-
Notifications
You must be signed in to change notification settings - Fork 23
allow installing ruby shims in a separate prefix #413
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
|
To me, it would also be fine to always install to |
|
@doudou Ping? |
|
The shims are global even with separate_prefixes? is meant to isolate the packages, not the primary tooling from the workspace. In any case, this patch would break a lot of stuff, no ? You would at least need to add the new path to Autoproj's internal path, and make sure it ends up where it should in env.sh. |
That's not the goal. The issue I'm trying to fix is that cmake will add the parent folder of each folder in
I don't see why, and it didn't in my tests but I can put it behind a configuration flag if you prefer.
It's there already (the second line following the last line of the patch) |
|
@doudou Ping? (Sorry for insisting, I'm in the process of resyncing our internal mirrors and would like to get these PRs in first) |
Apologies on my side as well ... The last two weeks were pretty intense. Fine then. But why not move it in both cases (separate_prefixes and normal) ? Wouldn't the same issue apply with separate_prefixes = false ? |
|
I'm also worried about this breaking existing builds ... I'd rather not publish autoproj 3 just for this. So please protect the new behaviour with a configuration flag. |
I'm fine with that.
It would, but that's kind of expected in that case because all packages share the same prefix anyway. I will change the patch to have the shims always in a separate prefix, regardless of @doudou Could you please have a look at #395 ? That one has been open since 2023. |
8f351e6 to
3dbf1a2
Compare
|
@doudou Done! |
CMake automatically adds the parent folder of each path in
PATHto its search paths, (partially) defeating the purpose of the setting.