path: root/2025/captions/emacsconf-2025-schemacs--one-year-progress-update-schemacs-formerly-gypsum--ramin-honary--main.vtt

WEBVTT captioned by sachac

00:00:01.060 --> 00:00:05.639
Hi, EmacsConf 2025. My name is Ramin Honary,

00:00:05.640 --> 00:00:07.559
and I'd like to talk to you today

00:00:07.560 --> 00:00:10.399
about my project called Schemacs

00:00:10.400 --> 00:00:12.079
which I presented last year.

00:00:12.080 --> 00:00:13.879
Back then it was called "Gypsum"

00:00:13.880 --> 00:00:18.319
and the name has since changed.

00:00:18.320 --> 00:00:20.239
So my name is Ramin Honary.

00:00:20.240 --> 00:00:24.999
I'm an Emacs enthusiast, have been since 2017 or so.

00:00:25.000 --> 00:00:27.759
I'm a full stack software developer.

00:00:27.760 --> 00:00:29.300
I love Haskell, Scheme,

00:00:29.301 --> 00:00:31.433
anything functional programming related,

00:00:31.434 --> 00:00:32.959
and of course Emacs.

00:00:32.960 --> 00:00:36.199
I started learning Scheme about three years ago,

00:00:36.200 --> 00:00:37.999
and this is my third time presenting

00:00:38.000 --> 00:00:40.799
at EmacsConf.

00:00:40.800 --> 00:00:46.479
So the Schemacs project that I'm talking to,

00:00:46.480 --> 00:00:48.159
I'm talking about to you today,

00:00:48.160 --> 00:00:50.279
was originally called "Gypsum".

00:00:50.280 --> 00:00:53.359
The reason I did not call it "Schemacs"

00:00:53.360 --> 00:00:58.119
was that the name "Schemacs" was taken on GitHub.

00:00:58.120 --> 00:00:59.567
But in the past year,

00:00:59.568 --> 00:01:02.119
I was able to get the permission

00:01:02.120 --> 00:01:04.479
of the author of GitHub's Schemacs

00:01:04.480 --> 00:01:08.840
project to name my project the same thing,

00:01:08.841 --> 00:01:11.320
even though it's a very similar project.

00:01:11.321 --> 00:01:14.719
So I changed the name.

00:01:14.720 --> 00:01:19.700
Let me see if I can quickly show the screen.

00:01:19.701 --> 00:01:24.959
So yeah, I have archived the old project.

00:01:24.960 --> 00:01:27.719
It's still there, but there's an explanation in

00:01:27.720 --> 00:01:30.167
the readme file and a screen grab

00:01:30.168 --> 00:01:31.599
of the conversation I had

00:01:31.600 --> 00:01:36.119
with the original author of GitHub Schemacs.

00:01:36.120 --> 00:01:38.679
My Schemacs is not on GitHub at all.

00:01:38.680 --> 00:01:43.879
It's only on Codeberg. So please don't get confused.

00:01:43.880 --> 00:01:45.999
But yes, I received permission

00:01:46.000 --> 00:01:50.199
to change the name and so I did.

00:01:50.200 --> 00:01:52.839
And I would like to give a quick shout-out to

00:01:52.840 --> 00:01:57.239
user "Tusharhero" for helping me with that.

00:01:57.240 --> 00:02:02.639
This person really helped make that name change happen.

00:02:02.640 --> 00:02:07.199
So back to the slides.

NOTE The scope of the project

00:02:07.200 --> 00:02:11.319
And so now I'd like to clarify the scope of the project.

00:02:11.320 --> 00:02:13.000
I don't think I quite made it clear

00:02:13.001 --> 00:02:14.919
well enough last year,

00:02:14.920 --> 00:02:18.400
but... Although I'm definitely cloning

00:02:18.401 --> 00:02:20.167
the Emacs Lisp programming language,

00:02:20.168 --> 00:02:25.267
the actual scope of the Schemacs project

00:02:25.268 --> 00:02:29.759
is to make an Emacs-like app platform for Scheme.

00:02:29.760 --> 00:02:32.959
I don't consider Emacs to be a text editor.

