Thanks for all the great things you do for Haskell, IOG crew!
Highlight from the HF perspective: @doyougnu added a feature to GHC so that error codes are hyperlinked to their pages on the Haskell Error Index. This means that in terminals that support the appropriate feature (VTE-based terminals like GNOME’s, Konsole, iTerm2 on Mac, and recent versions of the new Windows terminal), it will be possible to click the code to get an explanation.
GHCup has had this feature for some time, and it’s great to see it getting to GHC too.
We look forward to using our access logs to prioritize error message improvements.
Unfortunately I didn’t get it in, in time for 9.8 so it’ll be in the feature list for 9.10.
Thank you so much for seeing this all the way through!
Update #18 is out!
I’m now reusing this same thread for all updates to avoid spamming the Discourse.
Update #19 is out!
Quite a lot of work-in-progress that we’re hoping to land before the next update.
We’ve also published a blog post written by Bartek about his internship with us this summer:
Update #20 is out!
Update #21 is out!
Update #23 is out!
Sadly no progress on this front. The branches @JoshMeredith worked on are:
- GitHub - JoshMeredith/ghcjs-dom at wip/js-backend
- GitHub - JoshMeredith/jsaddle at wip/js-backend
- GitHub - JoshMeredith/jsaddle-dom at wip/js-backend
Iirc progress stalled last year because of misc issues (e.g. missing implementation of
lstat in 9.8.1, issues with custom Cabal’s
Setup programs…). The upcoming GHC 9.10 should be much better to make another try at fixing the ghcjs ecosystem.
This is not a priority for us, however, so any contribution is welcome and we’d be happy to help or to advise! If someone is planning to work on this, the idea was to release a new major version of these libraries that wouldn’t retain the compatibility with GHCJS to avoid having too much CPP (e.g. to deal with differences due to the FFI syntax) and to make the task easier.
Update #24 is out!