Replies: 5 comments 12 replies
-
As I see it, there are two separate reasons for making a 3.6.2/3.7.0: Either we make it before the rust track is ready, because it's taking too long or we make it because some users are having issues upgrading to the rust track and so we support them for a bit while they get their situation sorted. For the first, I'm not sure how many features we'll even add - personally I've been holding back on doing any C++ work because I don't want to write things twice before they ever get used. So I guess we'll have to see how many important or easy bugs show up that we would want to backport the fixes for. For the second, I've said before that we could do that but there would have to be interest. As I've seen it, most users should be able to build the rust version by just using rustup, so there is less of an argument than there was for sticking to C++11 (because upgrading C++ compilers is a PITA). So, for the most part, this would be targeting packagers. In the past we've seen some packagers not upgrade even to patch releases (which is how e.g. 3.1.0 keeps turning up in bug reports), so I would like to see some interest in this before we waste our time. (and if it's for packagers, it's probably okay to have it as a git tag only, I've seen that work for other projects - which would also mean no macos packages, which should ease releasing a bit) |
Beta Was this translation helpful? Give feedback.
This comment was marked as off-topic.
This comment was marked as off-topic.
-
I think we'll take most of those additional items, and call it 3.7.0. |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
"in a few months" was a bit optimistic but 3.7.0 is on the way soon |
Beta Was this translation helpful? Give feedback.
-
I've not used Discussions much, but let's give it a crack.
The Rust rewrite is well underway and it seems pretty clear that releasing a half-C++, half-Rust hybrid is an explicit non-goal.
If there are significant bugs in fish 3.6.1 that need to be fixed, we can produce a 3.6.2 release. As part of this, we can roll up minor
fixes, new and improved completions, and improvements to the documentation.
What I'm less sure on is whether it's worth planning a 3.6.2 (or call it a 3.7.0 branched from 3.6.1) in a few months with minor bugs, new completions, documentation improvements and any safe new features. We've traditionally avoided new features in point releases, just by necessity, but as we don't support anything except the latest branch I think it's probably effectively the same thing.
Beta Was this translation helpful? Give feedback.
All reactions