path: root/2020/subtitles/emacsconf-2020--22-powering-up-special-blocks--musa-al-hassy.vtt

WEBVTT

00:00:04.080 --> 00:00:07.359
All right, then. Well, hello everyone.

00:00:07.359 --> 00:00:11.519
I hope you're all enjoying the EmacsConf.

00:00:11.519 --> 00:00:15.040
My name is Musa Al-hassy, and I hope you're

00:00:15.040 --> 00:00:19.840
excited to learn about powering up special blocks.

00:00:19.840 --> 00:00:25.574
Let's first off find out
what these special blocks are,

00:00:25.574 --> 00:00:27.920
and see what we can go from.

00:00:27.920 --> 00:00:30.240
Yesterday, I saw a lot of cool talks

00:00:30.240 --> 00:00:33.200
and people were chatting about

00:00:33.200 --> 00:00:37.200
how should you present? Should you
do it this way or that way?

00:00:37.200 --> 00:00:39.931
I thought maybe I should try a different way.

00:00:39.931 --> 00:00:42.567
But I'm talking about special blocks

00:00:42.567 --> 00:00:45.039
and if I show you an Emacs, then I have to export

00:00:45.039 --> 00:00:47.840
the HTML so you can see what it looks like

00:00:47.840 --> 00:00:51.920
or export to a PDF so you can see what it looks like.

00:00:51.920 --> 00:00:54.239
So I ended up writing in org-reveal,

00:00:54.239 --> 00:00:58.233
and joyously, this just works.

00:00:58.960 --> 00:01:00.879
You can just see things here.

00:01:00.879 --> 00:01:03.452
I was worried that I'd have to take pictures

00:01:03.452 --> 00:01:09.760
and then insert pings, so that was a delight.

00:01:09.760 --> 00:01:15.704
Okay. Special blocks are these things like a center small quote.

00:01:15.704 --> 00:01:17.280
That's what a special block is,

00:01:17.280 --> 00:01:19.733
and with a bit of Lisp, we can make

00:01:19.733 --> 00:01:22.741
special blocks and link types.

00:01:22.741 --> 00:01:24.799
Right. Using a single interface.

00:01:24.799 --> 00:01:27.344
The interface is going to be similar

00:01:27.344 --> 00:01:29.281
to one many people are familiar with.

00:01:29.281 --> 00:01:34.712
In particular, Org Babel's src interface

00:01:34.712 --> 00:01:37.840
as well as using global header arguments for link types.

00:01:37.840 --> 00:01:39.450
The idea is to write it once

00:01:39.450 --> 00:01:41.200
and generate many different kinds.

00:01:41.200 --> 00:01:45.213
You write in Org markup and you can have HTML,

00:01:45.213 --> 00:01:49.767
you can have PDF, and joyously, org-reveal.

00:01:49.767 --> 00:01:53.600
That was an unexpected delight.

00:01:53.600 --> 00:01:57.840
Here are a few that you'll just see

00:01:57.840 --> 00:02:01.759
in this presentation.

00:02:01.759 --> 00:02:05.040
I won't show some of these link-only ones,

00:02:05.040 --> 00:02:06.799
but we'll see a few of these other ones

00:02:06.799 --> 00:02:09.500
just to make the presentation look nice

00:02:09.500 --> 00:02:11.520
So the presentation is really going to

00:02:11.520 --> 00:02:14.000
present these blocks and the mechanism

00:02:14.000 --> 00:02:15.120
at the same time.

00:02:15.120 --> 00:02:18.400
No HTML was written.

00:02:18.400 --> 00:02:21.280
Look, Ma! No HTML, just pure Org Mode,

00:02:21.280 --> 00:02:25.840
and you get all these beautiful boxes and things.

00:02:25.840 --> 00:02:27.680
The motivation for this is...

00:02:29.120 --> 00:02:31.200
you're online, you run into a blog,

00:02:31.200 --> 00:02:32.120
and you see something you like,

00:02:32.120 --> 00:02:33.120
and you're like, man,

00:02:33.120 --> 00:02:36.640
you know, I wish I could produce that.

00:02:36.640 --> 00:02:41.639
But you check, and the author wrote raw HTML.

00:02:41.639 --> 00:02:44.239
You know, #+HTML: everywhere.

