Fork me on GitHub
#cider
<
2019-09-20
>
bozhidar07:09:44

> These in general should be warnings somewhere. These are approximations and should guide people when things fail but not fail the process on their own

bozhidar07:09:21

How can this be a warning? Generally if the required dependency is missing all you get is a weird error. The idea of those checks was to actually repackage the error into something more understandable.

bozhidar07:09:19

> I would just edit the source of the function and remove the body. Maybe I can change all those user errors to be warnings or stings spit out in the repl buffer It used to be like this, but I wanted to prevent the creation of an useless buffer that doesn’t work as expected.

bozhidar07:09:05

> can someone update that check? shadow.replwas never the implementation namespace

bozhidar07:09:15

@thheller Sure, I can update that.

bozhidar07:09:24

I just needed some ns from the project to check, as this was the only way to figure out if shadow-cljs is available or not. What happened with this ns? Its name sounded somewhat central to me. 😄

thheller08:09:54

@bozhidar it was related to a REPL experiment that never went anywhere, I removed it in a cleanup since nobody was accessing what it provided anyways (not even me). the actual CLJS REPL implementation lives in shadow.cljs.repl but better check for an "official" API namespace like shadow.cljs.devtools.api

bozhidar08:09:36

Understood.

bozhidar08:09:46

I’ve just updated this check in CIDER’s master.

👍 4
erwinrooijakkers09:09:56

When I do cider-jack-in-clj&cljs I get this message:

[nREPL] Starting server via /usr/local/bin/lein update-in :dependencies conj \[nrepl\ \"0.6.0\"\] -- update-in :dependencies conj \[cider/piggieback\ \"0.4.1\"\] -- update-in :plugins conj \[refactor-nrepl\ \"2.5.0-SNAPSHOT\"\] -- update-in :plugins conj \[cider/cider-nrepl\ \"0.22.0-beta6\"\] -- repl :headless :host localhost...
And Figwheel does not work then, it shows the Figwheel Dev page instead of the wanted index.html. A colleague has a working dev environment and has this output:
[nREPL] Starting server via /usr/local/bin/lein update-in :dependencies conj \[nrepl\ \"0.6.0\"\] -- update-in :dependencies conj \[cider/piggieback\ \"0.4.0\"\] -- update-in :plugins conj \[refactor-nrepl\ \"2.4.0\"\] -- update-in :plugins conj \[cider/cider-nrepl\ \"0.21.1\"\] -- repl :headless :host localhost...
As can be seen he uses refactor-nrepl 2.4.0 instead of 2.5.0-SNAPSHOT and cider/cider-nrepl-0.21.1 instead of 0.22.0-beta6. How can this be configured?

practicalli-johnny16:09:28

Did you get this working? I don't think the minor version numbers should make a difference. Sometimes clearing the browser cache in the browser dev console can help

erwinrooijakkers20:09:23

Thanks. Not yet. Reboot and clearing things does not help.

erwinrooijakkers09:09:16

And also piggieback 0.4.0 instead of 0.4.1

bozhidar10:09:53

@jumar I have to check this, but I have we try to update the cache every couple of days.

👍 4
jumar10:09:13

would be great if it could be just skipped if connection doesn't work.

yuhan11:09:32

Is there any reason why cider-eval-last-sexp displays its overlay result at the end of the line instead of the sexp itself?

yuhan12:09:08

I traced the source and found that cider--make-result-overlay takes a cons cell, but cider--display-interactive-eval-result only ever passes single points/markers to it

bozhidar12:09:01

I don’t remember our reasoning from back then at all. 🙂 Might have been some oversight on our part.

yuhan12:09:53

ok, will open a PR to fix it 🙂