Fork me on GitHub
#cljs-dev
<
2016-02-01
>
slipset14:02:38

@dnolen: so the gist of the issue I mentioned is:

slipset14:02:50

in Clojure, #{1 2 1} throws an exception

slipset14:02:08

in Clojurescript #{1 2 1} returns #{1 2}

dnolen14:02:24

@slipset: the usual order of business is to make a minimal case

dnolen14:02:35

minimals cases involve zero tooling - only ClojureScript

dnolen14:02:41

no projects, no links, no gists

dnolen14:02:57

file an issue in JIRA with the step to reproduce

dnolen14:02:08

this issue should get minor priority

dnolen14:02:19

and preferable a patch

dnolen14:02:35

before doing that it should also be explained if and how cljs.tools.reader plays into it

slipset14:02:42

Ok, will do my best.

mfikes14:02:19

@slipset: My recommendation would be to add a new eval-str unit test.

mfikes15:02:09

I’ve added a new wiki that hopefully helps with clean reports of bootstrapped issues: https://github.com/clojure/clojurescript/wiki/Reporting-Bootstrap-Issues

mfikes15:02:34

It is linked into the main “Reporting Issues” page: https://github.com/clojure/clojurescript/wiki/Reporting-Issues

mfikes15:02:24

@slipset ^ With the above, you can directly show in a shipping REPL that #{1 2 1} evaluates to #{1 2}, which is big first step. The bigger picture of the reader, etc. is perhaps special for this particular issue.

richiardiandrea17:02:29

I confirm ^ happens in replumb (and http://clojurescript.io) as well