00:02:44.239 --> 00:02:49.533
That's going to obscure your real content.

00:02:49.533 --> 00:02:51.200
It's going to be surrounded by all this

00:02:51.200 --> 00:02:54.239
styling information. That's unfortunate.

00:02:54.239 --> 00:02:57.033
The author decides to use an Org macro.

00:02:57.033 --> 00:02:58.959
All right, a bit better,

00:02:58.959 --> 00:03:00.333
but then what if you decide,

00:03:00.333 --> 00:03:04.667
hey I want to make a PDF? Not great.

00:03:04.667 --> 00:03:06.000
And then the worst of all,

00:03:06.000 --> 00:03:07.840
the author doesn't give you the source,

00:03:07.840 --> 00:03:10.159
and then you have to view page source,

00:03:10.159 --> 00:03:13.760
and learn cascading style sheets,

00:03:14.239 --> 00:03:15.767
and sit in a corner and cry,

00:03:15.767 --> 00:03:18.080
and decide to do other things with your life.

00:03:18.080 --> 00:03:22.640
We want to give you Org users numerous styles

00:03:22.640 --> 00:03:27.200
and an extensible mechanism to add more of these

00:03:27.200 --> 00:03:28.799
aesthetically pleasing styles,

00:03:28.799 --> 00:03:31.200
to have really nice things

00:03:31.200 --> 00:03:33.933
look one way in the HTML

00:03:33.933 --> 00:03:36.567
and look almost the same way in the PDF

00:03:36.567 --> 00:03:38.667
and other back ends.

00:03:38.667 --> 00:03:41.680
And if by having these newer ones,

00:03:41.680 --> 00:03:45.519
people might be encouraged to try making new ones,

00:03:45.519 --> 00:03:49.040
especially when the interface is not so difficult,

00:03:49.040 --> 00:03:54.159
that's the aim.

00:03:54.159 --> 00:03:59.120
So let's have a real story to motivate this even more.

00:03:59.120 --> 00:04:03.533
Here's three friends. I hope I don't butcher their names,

00:04:03.533 --> 00:04:04.640
but these friends are called

00:04:04.640 --> 00:04:08.720
Amin, Sacha, and Corwin. They're organizing a conference,

00:04:08.720 --> 00:04:12.080
EmacsConf 2020.

00:04:12.080 --> 00:04:14.239
So Sacha decides to write an Org file

00:04:14.239 --> 00:04:16.479
and she would like some feedback.

00:04:16.479 --> 00:04:19.840
Okay. Just to make it clear, there's no...

00:04:19.840 --> 00:04:22.960
just how easy this looks,

00:04:22.960 --> 00:04:27.280
let's look at the source for this block.

00:04:27.280 --> 00:04:30.720
Notice it's just the word "green," then a colon,

00:04:30.720 --> 00:04:37.333
then Amin. No div style coloring,

00:04:37.333 --> 00:04:38.479
just green:Amin.

00:04:38.479 --> 00:04:41.520
A very pleasant Org markup.

00:04:41.520 --> 00:04:44.960
So that's quite nice. Put some bold around it.

00:04:44.960 --> 00:04:46.433
Not too difficult.

00:04:46.433 --> 00:04:49.900
Hopefully, this will be useful to other people as well.

00:04:49.900 --> 00:04:52.240
So what kind of feedback

00:04:52.240 --> 00:04:55.233
would Sacha expect to get?

00:04:55.233 --> 00:04:59.120
Maybe she would expect top-level remarks

00:04:59.120 --> 00:05:00.833
visible in the export.

00:05:00.833 --> 00:05:04.400
When she makes an HTML, she can see right there a big block.

00:05:04.400 --> 00:05:09.533
Right. Maybe Amin will suggest to Sacha,

00:05:09.533 --> 00:05:11.120
please replace this part

00:05:11.120 --> 00:05:12.267
with this other part

00:05:12.267 --> 00:05:13.333
or replace this word

00:05:13.333 --> 00:05:14.639
with this other word.

00:05:14.639 --> 00:05:16.960
This is not really possible

00:05:16.960 --> 00:05:21.919
with raw HTML or with even LaTeX.

