Add functions to restore custom buffer bookmark properties

[shipmints]

• New defcustoms 'bufferlo-bookmark-buffer-locals',
  'bufferlo-bookmark-buffer-handler-functions'.
• New defuns 'bufferlo-bookmark-map-function-buffer-locals',
  'bufferlo-bookmark-buffer-handler-buffer-locals',
  'bufferlo-bookmark-map-function-text-scale-mode-amount',
  'bufferlo-bookmark-buffer-handler-text-scale-mode-amount'.
• Update documentation.

[shipmints]

I suppose the configuration of these two features could be simplified like we handle ibuffer integration. The hooks need to be installed in pairs so we could automate this. Nobody will want to have only half the hooks installed, right?

[shipmints]

Done.

[shipmints]

Flo, should I merge this or do you have anything outstanding?

[florommel]

Just a nit, bufferlo-bookmark-buffer-locals and bufferlo-bookmark-text-scale-mode-amount seem to be flipped in the README's code examples.

[florommel]

This should be bufferlo-bookmark-text-scale-mode-amount, right?

[florommel]

and this should be bufferlo-bookmark-buffer-locals

[shipmints]

Looks like they are flipped, yep.

[florommel]

I also think that we don't need e2cf46e anymore.
Have a look at this branch: florommel-bufferlo-bookmark-buffer-handler-functions
I rebased it, removed e2cf46e and fixed the meantioned README nit. The bookmark handler feature hooks work fine for me (very cool, now I can save my but also have a quick look, then we can merge it.

PS: Great stuff, I just discovered the power of this.. now I can persist my symbol overlays with:

(setq bufferlo-bookmark-buffer-local-variables '(symbol-overlay-keywords-alist))

[shipmints]

e2cf46e

This is still required e2cf46e and I don't see it removed from the proposed branch. I still use it. It's independent of buffer locals so mentioning it in this PR seems a bit confusing?

[florommel]

e2cf46e

This is still required e2cf46e and I don't see it removed from the proposed branch. I still use it. It's independent of buffer locals so mentioning it in this PR seems a bit confusing?

Yes, sorry I was not precise; we still need it, of course. What I wanted to say is that this behavior is already on main. And you are right this has nothing to do with buffer locals. e2cf46e is just listed in the commits for this PR.

But anyways, do you agree with the rebased branch? I use it and it works.

[shipmints]

e2cf46e

This is still required e2cf46e and I don't see it removed from the proposed branch. I still use it. It's independent of buffer locals so mentioning it in this PR seems a bit confusing?

Yes, sorry I was not precise; we still need it, of course. What I wanted to say is that this behavior is already on main. And you are right this has nothing to do with buffer locals. e2cf46e is just listed in the commits for this PR.

But anyways, do you agree with the rebased branch? I use it and it works.

Seems fine to my eye, though I didn't test it because you did 👍 Not sure how e2cf46e got all mixed up in there but if it's in main, we're good. I'll adopt the main main once this is merged in. I can make some time to look at the Emacs 31 improvements this week.