00:02:32.960 --> 00:02:38.199
I consider Emacs to be a Lisp app platform.

00:02:38.200 --> 00:02:39.839
So it's similar to something like the

00:02:39.840 --> 00:02:48.539
World Wide Web, or Microsoft's .NET app platform, or Java.

00:02:48.540 --> 00:02:52.559
These are all examples of app platforms.

00:02:52.560 --> 00:02:55.119
I would like Schemacs to make it easy

00:02:55.120 --> 00:02:59.399
for not only people to use it for

00:02:59.400 --> 00:03:03.479
things like editing text or, you know, for

00:03:03.480 --> 00:03:06.439
using your computer through a command line

00:03:06.440 --> 00:03:10.599
or manipulating your Git repository.

00:03:10.600 --> 00:03:13.159
I'd also like you to be able to create simple

00:03:13.160 --> 00:03:16.119
GUIs or TUIs using Scheme.

00:03:16.120 --> 00:03:19.319
So that's also one of the goals of this project.

00:03:19.320 --> 00:03:23.079
It will of course have an Emacs-like text editor,

00:03:23.080 --> 00:03:24.999
and I will clone Emacs Lisp.

00:03:25.000 --> 00:03:29.879
So hopefully GNU Emacs users

00:03:29.880 --> 00:03:32.779
will feel comfortable moving over to Schemacs

00:03:32.780 --> 00:03:35.679
because they'll be able to use your init.

00:03:35.680 --> 00:03:40.399
You'll be able to use your init.el file.

00:03:40.400 --> 00:03:43.239
So configuring and scripting Schemacs

00:03:43.240 --> 00:03:44.479
should be done in Scheme.

00:03:44.480 --> 00:03:47.679
I'd like to encourage scripting in Scheme

00:03:47.680 --> 00:03:51.199
and creating new workflows and macros in Scheme.

00:03:51.200 --> 00:03:54.267
It will support Emacs Lisp depending on

00:03:54.268 --> 00:03:59.319
how much of the Emacs Lisp interpreter I can clone.

00:03:59.320 --> 00:04:03.039
That will be supported but not encouraged.

00:04:03.040 --> 00:04:06.319
But you should still be able to run your init.el.

00:04:06.320 --> 00:04:08.839
And I would like it to be good enough,

00:04:08.840 --> 00:04:09.999
this Emacs Lisp interpreter

00:04:10.000 --> 00:04:12.999
should be good enough to run packages from ELPA.

00:04:13.000 --> 00:04:15.879
Although it will probably be some time

00:04:15.880 --> 00:04:17.559
before it will be able to run

00:04:17.560 --> 00:04:24.759
something as large as Org Mode or Magit.

NOTE Difference with Robin Templeton's project (Guile-Emacs)

00:04:24.760 --> 00:04:27.439
It is slightly different from the Guile-Emacs project.

00:04:27.440 --> 00:04:30.333
This is the work of Robin Templeton

00:04:30.334 --> 00:04:32.219
who presented last year.

00:04:32.220 --> 00:04:36.033
Guile-Emacs links the Guile runtime

00:04:36.034 --> 00:04:38.500
into the Emacs executable.

00:04:38.501 --> 00:04:41.580
It's not a Scheme application. Emacs,

00:04:41.581 --> 00:04:44.200
the core of Emacs is written in C.

00:04:44.201 --> 00:04:48.120
Guile, the core of Guile is written in C.

00:04:48.121 --> 00:04:53.700
What Robin Templeton has done is, at the C level, linked

00:04:53.701 --> 00:04:56.833
"libguile.so" into Emacs and then provided

00:04:56.834 --> 00:04:59.500
a programming layer where you can

00:04:59.501 --> 00:05:04.759
call the Scheme interpreter from Emacs Lisp

00:05:04.760 --> 00:05:11.279
so that you can run Scheme programs from within Emacs

00:05:11.280 --> 00:05:13.919
without having to launch a separate process

00:05:13.920 --> 00:05:18.039
and communicate over a channel such as a socket.

00:05:18.040 --> 00:05:19.839
You won't need "SLIME" or anything.