00:05:21.919 --> 00:05:24.000
You'd have to have multiple arguments:

00:05:24.000 --> 00:05:25.360
the first argument, and then

00:05:25.360 --> 00:05:28.800
the replacement argument. It's a bit clunky.

00:05:28.800 --> 00:05:32.080
But with our setup, you just write some text,

00:05:32.080 --> 00:05:34.367
write #+replace_with

00:05:34.367 --> 00:05:36.240
and then write more text, and you're good to go.

00:05:36.240 --> 00:05:39.440
Normal Org markup.

00:05:39.440 --> 00:05:40.720
Everyone speaks different languages.

00:05:40.720 --> 00:05:43.833
Maybe they want to use one word,

00:05:43.833 --> 00:05:45.919
or they're arguing about

00:05:45.919 --> 00:05:48.000
whether we talk about frames or windows,

00:05:48.000 --> 00:05:50.560
so maybe they want to have some translations.

00:05:50.560 --> 00:05:53.433
So there are different kinds of feedback.

00:05:53.433 --> 00:05:55.360
Let's take an example.

00:05:55.360 --> 00:05:59.360
Look at what they are.

00:05:59.360 --> 00:06:02.560
For example, Sacha might write

00:06:02.560 --> 00:06:06.160
this Org Mode right here,

00:06:06.160 --> 00:06:09.759
and then in her HTML exports, you might see this,

00:06:09.759 --> 00:06:12.733
and her feedback might look really nicely

00:06:12.733 --> 00:06:13.840
from anyone who says

00:06:14.560 --> 00:06:17.400
let's do some Lisp instead of mathematics.

00:06:17.400 --> 00:06:18.560
Let's just do some Lisp.

00:06:18.560 --> 00:06:22.479
Corwin says, let's not be so silly.

00:06:22.479 --> 00:06:25.120
Let's just say 9 a.m. and move on.

00:06:28.080 --> 00:06:31.360
Amin likes to export to PDF,

00:06:31.360 --> 00:06:34.333
and so he writes his top-level remarks using LaTeX.

00:06:34.333 --> 00:06:36.900
That's how. To get this square

00:06:36.900 --> 00:06:38.960
Amin: please change whatever,

00:06:38.960 --> 00:06:43.120
he might write like this: #+latex:.

00:06:43.120 --> 00:06:50.880
But then Sacha only exports to HTML, for example,

00:06:50.880 --> 00:06:54.880
so she doesn't look at the PDF,

00:06:54.880 --> 00:06:57.867
and she may not see his top-level feedback

00:06:57.867 --> 00:07:00.667
with those nice brackets and and bold.

00:07:00.667 --> 00:07:03.120
She might think everything's good.

00:07:03.120 --> 00:07:06.160
That can be a bit disastrous.

00:07:06.160 --> 00:07:08.600
So maybe Sacha will then

00:07:08.600 --> 00:07:13.199
make some of her own feedback.

00:07:13.199 --> 00:07:16.160
To produce it, she might write

00:07:16.560 --> 00:07:21.680
HTML commands, #+html: to get that.

00:07:21.680 --> 00:07:24.100
But then Amin will make a PDF,

00:07:24.100 --> 00:07:25.680
and this won't stick out.

00:07:25.680 --> 00:07:28.367
So he might think everything's okay,

00:07:28.367 --> 00:07:30.160
even though it's not.

00:07:30.160 --> 00:07:33.100
Then Corwin actually decides,

00:07:33.100 --> 00:07:35.900
"Hey, let me read the exported result

00:07:35.900 --> 00:07:38.867
and there's all those feedback
from two people

00:07:38.867 --> 00:07:42.633
who haven't read anything,
because maybe they were in a rush,

00:07:42.633 --> 00:07:45.167
and didn't see the top-level feedback.

00:07:45.167 --> 00:07:50.000
So they agree. "Hey, let's have a
uniform Org interface that exports

00:07:50.000 --> 00:07:52.567
to both HTML and PDF.

00:07:52.567 --> 00:07:54.080
Make both of us happy."

00:07:54.080 --> 00:07:58.160
Okay. So they decide to use
Org special blocks.

00:07:58.160 --> 00:08:01.300
Right. To set this up,

00:08:01.300 --> 00:08:04.400
they need to read a little bit of Lisp,

00:08:04.400 --> 00:08:08.879
hooks, advice, macros to get all of this
set up,

00:08:08.879 --> 00:08:12.319
and then they'll use Org as the main
interface.

