Fork me on GitHub
#cider
<
2017-02-11
>
richiardiandrea00:02:39

very experimental but these are the things you can achieve in emacs thanks to code you can find in cider clojure-mode or clj-refactor

onetom00:02:37

actually i was envying the capabilities of clj-refactor and the clj formatter libraries. it's a pity u cant rely on these when using intellij...

richiardiandrea00:02:42

yes they are fundamentally two ways of thinking about editors

dpsutton00:02:53

i know just about nothing about recognizing security certs

dpsutton00:02:14

but i would accept that

dpsutton00:02:17

melpa is fine

onetom00:02:21

i was just wondering if it's expected or there is something im missing

dpsutton00:02:28

i've never seen it before

dpsutton00:02:31

doesn't pop up fo rme

onetom00:02:36

because it feels like this can't be the default...

dpsutton00:02:51

possible the NSA is investigating you?

tanzoniteblack00:02:52

@onetom what's the emacs version?

dpsutton00:02:59

^ more likely

tanzoniteblack00:02:24

M-x emacs-version will give you that from within emacs, in case you weren't aware of that

dpsutton00:02:43

i was just kidding about the NSA, of course

dpsutton00:02:06

but is it possible you're on a network or you have yourself accepted or rejected the certificate issuing authorities?

dpsutton00:02:13

i think i've heard you can add and remove those yourself

dpsutton00:02:25

it sounds like you don't recognize the authority of let's encrypt

dpsutton00:02:59

> The gnutls-trustfiles variable is a list of trustfiles (certificates for the issuing authorities). This is global, not per host name (although gnutls-negotiate supports a trustfile per connection so it could be done if needed). The trustfiles can be in PEM or DER format and examples can be found in most Unix distributions. By default the following locations are tried in this order: /etc/ssl/certs/ca-certificates.crt for Debian, Ubuntu, Gentoo and Arch Linux; /etc/pki/tls/certs/ca-bundle.crt for Fedora and RHEL; /etc/ssl/ca-bundle.pem for Suse; /usr/ssl/certs/ca-bundle.crt for Cygwin; /usr/local/share/certs/ca-root-nss.crt for FreeBSD. You can easily customize gnutls-trustfiles to be something else, but let us know if you do, so we can make the change to benefit the other users of that platform.

dpsutton00:02:24

but you can accept that key and continue on, correct?

dpsutton00:02:34

or do you want to diagnose this more fully before doing anything else?

onetom00:02:36

im on regular macOS installations. browsers didnt complain about anything, though i cant remember any letsencrypt protected site now to doublecheck

onetom00:02:05

it just feels abnormal

onetom00:02:28

and if i were to recommend this setup to friends they would have the same questions and i wouldnt know what to answer

dpsutton00:02:56

seems relevant. I guess OS X uses some system keychain for SSL certs that is opaque to Emacs. Perhaps it should learn to understand it, if that's even possible. There's a suggested workaround related to libressl there.

dpsutton00:02:39

that's exactly your issue

dpsutton00:02:54

and i think they are saying that os x is being a little secretive with its accepted certs

dpsutton00:02:01

so that emacs can't see if the os uses them

dpsutton00:02:09

but there's a way to access it

onetom00:02:56

my googling havent yielded this result...

dpsutton00:02:20

i googled "emacs and let's encrypt"

dpsutton00:02:32

first result was a gnu bug report about this

dpsutton00:02:41

i'm not good but google is outstanding 🙂

onetom00:02:09

i just googled for emacs letsencrypt ...

sylvain09:02:41

I have the vague recollection of running into the same issue some time ago. I think I just switched to accessing MELPA via http instead of https

donavan11:02:39

Hi, sorry for the newb question but does cider/nrepl get the same log messages as lein repl does?

bhagany14:02:07

@donavan if I understand you correctly, the answer depends on what version of cider you're using. A bug was recently fixed that caused messages logged to stdout or stderr to get put in the server process buffer instead of the repl buffer.

bhagany14:02:46

so, if that's the behavior you want, you'll want to use 0.15.0-SNAPSHOT

bhagany14:02:08

where "that behavior" means - log messages going to the repl buffer

donavan14:02:31

by server process buffer you mean the nrepl-messages {ID} buffer?

donavan14:02:16

Yeah, my verison is 0.14

bhagany14:02:34

hmm, my installation names it *nrepl-server {ID}*, but I'm not sure if that's configurable

donavan14:02:24

it contains messages like:

(<-- 
  id  "237"
  session  "881875ca-f549-4d02-8778-f15514c7e38c"
  status  ("done")
)

bhagany14:02:06

ah, yes, so logging stuff would show up there, I think, but it would be within those message blocks

donavan14:02:22

Maybe this is all academic till I use v0.15... I'm currently using lein repl and cider-connect

bhagany14:02:31

ah, yes, that changes thigns

donavan14:02:33

thanks for the tip about he bug

bhagany14:02:00

in the cider-connect case, your log messages will end up in your lein repl

donavan14:02:10

Yeah, I'm using it as a workaround

donavan14:02:39

The original quiestion refered to the cider-jack-in case

bhagany14:02:43

okay, I see. 0.15.0-SNAPSHOT should do what you want then, if you'd rather be jacking in

donavan14:02:51

Having used lein repl and cider-connect I'm now thinking it's better to split your repl (emacs) and just have logs in terminal... though it stumped me for a while! Thanks again for the help

bhagany14:02:05

sure, no problem