00:05:19.840 --> 00:05:23.419
The Guile interpreter is just right there inside of Emacs.

00:05:23.420 --> 00:05:25.999
But my project is not like this at all.

00:05:26.000 --> 00:05:28.879
Schemacs is written completely from the ground up

00:05:28.880 --> 00:05:34.999
in R7RS-compliant Scheme. And because it's R7RS-compliant,

00:05:35.000 --> 00:05:37.999
it's not bound to any one particular Scheme implementation,

00:05:38.000 --> 00:05:39.879
although Guile is the reference implementation.

00:05:39.880 --> 00:05:42.359
One goal of this project is to be able to run

00:05:42.360 --> 00:05:49.719
Schemacs on any R7RS-compliant Scheme implementation.

NOTE Progress made since last year

00:05:49.720 --> 00:05:56.259
The work that I've done this past year mostly is internal.

00:05:56.260 --> 00:06:00.939
There's not much that you can see on the surface.

00:06:00.940 --> 00:06:04.519
But the most... One of the most important things that I

00:06:04.520 --> 00:06:06.839
did was I rewrote the parser in R7RS Scheme,

00:06:06.840 --> 00:06:07.919
so it no longer depends on

00:06:07.920 --> 00:06:11.999
the Guile regular expressions library.

00:06:12.000 --> 00:06:14.959
The parser now also provides source locations,

00:06:14.960 --> 00:06:18.939
so if an error occurs in Emacs Lisp,

00:06:18.940 --> 00:06:20.567
there will be a stack trace

00:06:20.568 --> 00:06:23.633
and it will show you where in the source code

00:06:23.634 --> 00:06:28.319
the error occured. This was not possible last year.

00:06:28.320 --> 00:06:30.860
And because it no longer depends on Guile,

00:06:30.861 --> 00:06:34.520
I can make it work on multiple Scheme implementations.

00:06:34.521 --> 00:06:36.820
So far, I've been able to get it to run on

00:06:36.821 --> 00:06:38.920
the Chibi Scheme interpreter

00:06:38.921 --> 00:06:41.280
and the Gauche Scheme interpreter, as well as

00:06:41.281 --> 00:06:44.279
Guile, which is the reference implementation.

00:06:44.280 --> 00:06:48.559
For a short time, it did work also on Chez Scheme,

00:06:48.560 --> 00:06:53.179
the Chez Scheme compiler, using Gwen Weinholt's "Akku,"

00:06:53.180 --> 00:06:59.299
which is a program that translates R7RS Scheme to R6RS Scheme.

00:06:59.300 --> 00:07:04.519
And with that translation, because Chez Scheme

00:07:04.520 --> 00:07:07.319
is pretty strictly an R6RS compiler,

00:07:07.320 --> 00:07:11.519
the translation allows you to run R7RS programs.

00:07:11.520 --> 00:07:15.219
But due to some change, I'm not sure where,

00:07:15.220 --> 00:07:17.119
it may have been changed in the Schemacs source code,

00:07:17.120 --> 00:07:19.639
or it may have been a change to Akku,

00:07:19.640 --> 00:07:21.239
but it no longer builds on Chez.

00:07:21.240 --> 00:07:28.039
It did at one point. I'd like to try to fix that.

NOTE Portable React-like GUI

00:07:28.040 --> 00:07:30.719
The second most important thing that I've worked on is a

00:07:30.720 --> 00:07:36.439
portable React-like GUI. And so React,

00:07:36.440 --> 00:07:40.999
for anyone who has done web programming,

00:07:41.000 --> 00:07:46.839
is a very popular framework for programming web applications.

00:07:46.840 --> 00:07:48.233
And I've provided something

00:07:48.234 --> 00:07:49.599
very similar to that in Scheme now.

00:07:49.600 --> 00:07:54.679
So it works. I have constructed a DOM data structure

00:07:54.680 --> 00:07:59.079
in Scheme. It's just an ordinary Scheme data structure.

00:07:59.080 --> 00:08:01.519
It works like the web's "Document Object Model"

00:08:01.520 --> 00:08:03.960
or the "DOM" data structure.