00:08:12.319 --> 00:08:16.479
It's a lot of work, but it's worth it,
right? maybe?

00:08:16.479 --> 00:08:19.360
But then Corwin's a bit terse.

00:08:19.759 --> 00:08:24.333
Corwin maybe doesn't want to
write using blocks.

00:08:24.333 --> 00:08:26.080
He thinks they're overkill.

00:08:26.080 --> 00:08:32.560
Sacha wants HTML, and Amin wants PDF,
and Corwin wants org-reveal.

00:08:32.560 --> 00:08:35.533
So now they have to reformat
all their code.

00:08:35.533 --> 00:08:36.900
And then they need to use org link types

00:08:36.900 --> 00:08:38.867
to reduce the overkill,

00:08:38.867 --> 00:08:41.367
so they can try to avoid duplication

00:08:41.367 --> 00:08:46.800
by factoring things out into
self-contained functions.

00:08:46.800 --> 00:08:50.320
But now, to set up our links,

00:08:50.320 --> 00:08:55.467
we'll have to learn a new interface, org
setup link.

00:08:55.467 --> 00:09:02.160
Learn a little bit about fonts, follow
links, export handlers...

00:09:02.160 --> 00:09:05.040
It's so much. That's so much. But then,

00:09:05.040 --> 00:09:06.800
the friends, they learn a lot.

00:09:06.800 --> 00:09:08.399
They learn about defun.

00:09:08.399 --> 00:09:11.120
So these words are red.

00:09:11.120 --> 00:09:12.185
You get a little explanation.

00:09:12.185 --> 00:09:14.320
I think it's a bit too small for anyone
to read.

00:09:14.320 --> 00:09:18.000
This is Lisp documentation for defun.

00:09:18.000 --> 00:09:21.600
advice-add. There's some Lisp
documentation.

00:09:21.600 --> 00:09:23.600
They learn about destructuring -let.

00:09:23.600 --> 00:09:25.279
This is from the dash library.

00:09:25.279 --> 00:09:26.959
Here's all that glorious,

00:09:26.959 --> 00:09:28.800
glorious documentation with examples.

00:09:28.800 --> 00:09:30.300
Sorry. I like that.

00:09:30.300 --> 00:09:32.467
They might make
an ad-hoc mechanism

00:09:32.467 --> 00:09:35.400
to simulate arguments for special blocks,

00:09:35.400 --> 00:09:38.500
so something maybe called
extract-arguments,

00:09:38.500 --> 00:09:41.533
and then, of course, to make
new link types,

00:09:41.533 --> 00:09:42.480
they have to learn about

00:09:42.480 --> 00:09:47.400
org-link-set-parameters and
its numerous bits and pieces.

00:09:49.920 --> 00:09:53.600
Let's close all these ones down.

00:09:53.600 --> 00:09:55.800
Of course they also need to be
comfortable

00:09:55.800 --> 00:09:59.920
with loops and maps and matching and
string functions.

00:09:59.920 --> 00:10:02.560
So it's a bit of a pain.

00:10:03.360 --> 00:10:05.839
It's probably not worth it.

00:10:05.839 --> 00:10:07.767
Maybe I'll just rush things quickly,

00:10:07.767 --> 00:10:09.360
or do it ad-hoc...

00:10:10.320 --> 00:10:13.680
We have things to do.

00:10:13.680 --> 00:10:19.367
But maybe the squad wants to have a
modular and unified interface

00:10:19.367 --> 00:10:23.700
so everyone's comfortable with defun to
define a function

00:10:23.700 --> 00:10:29.440
and they say, "It would be nice if we
could just define simultaneously

00:10:29.440 --> 00:10:32.959
both a block and the link type."

00:10:32.959 --> 00:10:36.000
That way, we have a single interface

00:10:36.000 --> 00:10:37.867
Org mode, for these things.

00:10:37.867 --> 00:10:39.767
It would be nice if it was modular.

00:10:39.767 --> 00:10:44.633
If I defined a one kind of block and
you defined another,

00:10:44.633 --> 00:10:45.519
we could compose them,

00:10:45.519 --> 00:10:49.360
then get a nice bigger block, like LEGO.

