search

Skrylar / red.el

Emacs editing support for the Red and Red/System programming languages.
9 stars 5 forks source link

Continued development? #2

Open xificurC opened 8 years ago

xificurC commented 8 years ago

Hi @Skrylar :)

Will you continue to work on this repo? I want to add org-babel support (although that might need to be a PR into their git, will need to check) and add this mode into MELPA, and hopefully keep the repo updated (hopefully because I tend to hop languages a lot). Anyway, I'm OK with sending you PR's if you want to, or you can add me as a contributor. But if you don't plan on working on this anymore I might as well work on the fork. Let me know so I can make the proper steps ;)

Skrylar commented 8 years ago

@xificurC I wasn't aware there was a fork.

I haven't used Red in a while. The short version is, I have a strong disagreement with their current priorities. Getting off a proprietary language stack is probably an A level issue since it gates off potential contributors, and proper memory management is far more important than flashy GUI demos.

That being said, emacs modes aren't that difficult to maintain. So if you want to send in patches, or if something has changed that I missed, I'm okay with supporting those.

As for org-babel, you don't have to patch it upstream to support new languages. There's just a table of known languages, and you can hook in to that. Supporting literate programming is probably simple. Making use of the evaluation system (you can have org run snippets and store their outputs, ex. for documentation purposes) is trickier. But both ought to be doable.

xificurC commented 8 years ago

Hi @Skrylar, there's only a fork I made and started working on some of the things I mentioned above.

Would you mind elaborating on the issues you mentioned? I see BSL and BSD licenses mentioned which should be fine, no? Re memory management - are they just postponing it or is there something you didn't agree upon?

Thanks for the pointers. As far as I understand you don't plan to work on Red or this emacs mode right now, so I suggest the following - I will work on my things in a fork (and see how far I get) and if the need arises for us to collaborate in the future we can merge our repos, or create a new one.

Awaiting your reply ;)

Skrylar commented 8 years ago

@xificurC

Red is BSD, yes. It also requires the Rebol compiler from upstream, which is proprietary. And to make distributables as Doc does requires an SDK license. This fails both the Stallman and Debian tests. While FOSS purity is not important to everyone, compliance with Debian's "us and everyone downstream can build a package" policy is useful for growth in to linux spheres. And linux developers are the ones who tend to offer up free language work (ex. red.el)

Doc told me that this was a priority after the language was more stable. Fine. Yet language stability (ex. a GC) has been postponed for flashy features to appeal to investors, so it's clear that we're second class citizens for the time being. Those of us who wanted lisp-like macros paired with parse were exiled to a separate chat room. So GC is "in the works" at the time of my last discussion.

Things I wouldn't mind: An easier way to get the list of words for making font-lock-faces lists. I had to make the initial word list by manually copying text from the terminal's help command and fixing it by hand.

Things I still need: An example of "proper" syntax indention and layout. Right now the indent function is winged from then-personal preference and informally assumed from a few REBOL examples. It really ought to be based on a community-blessed default so I have something to base unit tests on.

Stuff that might be relevant: We might need to add separate modes for Red/View, Red/System and Red. This is not so hard to do, as we can simply use a red-base and derive some modes underneath that. All that changes is a set of font lock highlights, so given the word lists its something that could be done in one or two days. Now that multi-modal Red is formally a thing, that's probably important (it really wasn't when I first wrote this mode.)

If red.el is out of date, I wouldn't mind trolling around chat for a bit until its up to date again. I'm still interested in Red, I just don't use it while it lacks basic necessities for production work.

xificurC commented 8 years ago

I understand. I hope everything smoothens in the next 1-2 years as well.

Since you're not using Red right now I wouldn't want you to take away any of your time. Let me work in my fork for now and let me know when you start doing some Red again. As I said before I have no problem e.g. making a group where we can collaborate later, or merge things back here and add me as a contributor. We can iron out the details once you get back to Red. For now I'll just try to make my own experience writing Red in Emacs more pleasant (e.g. I just pushed some commits that mostly handle bracketed strings).