00:08:03.961 --> 00:08:09.999
And then this Scheme DOM data structure can be rendered

00:08:10.000 --> 00:08:13.059
using any GUI framework that is convenient

00:08:13.060 --> 00:08:16.239
for the Scheme implementation that you're targeting.

00:08:16.240 --> 00:08:18.879
And you should be able to implement

00:08:18.880 --> 00:08:22.919
also rendering to a CLI as well.

00:08:22.920 --> 00:08:24.600
The current reference implementation

00:08:24.601 --> 00:08:27.759
is using a framework called Guile-GI.

00:08:27.760 --> 00:08:30.639
This is the "GObject Introspection" framework.

00:08:30.640 --> 00:08:31.967
It's a very simple

00:08:31.968 --> 00:08:36.119
GObject Introspection framework for Guile,

00:08:36.120 --> 00:08:40.979
and it binds to GTK3 on Linux.

00:08:40.980 --> 00:08:42.919
There's a similar framework called G-Golf

00:08:42.920 --> 00:08:48.359
which I'd like to begin using as well, also for Guile.

00:08:48.360 --> 00:08:53.739
G-Golf seems to be a bit more well-maintained, a bit...

00:08:53.740 --> 00:08:57.799
It has better features. G-Golf may be a

00:08:57.800 --> 00:09:00.039
better rendering backend for the reference

00:09:00.040 --> 00:09:06.039
implementation, but I would like to provide both.

NOTE Demo

00:09:06.040 --> 00:09:07.933
I will give a demo of this now.

00:09:07.934 --> 00:09:12.999
Unfortunately not a whole lot

00:09:13.000 --> 00:09:18.139
to see compared to last year.

00:09:18.140 --> 00:09:20.639
First thing I'd like to show is

00:09:20.640 --> 00:09:24.279
that I now have a Makefile. You can look inside

00:09:24.280 --> 00:09:28.400
this Makefile and if you're able to read a Makefile,

00:09:28.401 --> 00:09:31.967
you can see that I have several targets now available.

00:09:31.968 --> 00:09:35.000
You can build Schemacs for Guile,

00:09:35.001 --> 00:09:36.667
you can build Schemacs for Gambit,

00:09:36.668 --> 00:09:40.333
or Stklos, or Chicken, or Chez,

00:09:40.334 --> 00:09:42.900
although none of these (except for Guile)

00:09:42.901 --> 00:09:51.167
currently works. These targets will actually

00:09:51.168 --> 00:09:56.000
build the source code, but then you would have to

00:09:56.001 --> 00:09:59.433
load it into the REPL separately.

00:09:59.434 --> 00:10:02.467
There are targets for launching

00:10:02.468 --> 00:10:06.467
a Gauche REPL and a Chibi REPL.

00:10:06.468 --> 00:10:10.867
You can also run the Emacs Lisp tests

00:10:10.868 --> 00:10:13.067
in Gauche and Chibi.

00:10:13.068 --> 00:10:19.079
You can also start a Guile REPL through this Makefile.

00:10:19.080 --> 00:10:27.499
So I will do that right now in the shell.

00:10:27.500 --> 00:10:31.833
(...make the text larger...there we go...)

00:10:31.834 --> 00:10:38.479
OK, so we have this directory of the source code.

00:10:38.480 --> 00:10:44.559
Let's just begin by running "guile.sh".

00:10:44.560 --> 00:10:52.600
This will launch a REPL and you can load "main-guile".

00:10:52.601 --> 00:10:59.719
This will launch the GUI. This is the basic

00:10:59.720 --> 00:11:04.800
proof of concept GUI that uses Guile-GI.

00:11:04.801 --> 00:11:10.519
So it may be hard to see.

00:11:10.520 --> 00:11:14.559
I cannot change the size of the text yet.

00:11:14.560 --> 00:11:17.333
I've implemented the M-: feature

00:11:17.334 --> 00:11:24.067
where you can eval in a minibuffer some Scheme code.

00:11:24.068 --> 00:11:34.439
(string-append "hello" ...)