00:10:49.360 --> 00:10:52.320
That would be nice. Building blocks.

00:10:52.320 --> 00:10:56.240
This is what we have come up with,
called defblock.

00:10:56.240 --> 00:11:01.760
It also has a long documentation string
containing examples and things.

00:11:01.760 --> 00:11:04.800
So that way, it can try to be useful.

00:11:04.800 --> 00:11:10.880
Let's look at a solution to these
friends' trilemma.

00:11:10.880 --> 00:11:14.320
So here's a way to define a block.

00:11:14.320 --> 00:11:22.320
It doesn't look that difficult, but this
is how they can define a block

00:11:22.320 --> 00:11:25.920
for their top-level feedback.

00:11:25.920 --> 00:11:28.959
Let's look at the three main parts
together.

00:11:28.959 --> 00:11:31.233
It's not that difficult, I hope.

00:11:31.233 --> 00:11:35.300
Just six lines, and that's including a
documentation string,

00:11:35.300 --> 00:11:37.633
newlines and things.

00:11:37.633 --> 00:11:43.300
So in line 1, we define the block just
like you define a function.

00:11:43.300 --> 00:11:44.880
We define a block.

00:11:44.880 --> 00:11:47.433
The block name is going to be called
"feedback."

00:11:47.433 --> 00:11:49.680
It has an author, "who."

00:11:49.680 --> 00:11:54.133
The author has no default value.

00:11:54.133 --> 00:11:57.760
It has a color, and the color has a
default value of red.

00:11:57.760 --> 00:12:01.680
So just as when you define functions,

00:12:01.680 --> 00:12:06.233
you start by define or defblock,

00:12:06.233 --> 00:12:13.440
then the name, some mandatory argument,
and some optional arguments.

00:12:13.440 --> 00:12:18.480
Then the next stage is definition.
Documentation.

00:12:18.480 --> 00:12:20.133
The people who use this,

00:12:20.133 --> 00:12:22.880
which are future you or future me,

00:12:22.880 --> 00:12:25.519
might want to know what this is.

00:12:25.519 --> 00:12:27.839
So let's get to document this.

00:12:27.839 --> 00:12:32.079
For Corwin, who might want to use
tooltips...

00:12:32.079 --> 00:12:37.120
When Corwin writes feedback in Emacs,
they'll see a nice little tooltip,

00:12:37.120 --> 00:12:38.639
and the tooltip will have

00:12:38.639 --> 00:12:41.279
this documentation string.

00:12:41.279 --> 00:12:43.279
That'll be nice.

00:12:43.279 --> 00:12:45.980
And then here's the third part.

00:12:45.980 --> 00:12:48.067
The last three lines are not so
difficult.

00:12:48.067 --> 00:12:52.800
If the backend is HTML,

00:12:52.800 --> 00:12:55.360
please use this template string.

00:12:55.360 --> 00:12:57.440
Otherwise, use the other string.

00:12:57.440 --> 00:13:01.279
For each of these string markers,

00:13:01.279 --> 00:13:04.959
please put in the color, who wrote it,
and then the contents

00:13:04.959 --> 00:13:07.279
of the special block or the link type.

00:13:08.160 --> 00:13:11.600
So that's pretty neat. Not so difficult.

00:13:11.600 --> 00:13:14.639
I thought that was kind of cool,

00:13:14.639 --> 00:13:16.600
then noticed it's anaphoric.

00:13:16.600 --> 00:13:21.033
This defblock gives you two new names.

00:13:21.033 --> 00:13:23.433
It gives you a name called contents,

00:13:23.433 --> 00:13:26.480
and it gives you a name called backend.

00:13:26.480 --> 00:13:29.733
So even if you're writing a defblock

00:13:29.733 --> 00:13:32.560
and you intend it to be used
only for links...

00:13:32.560 --> 00:13:35.440
Like these colors, for example.

00:13:35.440 --> 00:13:38.399
These colors were defined using defblock.

00:13:38.399 --> 00:13:41.279
I used them as links right here.

00:13:41.279 --> 00:13:43.360
You don't need to worry

00:13:43.360 --> 00:13:45.300
where does the text come from
in the link.

