Clojurians Log v2

my compile error just popped up again, three files are affected, all missing just the namespace in one function def:

https://gist.github.com/mk/6d921f6b0c5a14e0463a4b425b830f91

it’s just missing the namespace, so somewhere swallowing com.nextjournal.journal.views.

e.g. compare the generated js from https://gist.github.com/mk/6d921f6b0c5a14e0463a4b425b830f91#file-billing-js-L344 with the clojurescript source https://gist.github.com/mk/6d921f6b0c5a14e0463a4b425b830f91#file-billing-cljc-L376

or https://gist.github.com/mk/6d921f6b0c5a14e0463a4b425b830f91#file-settings-js-L59 with https://gist.github.com/mk/6d921f6b0c5a14e0463a4b425b830f91#file-settings-cljs-L36

@mkvlr that looks very much like a threading issue. like 2 threads writing the same file. do you use :parallel-build at all?

return cljs.core.reset_BANG_.call(null,notifications_popover_ifications_form,current_setting,map__134483,map__134483__$1,runner_log))

one extra ) at the end. but also name overlap before, as if the output slightly shifted because some other thread got a few bytes in before

the first part is this call https://gist.github.com/mk/6d921f6b0c5a14e0463a4b425b830f91#file-status-cljs-L311

then suddenly some other bits appear that don't belong there

I would suspect threads. either just actually writing to the same file or just messing with *out* somehow (unlikely given its a thread-local but who knows)

the others look like just the ns is missing but I suspect there might be more errors in the files. it just stops parsing at the first one. did you try manually adding the ns and see how far it gets?

also note

return (new com.nextjournal.journal.views.billing.t_com$nextjournal$journal$views$billing135573(null));
})()
,new cljs.core.Keyword(null,"initial-data","initial-data",-1315709804),null,new cljs.core.Keyword(null,"options","options",99638489),new cljs.core.PersistentArrayMap(null, 1, [new cljs.core.Keyword(null,"watch-atom","watch-atom",-2134031308),false], null)], null));
})], null));

//# sourceMappingURL=billing.js.map?rel=1573720735583
65");
});

/**
 * Positional factory function for com.nextjournal.journal.views.billing/t_com$nextjournal$journal$views$billing135465.
 */
com.nextjournal.journal.views.billing.__GT_t_com$nextjournal$journal$views$billing135465 = (function com$nextjournal$journal$views$billing$__GT_t_com$nextjournal$journal$views$billing135465(meta135466){
return (new com.nextjournal.journal.views.billing.t_com$nextjournal$journal$views$billing135465(meta135466));
});

}

return (new com.nextjournal.journal.views.billing.t_com$nextjournal$journal$views$billing135465(null));
})()
,new cljs.core.Keyword(null,"initial-data","initial-data",-1315709804),null,new cljs.core.Keyword(null,"options","options",99638489),new cljs.core.PersistentArrayMap(null, 1, [new cljs.core.Keyword(null,"watch-atom","watch-atom",-2134031308),false], null)], null));
})], null));

//# sourceMappingURL=billing.js.map?rel=1573720735531

definitely overlap in the billing.js. the same code repeating in the file pretty much makes it clear its a thread issue?

@thheller yes, we’re using :parallel-build. Let me try to produce the error again and see if there’s more errors… I didn’t save the build folder 🙈

@mkvlr note that :parallel-build is unlikely to create issues in the same file. that should still only compile once. so more likely that it is something related to your build setup in general. running multiple watches or so.

Tested master on Pitch code base, seems to work fine.

@mkvlr you're not trying to use the REPL at the same time as the build is running are you?

@dnolen figwheel builds on boot and we do connect emacs automatically on boot. Should we delay that until the first compile is done?

will check in what order things are happening exactly…

well if the REPL itself triggers some compilation this won't be coordinated with the first build pretty sure

fixed the node lib load-file issues at the REPL

applying 3077

anything else we should consider?

(for a release)

https://clojure.atlassian.net/browse/ASYNC-230? 🙂

well that's for core.async 🙂 there's no need to coordinate that

but yes, I think that looks ok will get that in my queue soon

Maybe https://clojure.atlassian.net/browse/CLJS-2823 ?

maybe https://clojure.atlassian.net/browse/CLJS-3003 dunno about the state of the patch though

Also https://clojure.atlassian.net/browse/CLJS-3102 https://clojure.atlassian.net/browse/CLJS-3086 https://clojure.atlassian.net/browse/CLJS-2874 https://clojure.atlassian.net/browse/CLJS-3106 https://clojure.atlassian.net/browse/CLJS-3105 https://clojure.atlassian.net/browse/CLJS-3104

sooo... I created a few patches what seems like a long time ago... I am foggy now and forget how to tell which patches are being considered for release.