00:11:34.440 --> 00:11:38.279
It outputs the result in the buffer.

00:11:38.280 --> 00:11:40.959
This is basically the "*Messages*" buffer.

00:11:40.960 --> 00:11:42.619
And that's all the more that I have.

00:11:42.620 --> 00:11:45.479
This is the same state it was in last year.

00:11:45.480 --> 00:11:48.699
It hasn't changed a whole lot since back then.

NOTE Additional changes

00:11:48.700 --> 00:11:52.819
But I have made additional changes.

00:11:52.820 --> 00:12:05.379
So first of all, you can run

00:12:05.380 --> 00:12:10.200
(let me just go back into the Guile)... you can

00:12:10.201 --> 00:12:16.619
run the Emacs Lisp interpreter tests, so "elisp-tests".

00:12:16.620 --> 00:12:18.919
As you can see, it gives you a stack trace.

00:12:18.920 --> 00:12:22.039
So this is an error that I've been able to reproduce.

00:12:22.040 --> 00:12:24.599
I know exactly what the cause of this error is.

00:12:24.600 --> 00:12:27.599
It is not finding a variable

00:12:27.600 --> 00:12:30.319
because the closure is not correctly

00:12:30.320 --> 00:12:35.199
capturing its environment. So there should be a variable

00:12:35.200 --> 00:12:38.719
in the closure, but that variable has not been captured

00:12:38.720 --> 00:12:43.459
and so it is causing an error.

00:12:43.460 --> 00:12:48.080
It is currently loading "byte-run.el".

00:12:48.081 --> 00:12:49.820
Let me show you what code that is here.

00:12:49.821 --> 00:12:53.500
So I've copied into the source repository

00:12:53.501 --> 00:12:58.760
for Schemacs some of the Elisp code from GNU Emacs.

00:12:58.761 --> 00:13:04.420
So I have this "subr.el".

00:13:04.421 --> 00:13:09.860
This declares most of the core of Emacs Lisp

00:13:09.861 --> 00:13:12.860
that's not written in C.

00:13:12.861 --> 00:13:18.999
There's also "byte-run.el". Schemacs Emacs Lisp can now

00:13:19.000 --> 00:13:24.379
evaluate this. This is where functions like "defun" are

00:13:24.380 --> 00:13:28.359
defined, and "defmacro". So as you can see,

00:13:28.360 --> 00:13:30.799
defun itself is a defmacro defined right here.

00:13:30.800 --> 00:13:34.859
It's written in Emacs Lisp itself,

00:13:34.860 --> 00:13:37.719
defined in terms of defalias.

00:13:37.720 --> 00:13:40.239
So I can evaluate "byte-run",

00:13:40.240 --> 00:13:42.739
I can evaluate "macroexp",

00:13:42.740 --> 00:13:46.019
and the failure occurs somewhere in "subr.el".

00:13:46.020 --> 00:13:48.959
Although if you look at the stack trace,

00:13:48.960 --> 00:13:51.159
it doesn't provide all the necessary information.

00:13:51.160 --> 00:13:56.439
So it appears to be happening in byte-run.el.

00:13:56.440 --> 00:14:00.619
Really, it's an error that's occurring inside of a macro,

00:14:00.620 --> 00:14:05.799
and the macro call site is somewhere in subr.el.

00:14:05.800 --> 00:14:08.639
Anyway, take note of this stack trace.

00:14:08.640 --> 00:14:12.019
This was run from within Guile.

NOTE Other Scheme implementations

00:14:12.020 --> 00:14:14.199
Now what I've done this past year

00:14:14.200 --> 00:14:19.479
is make it work on other Scheme implementations.

00:14:19.480 --> 00:14:23.833
Use "make" to launch a Gauche REPL.

00:14:23.834 --> 00:14:25.999
Now I'm inside of Gauche.

00:14:26.000 --> 00:14:27.233
This is the command that

00:14:27.234 --> 00:14:30.079
you would use to launch a Gauche REPL.

00:14:30.080 --> 00:14:38.199
And I can load the same program (load "elisp-tests.scm").

00:14:38.200 --> 00:14:42.759
You get the exact same result as Guile.