00:13:45.300 --> 00:13:48.959
If I say "red:Bob," is it Bob?

00:13:48.959 --> 00:13:52.000
Or if I put a description, is it the
description?

00:13:52.000 --> 00:13:54.000
So it's whatever is available will

00:13:54.000 --> 00:13:56.720
become the value of contents.

00:13:56.720 --> 00:13:59.199
If you're really interested

00:13:59.199 --> 00:14:02.433
and you want to do some intricate stuff,

00:14:02.433 --> 00:14:06.933
defblock also gives you something called
raw-contents,

00:14:06.933 --> 00:14:08.633
if you really want to touch

00:14:08.633 --> 00:14:12.639
the raw contents with all of the Org
markups still there.

00:14:12.639 --> 00:14:19.440
Let's see how everyone can communicate
amongst themselves

00:14:19.440 --> 00:14:22.480
using this new interface.

00:14:22.480 --> 00:14:26.000
So, Sacha speculates and she... How does

00:14:26.000 --> 00:14:28.399
she speculate for her Org HTML?

00:14:28.399 --> 00:14:33.733
She might just write. Hey look at that,
no HTML, nice.

00:14:34.800 --> 00:14:36.833
Amin wants to have some green,

00:14:36.833 --> 00:14:39.600
and so he just says, hey here's some
color green.

00:14:39.600 --> 00:14:42.959
There you go. It looks almost the same.

00:14:42.959 --> 00:14:48.267
Notice that the main argument is right
here.

00:14:48.267 --> 00:14:49.680
defblock took an author,

00:14:49.680 --> 00:14:51.333
and here's the author again.

00:14:51.333 --> 00:14:53.920
And now the optional argument

00:14:53.920 --> 00:14:57.007
uses the org babel source interface

00:14:57.007 --> 00:15:02.867
You just say :, then a key, and then the
argument. Quite nice.

00:15:02.867 --> 00:15:07.920
Corwin doesn't want to use blocks.
It's a bit of an overkill.

00:15:07.920 --> 00:15:12.959
He can just write a link.

00:15:12.959 --> 00:15:17.440
So the main argument is now
the label of the link,

00:15:17.440 --> 00:15:23.667
and the description of the link is the
contents of the feedback.

00:15:23.667 --> 00:15:25.680
So that was quite nice.

00:15:25.680 --> 00:15:29.360
So it looks like everyone uses the same
interface on the left

00:15:29.360 --> 00:15:32.800
and can have varying outputs.

00:15:32.800 --> 00:15:34.480
I think it looks quite nice,

00:15:34.480 --> 00:15:36.639
and I hope you do too.

00:15:36.639 --> 00:15:38.800
There's a few more.

00:15:38.800 --> 00:15:41.800
Maybe, as you saw in some previous ones,

00:15:41.800 --> 00:15:43.920
we had text side beside side,

00:15:43.920 --> 00:15:47.440
or we folded some regions away.

00:15:47.440 --> 00:15:50.959
We put some things in pretty boxes.

00:15:50.959 --> 00:15:57.120
We had some spoilers at the very
beginning that we hid some text.

00:15:57.120 --> 00:16:01.680
We demoed some texts. Here's some Org
and here's what it looks like,

00:16:01.680 --> 00:16:05.199
and most importantly, they compose.

00:16:05.199 --> 00:16:12.639
There's a a macro called thread-block.

00:16:12.639 --> 00:16:17.000
thread-block call, and it lets you
thread the contents

00:16:17.000 --> 00:16:18.000
through a number of blocks,

00:16:18.000 --> 00:16:20.639
treating them as if they were functions.

00:16:20.639 --> 00:16:22.480
So, really, you can think of a block

00:16:23.680 --> 00:16:25.567
as a string-valued function.

00:16:25.567 --> 00:16:28.533
That's pretty neat, I think.

00:16:28.533 --> 00:16:30.959
Thank you for listening.

00:16:31.759 --> 00:16:34.320
I hope you've enjoyed this little

00:16:34.880 --> 00:16:38.160
happy fun time with the Emacs and
friends.

00:16:38.160 --> 00:16:43.730
I'll happily answer questions right now.

