They are still on my radar, but my free time has been more fragmented
than usual the past few months.
Is there anyone else who could help? Having patches lingering for months is what I find the most problematic. It's strainful to keep track of them and update/rebase them. It also makes it more difficult to work on new patches when these touch the same code as pending ones. It would be great if someone else could take care of the detailed code reviews such that you can focus your time and attention on higher-level aspects.
Among the pending patches,
Simplify resolve_nameref and setscope -
54063 is the one I would really like to see committed. I already had to rebase it and that was a pain. In order to review it, I would strongly advise against trying to review the full diff. Instead, I would rather review on GitHub the
23 commits that compose it and convince yourself that each of them makes sense and preserves the current behavior, and/or review the updated
resolve_nameref,
setscope, and
upscope functions whose code is now rather straightforward.
Yes, one of the things I've been trying to get straight is how the set
of patches in your list interact with ripping out the subscripting
code.
I don't think they introduce any fundamental change. The width field is still used to distinguish between references to plain variables and to subscripted ones. However,
Simplify resolve_nameref and setscope -
54063 would obviously conflict with any existing patch that removes subscripting code. If you already have such a patch, you can give it to me and I will try to rebase it on top of
54063 (or vice versa, if it looks easier). If you haven't yet created such a patch, I can try to create one on top of
54063.
One thing to keep in mind is that removing support for named references to subscripted variables could be as simple as forbidding the definition of such named references. I'm not aware of any case where the subscripting code affects the handling of references to plain variables. That code could in principle remain in place. It would simply be dead. Obviously, it would be much better/cleaner to rip it out but there is no necessity of doing it right away.
Philippe