Fork me on GitHub
#clojure-dev
<
2019-06-07
>
mfikes00:06:22

Oh well, I lost the ability to repro it now. The prefix appeared to be being passed to install.sh:

==> ./install.sh /usr/local/Cellar/clojure/1.10.1.447
šŸŗ  /usr/local/Cellar/clojure/1.10.1.447: 9 files, 21.1MB, built in 3 seconds
Mikes-MacBook-Pro:~$ clj
cp: PREFIX/example-deps.edn: No such file or directory
I had Graal as my JVM, so on a hunch, I removed it and put in the stock 1.8 JVM and brew install clojure started behaving correctly. Then removing the stock 1.8 and going back to Graal did not lead to the ability to repro it. No clue what happened at this point, but sharing in case others hit it.

alexmiller13:06:13

@gfredericks :/ is "open for the language to start interpreting" and not an invalid keyword so should be ok to generate. and cljs should fix it's weirdness.

gfredericks14:06:41

@alexmiller That's an intriguing way to string words together, but your interpretation of it makes me happy; primarily because I don't have to make any hard decisions, but also because the impl and official definition can match

seancorfield17:06:56

I noticed in a comment on a blog, that clojure.tools.namespace has been "stuck" at 0.3.0-alpha4 for two years, and while that's technically the "latest development release", pretty much everyone using c.t.n is using the alpha. Is there any real reason for not declaring it "done" at this point and making an 0.3.0 release? Or maybe just declaring 0.3.0-alph4 the "latest stable release" in the readme? /cc @stuartsierra

alexmiller17:06:17

data.xml is similarly afflicted

gfredericks18:06:16

these differences matter when using tools like lein-ancient

gfredericks18:06:41

(which relates to a complaint I have about tools.logging, if anybody's interested)

gfredericks18:06:07

(the latest release is 0.5.0-alpha.1, which is nonstandard because of the extra ., and causes lein-ancient (or whatever lib it uses) to classify it as a proper release)

dominicm18:06:42

Ah, it's not following the semantic rules properly. Interesting.

gfredericks18:06:07

I suspect it was a typo

alexmiller19:06:43

that is pretty weird

gfredericks19:06:16

I don't know for sure that releasing 0.5.0 would fix it -- the algorithm might consider 0.5.0-alpha.1 to be a later release, in which case it'd require a 0.5.1

andy.fingerhut19:06:25

I suspect test.check for creating the pom.xml file šŸ™‚

alexmiller19:06:39

I emailed the project owners of those projects

thanks3 3