00:14:42.760 --> 00:14:45.799
So we have two different Scheme implementations

00:14:45.800 --> 00:14:46.799
producing the same result.

00:14:46.800 --> 00:14:53.079
Let's try "make" a Chibi REPL. This is Chibi Scheme.

00:14:53.080 --> 00:15:00.219
And you can (load "elisp-tests.scm").

00:15:00.220 --> 00:15:04.080
Chibi is a bit slower, but you get the exact same result.

00:15:04.081 --> 00:15:07.400
So we have three different Scheme implementations

00:15:07.401 --> 00:15:11.539
all running Emacs Lisp,

00:15:11.540 --> 00:15:14.039
and all producing the same result.

00:15:14.040 --> 00:15:14.767
I think that's...

00:15:14.768 --> 00:15:17.099
I'm fairly proud of that accomplishment.

00:15:17.100 --> 00:15:21.200
I was able to get the code written to the point

00:15:21.201 --> 00:15:24.879
where it actually runs on multiple implementations.

00:15:24.880 --> 00:15:30.599
You can also try making it for other Scheme compilers

00:15:30.600 --> 00:15:35.959
like "schemacs-mitscheme" for example,

00:15:35.960 --> 00:15:40.019
but this will fail.

00:15:40.020 --> 00:15:46.679
You can try building it for "schemacs-chez",

00:15:46.680 --> 00:15:51.800
Let's try Chez... there we go.

00:15:51.801 --> 00:15:54.233
And it will use Akku,

00:15:54.234 --> 00:15:58.000
and it will fetch the necessary dependencies.

00:15:58.001 --> 00:16:03.433
But it fails, and I haven't been able to

00:16:03.434 --> 00:16:05.959
debug that quite yet.

00:16:05.960 --> 00:16:13.139
Stklos fails for a similar reason.

00:16:13.140 --> 00:16:21.699
Gambit... Chicken still doesn't build all the way yet.

00:16:21.700 --> 00:16:27.120
The Makefile at least has places for it.

00:16:27.121 --> 00:16:30.967
If anyone can help me out and get Schemacs to compile

00:16:30.968 --> 00:16:32.733
on these other Scheme implementations,

00:16:32.734 --> 00:16:34.000
I'd appreciate it.

00:16:34.001 --> 00:16:35.467
I can probably figure it out myself,

00:16:35.468 --> 00:16:37.000
but that will take more time.

00:16:37.001 --> 00:16:40.120
And let me just show you quickly

00:16:40.121 --> 00:16:41.720
the test program.

00:16:41.721 --> 00:16:44.500
Basically this is the Emacs Lisp test program

00:16:44.501 --> 00:16:48.580
that I was just running, and it produces an error.

00:16:48.581 --> 00:16:53.220
All it does is it loads these files here in this order,

00:16:53.221 --> 00:16:55.180
and it fails right around here.

00:16:55.181 --> 00:16:58.319
So it's able to load these two.

00:16:58.320 --> 00:17:01.360
And yeah, that's what I've accomplished

00:17:01.361 --> 00:17:06.199
on the Emacs Lisp side of things.

NOTE GUI framework

00:17:06.200 --> 00:17:09.119
The next thing I want to show you is the GUI framework

00:17:09.120 --> 00:17:13.199
that I've written, which I'm fairly proud of so far.

00:17:13.200 --> 00:17:15.833
So this is the GUI framework

00:17:15.834 --> 00:17:19.919
(oops, I better launch it again, OK...)

00:17:19.920 --> 00:17:22.600
and let me show you the tests.

00:17:22.601 --> 00:17:25.700
So here in the tests, you can start to see

00:17:25.701 --> 00:17:29.067
some examples of how you use it.

00:17:29.068 --> 00:17:31.833
So here is a "counter" test, and this is kind of like

00:17:31.834 --> 00:17:35.940
the "hello world" of reactive programming frameworks,

00:17:35.941 --> 00:17:38.000
where you have a state variable,

00:17:38.200 --> 00:17:41.659
sometimes called an "observable."

