gyp ERR! stack Error: Can't find Python executable "C:\Users\Aristeidis\AppData\Local\Programs\Python\Python36-32\python.EXE", you can set the PYTHON env variable.
is asking for a file that exists!
If gyp cannot see python3 (if 36-32 is python3), then why was it working before with julia v0.6?
Last time I upgraded julia from v0.4 to v0.6, nothing worked again, I had to uninstall everything.
Thank you for the help in advance.
Same issue here. I remove my configuration folder and try to install the package.
I get other error message:
/snap/core/current/lib/x86_64-linux-gnu/libc.so.6: version 'GLIBC_2.28' not found (required by ~/.atom/packages/julia-client/node_modules/node-pty-prebuilt/build/Release/pty.node)
Hi, it’s a problem with the versions of some dependencies, see this issue. Happened to me as well. Simple and quick solution, without any cost for reverting it once it’s fixed:
You can just download the ZIP version (e.g. atom-windows.zip) of Atom from release 1.38.2, unzip it somehwere and run atom.exe. It will use your config (packages, settings, etc.) as a “native” installation, but not auto-update itself
And, once the problem is fixed, you can switch back to your actually installed Atom without re-installing anything.
I had to go though my hard drive and remove several folders belonging to Atom, they don’t remove on an uninstall.
The fix by asprionj seems better because its a standalone version (if I understand correctly).
[Obsolete now; the issue is solved, see below.]
The only workaround that works for me so far (on macOS 10.14.6) is to
completely uninstall Atom 1.39.1, including ~/.atom;
install Atom 1.38.2 from here: https://github.com/atom/atom/releases/tag/v1.38.2;
go to Preferences and reinstall uber-juno until the error message for julia-client occurs;
then, in the Terminal, run: apm install [email protected];
and deactive the auto-update until the issue will be fixed. :o)
Trying to install julia-client 0.8.12-alpha immediately crashes Atom 1.39.1.
Alright, I got a couple reports that [email protected] works fine on MacOS (as well as on Windows and Linux).
Sorry for the trouble!