for new contributors getting setup

dnolen20:01:02

@danielcompton I applied and it an immediately ran into some problem

dnolen20:01:09

sadly I didn’t record what went wrong :disappointed:

dnolen20:01:34

somebody else should give that patch a try and leave feedback

danielcompton20:01:50

I'm just checking itnow

danielcompton20:01:02

I'll give it a workout against master

dnolen21:01:16

if you have time might want to give master a spin - I just landed this https://github.com/clojure/clojurescript/commit/94b4e9cdc845c1345d28f8e1a339189bd3de6971

dnolen21:01:22

long outstanding issue

anmonteiro21:01:55

IIRC there were some other issues that were blocked because of that one?

anmonteiro21:01:24

that can now be worked on? or am I confusing with something else?

dnolen21:01:42

hrm I don’t recall - but maybe that’s true?

mfikes21:01:12

@dnolen Something in that commit is breaking Planck… will try to sort out what it might be

anmonteiro21:01:23

interesting, trying it out in Lumo

dnolen21:01:01

well the signature of load-libs changed

dnolen21:01:13

so if you were depending on that you need to adjust

dnolen21:01:18

one more argument

mfikes21:01:53

OK… (for Planck it is breaking the build phase where it AOT-compiles its macro namespaces), but I’ll check for any load-libs calls.

dnolen21:01:49

@mfikes interesting - though that seems really weird

mfikes21:01:00

Yeah… it could be anything

anmonteiro21:01:28

@mfikes is it that specific commit?

anmonteiro21:01:41

or could it be related to the recent spec fixes?

anmonteiro21:01:46

(since you AOT clojure.spec macros)

mfikes21:01:15

I’m pretty sure it is that commit… I’ve been very closely following master as of late and it was good with the previous commit

anmonteiro21:01:54

sure, just confirming

anmonteiro21:01:00

I haven't been following it too closely

mfikes21:01:09

FWIW, script/test-self-parity fails with something related to ns processing:

#error {:message "Could not parse ns form cljs.clojure-alias-test", :data {:tag :cljs/analysis-error}, :cause #object[Error Error: No protocol method IDeref.-deref defined for type null: ]}

dnolen21:01:23

@mfikes hrm

mfikes21:01:57

Yeah… not too helpful… seeing if I can see how that change might affect self-hosted

dnolen21:01:43

actually something is wrong in that commit

dnolen21:01:52

line 1041 is suspicous

dnolen21:01:19

@mfikes yep, try master

dnolen21:01:26

the last commit caused several other problems

mfikes22:01:40

@dnolen now with master, Planck’s tests pass. I still get the same failure with script/test-self-parity. Perhaps there is a weakness in the test. I’ll take a look at it tonight (heading out for dinner)

anmonteiro22:01:44

self parity tests still fail for me with current master

anmonteiro22:01:07

but Lumo compiles correctly, which it didn't with HEAD1~

mfikes22:01:33

Cool. Perhaps script/test-self-parity is wrong in some way.

anmonteiro22:01:44

scratch that, Lumo doesn't compile with current master

anmonteiro22:01:04

gets stuck while AOTing macros

anmonteiro22:01:30

I don't have time to look into it now, but I suspect that if we get the self parity tests to pass everything will be OK downstream in Planck & Lumo

mfikes22:01:02

Yeah, for whatever reason, Planck is now OK.

dnolen22:01:40

ok yeah we should try to sort out the bootstrap issues before Friday

dnolen22:01:46

but I’m pretty sure the require order patch is OK now

angusiguess22:01:28

Oops, should have run all the suites, sorry about that.

danielcompton23:01:02

I've updated http://dev.clojure.org/jira/browse/CLJS-1841, ready for someone else to take a look