00:17:41.660 --> 00:17:43.439
I'm calling it "number",

00:17:43.440 --> 00:17:47.838
and it uses "=" to check if the state has updated.

00:17:49.000 --> 00:17:52.820
If an update occurs and the new value is different

00:17:52.821 --> 00:17:55.032
from the old value according to the "=" function,

00:17:55.033 --> 00:17:59.232
then trigger a state update in the GUI as well.

00:17:59.233 --> 00:18:03.099
Initialize to 0, bound to "number".

00:18:03.100 --> 00:18:06.132
I have a "button" function which creates a simple button.

00:18:06.133 --> 00:18:07.832
It takes a label and an action.

00:18:07.833 --> 00:18:10.865
Right here you see the "div" command.

00:18:10.866 --> 00:18:13.039
This is what creates a "div".

00:18:13.040 --> 00:18:18.999
Using the properties, I describe that this div is a

00:18:19.000 --> 00:18:22.599
push-button and the "on-button-push" is an action.

00:18:22.600 --> 00:18:25.265
The action is to update the variable "number"

00:18:25.266 --> 00:18:30.399
using whatever function or lambda was provided to it.

00:18:30.400 --> 00:18:31.965
And then the content that you see on screen,

00:18:31.966 --> 00:18:34.033
that you will see on screen when it runs,

00:18:34.034 --> 00:18:38.667
is here. You create a "div-pack cut-vertical".

00:18:38.668 --> 00:18:43.233
You declare two buttons and then you declare

00:18:43.234 --> 00:18:46.800
this "use-vars" which will take the content

00:18:46.801 --> 00:18:48.833
of this variable here, this observable,

00:18:48.834 --> 00:18:51.733
and place it into the GUI next to the buttons here.

00:18:51.734 --> 00:18:54.233
So what you will see on screen is

00:18:54.234 --> 00:18:57.067
a "plus" button which increments,

00:18:57.068 --> 00:18:59.320
here is the "increment" function,

00:18:59.321 --> 00:19:02.779
a "minus" button which decrements,

00:19:02.780 --> 00:19:05.479
and then the content of the variable that is

00:19:05.480 --> 00:19:09.699
being incremented and decremented.

00:19:09.700 --> 00:19:11.865
The advantage of these reactive frameworks is that

00:19:11.866 --> 00:19:13.065
with very few lines of code

00:19:13.066 --> 00:19:16.032
you can create fairly complex interfaces.

00:19:16.033 --> 00:19:18.599
The less code you have to write,

00:19:18.600 --> 00:19:21.232
the fewer chances you have to make mistakes.

00:19:21.233 --> 00:19:23.967
So let's just run this program.

00:19:23.968 --> 00:19:33.292
This was the "counter-test."

00:19:33.293 --> 00:19:35.199
And that is the debug window. Here's the "counter."

00:19:35.200 --> 00:19:38.452
I'm sorry it's not much larger than this.

00:19:38.453 --> 00:19:41.132
But here's the "plus" button, the "minus" button,

00:19:41.133 --> 00:19:43.919
and here's the "number", 0.

00:19:43.920 --> 00:19:49.792
And I can increment or decrement as much as I like.

00:19:49.793 --> 00:19:51.100
So yeah, that's kind of the hello world

00:19:51.101 --> 00:19:55.239
of reactive programming. (I'll reboot the REPL...)

00:19:55.240 --> 00:19:58.599
The next thing I want to show you is this layout test.

00:19:58.600 --> 00:20:02.192
And I'll just run the test first.

00:20:02.193 --> 00:20:05.019
So here we have basically

00:20:05.020 --> 00:20:08.539
a tiling window manager kind of thing,

00:20:08.540 --> 00:20:11.279
where you can resize the tiles

00:20:11.280 --> 00:20:12.933
and then by clicking on these buttons here,

00:20:12.934 --> 00:20:16.939
you can change the layout.

00:20:16.940 --> 00:20:20.167
So you can do two on the right,

00:20:20.168 --> 00:20:27.100
two up above, or three up above. So, yeah.

