Christopher Rowley

Make sure you read the full error message! Here is the relevant bit: ``` The following package could not be installed └─ pyfiles * does not exist (perhaps a typo...

Thanks. The slowness in CondaPkg is entirely down to setting up the PkgREPL mode. Some things we could do: - Use SnoopCompile to precompile this init function. - Put this...

CondaPkg adds new commands to the PkgREPL so we need to regenerate the help otherwise they don't show up in help.

What system did you get your timings on? On my Windows machine I don't observe a slow-down - loading CondaPkg takes about 20ms on 1.10 or 1.11-rc3 for me. However...

I have the same issue. CondaPkg isn't doing that, I think it's VSCode for some reason - but I've never worked out why. It's a mystery. I've never looked into...

Ah the docs maybe aren't clear, if you are trying to use a pre-installed Julia, set `PYTHON_JULIAPKG_EXE`.

I've not had to do that before so not sure, but some combination of `PYTHON_JULIAPKG_EXE`, `PYTHON_JULIAPKG_PROJECT`, `PYTHON_JULIAPKG_OFFLINE`, `JULIA_DEPOT` might help. For a bit on how juliapkg works: it will only...

Thinking a bit more: - Setting `PYTHON_JULIAPKG_EXE` will use a specific Julia, so won't download one. - Setting `PYTHON_JULIAPKG_OFFLINE=yes` will prevent any packages from being downloaded. - However this means...

If what you actually want is to have a read-only depot but let people create their own projects from it while offline, that may require some changes to juliapkg.

I think I'm going to change how offline mode behaves in juliapkg. Instead of just skipping installing packages entirely, I'll change it to do the same package install steps but...