Louis Gesbert

Adjusted for now released `asak.0.4` (removed the `pin-depends` in opam files)

Ahem, CI failure is due to the latest corpus update, not this PR (it works outside of this CI, the exercise failing is the one making use of gg/vg so...

I see this has no milestone: I think it's among the top most-wanted features and should be prioritised :) Maybe 1.1 ?

Just had this issue, which got fixed by removing `qobuz_dl.db` (inside the config directory, `~/.config/qobuz-dl` in my case)

I think this could be reported to https://github.com/ocaml/ocaml.org/issues (or maybe https://github.com/ocaml/odoc/issues ?) as this website is outside the control of `learn-ocaml`…

Ah, indeed, the latest .msi is severly dated, I need to find the time to do that. Thanks for reminding me!

This seems to only be the case on some setups; this needs more investigation, but seems to work on most Windows installations. Any details appreciated if you can reproduce (Windows...

I am not sure the executable itself can do anything in the case when it is containerised in this way (I don't understand what `--fork-replace` would do ?): AFAIK replacing...

Ah! I understand better your intent. Wouldn't `[ -d www ] && learn-ocaml serve & learn-ocaml build serve --replace` work then ?

After reading https://hynek.me/articles/docker-signals/ I still think signals should work, since we already use dumb-init... it may be worth testing anyway. Otherwise no strong objection to option 3 (maybe with a...