00:20:27.101 --> 00:20:30.267
Those tiling windows, as you can see,

00:20:30.268 --> 00:20:33.000
once I work this branch into the main branch of Schemacs,

00:20:33.001 --> 00:20:35.273
I can use that to implement

00:20:35.274 --> 00:20:38.265
the split window functionality for Schemacs,

00:20:38.266 --> 00:20:42.466
the editor. So here's what this split...

00:20:42.466 --> 00:20:44.865
Here's the layout test that you just saw.

00:20:44.866 --> 00:20:46.732
Let me make it a bit smaller

00:20:46.733 --> 00:20:48.199
so that it all fits on one screen.

00:20:48.200 --> 00:20:51.365
So basically we have the "button" command again,

00:20:51.366 --> 00:20:54.132
and then these are the button actions

00:20:54.133 --> 00:20:56.999
which basically just changes the layout,

00:20:57.000 --> 00:21:01.459
and then I have the layout. So this layout is a "div."

00:21:01.460 --> 00:21:04.919
The first div just places three buttons in a row.

00:21:04.920 --> 00:21:07.532
The next layout is a div within a div.

00:21:07.533 --> 00:21:12.999
So we have one div which places the button

00:21:13.000 --> 00:21:18.632
called "two right" buttons, and the div above it,

00:21:18.633 --> 00:21:21.492
which places the "three in a row" button

00:21:21.493 --> 00:21:22.660
or the "two above" buttons.

00:21:22.661 --> 00:21:26.000
And here's the next... So there's three different

00:21:26.001 --> 00:21:30.600
layouts, and clicking on one of their associated buttons

00:21:30.601 --> 00:21:32.100
will just change the layout.

00:21:32.101 --> 00:21:33.433
As you can see, very little code

00:21:33.434 --> 00:21:38.500
to create a somewhat complex user interface.

00:21:38.501 --> 00:21:42.867
That's the advantage of using reactive or declarative

00:21:42.868 --> 00:21:44.233
UI programming paradigms.

00:21:44.234 --> 00:21:47.232
So yeah, this has not been merged into Schemacs

00:21:47.233 --> 00:21:48.700
at the time of this recording,

00:21:48.701 --> 00:21:51.519
but will be soon hopefully.

NOTE Wrapping up

00:21:51.520 --> 00:21:55.165
So yeah, I think I've already gone on for 20 minutes.

00:21:55.166 --> 00:21:59.139
So I guess I'll just end my presentation here.

00:21:59.140 --> 00:22:00.465
I have lots more to talk about.

00:22:00.466 --> 00:22:03.979
I guess I will say one last thing before I go:

00:22:03.980 --> 00:22:07.065
that I would very much like for others

00:22:07.066 --> 00:22:09.199
to try and contribute to this project.

00:22:09.200 --> 00:22:14.232
I will do my best to try and help teach anybody

00:22:14.233 --> 00:22:16.832
or work with anybody, especially even

00:22:16.833 --> 00:22:18.599
if you don't have much experience with Scheme.

00:22:18.600 --> 00:22:22.759
I'd like to help everybody try to contribute.

00:22:22.760 --> 00:22:26.239
Basically I want to get this proof of concept working.

00:22:26.240 --> 00:22:30.673
I want to get a stable user interface up and running,

00:22:30.674 --> 00:22:33.065
and then we can start working on

00:22:33.066 --> 00:22:36.699
improving the Emacs Lisp interpreter all together.

00:22:36.700 --> 00:22:41.065
There are close to 1,400 built-in functions

00:22:41.066 --> 00:22:43.659
which need to be implemented.

00:22:43.660 --> 00:22:44.965
We don't need to get all of them

00:22:44.966 --> 00:22:48.465
in order to be able to run probably most of ELPA,

00:22:48.466 --> 00:22:50.865
but as much as possible.

00:22:50.866 --> 00:22:54.799
We would like to clone Emacs Lisp and I need help.

00:22:54.800 --> 00:23:02.579
So get a hold of me. My project is on Codeberg.

00:23:02.580 --> 00:23:06.919
Well, (oh, I can't show this here),

00:23:06.920 --> 00:23:14.113
but I will end it there. Thank you for listening.