00:16:45.360 --> 00:16:49.467
Someone says: "Why did you put
optional arguments

00:16:49.467 --> 00:16:50.480
in a separate list

00:16:50.480 --> 00:16:54.560
rather than using cl-style argument
lists?"

00:16:54.560 --> 00:16:58.399
So that's a very good question,

00:16:58.399 --> 00:17:00.000
and I will answer that

00:17:00.000 --> 00:17:05.467
by showing you a more involved
definition of feedback.

00:17:05.467 --> 00:17:14.567
Let's look at a more involved one right
here.

00:17:14.567 --> 00:17:19.280
So, for example, this one is
called rremark.

00:17:19.280 --> 00:17:23.439
Please let me know if my text is not
sufficiently big.

00:17:23.439 --> 00:17:28.033
Here is why we have two arguments.

00:17:28.033 --> 00:17:30.720
That takes two arguments instead of one

00:17:30.720 --> 00:17:33.360
for its argument list.

00:17:33.360 --> 00:17:34.799
You have def block,

00:17:34.799 --> 00:17:36.000
then you have the name,

00:17:36.000 --> 00:17:40.467
then you have the first argument list

00:17:40.467 --> 00:17:42.880
and the second argument list.

00:17:42.880 --> 00:17:46.080
The first argument list

00:17:46.080 --> 00:17:49.280
takes the text right after the begin.

00:17:49.280 --> 00:17:53.000
The text right after the begin is the
main argument.

00:17:53.000 --> 00:17:59.200
And then the remaining key-value pairs
are in the second argument list.

00:18:00.320 --> 00:18:03.280
Now the reason we have two is because

00:18:03.280 --> 00:18:08.880
in order to streamline the interface to
account for both special blocks

00:18:08.880 --> 00:18:13.360
and Org link types, what we do is we say,

00:18:13.360 --> 00:18:18.000
in the first argument list, you can give
a name to the first argument,

00:18:18.000 --> 00:18:19.633
give it a default value,

00:18:19.633 --> 00:18:28.800
and anything else you provide will
become part of the link information.

00:18:28.800 --> 00:18:30.861
For example, this link,

00:18:30.861 --> 00:18:32.833
we decided to make its face
angry red.

00:18:32.833 --> 00:18:36.433
You might want to give other
features to links.

00:18:36.433 --> 00:18:39.100
So we're trying to streamline
the interface

00:18:39.100 --> 00:18:41.733
for both special blocks and
Org link types,

00:18:41.733 --> 00:18:46.240
and we thought this way was quite nice.

00:18:46.240 --> 00:18:47.500
That was the main reason.

00:18:47.500 --> 00:18:52.480
Someone asks--

00:18:52.480 --> 00:18:55.039
if you have follow-ups, please ask--

00:18:55.039 --> 00:18:57.600
Someone asks, "Do you intend to try to

00:18:57.600 --> 00:19:00.559
upstream this amazing work into Org?"

00:19:00.559 --> 00:19:02.300
Well, I'm glad you like it.

00:19:02.300 --> 00:19:04.559
I don't know how to upstream,

00:19:04.559 --> 00:19:06.400
but I will look into it,

00:19:06.400 --> 00:19:09.833
and any advice or guidance
would be much appreciated.

00:19:11.840 --> 00:19:15.267
Lisp is awesome. Just as
defun is a macro,

00:19:15.267 --> 00:19:17.120
defblock is a macro, and then

00:19:17.120 --> 00:19:20.240
source blocks are awesome.

00:19:20.240 --> 00:19:22.467
Now maybe we can have arguments in
special blocks,

00:19:22.467 --> 00:19:28.799
and motivate and encourage more
people to learn Lisp.

00:19:28.799 --> 00:19:32.559
So another person asks,

00:19:32.559 --> 00:19:35.280
"What is used to produce colorful

00:19:35.280 --> 00:19:38.559
boxes around the cursor?"

00:19:38.559 --> 00:19:40.400
I'm not quite sure if you're asking...

00:19:40.400 --> 00:19:42.559
Are you talking about my cursor
right here,

00:19:42.559 --> 00:19:48.400
or are you talking about in the slide?

00:19:48.400 --> 00:19:53.767
So this cursor is some application
called Streambrush,

00:19:53.767 --> 00:19:55.440
that I had to purchase.

00:19:55.440 --> 00:19:59.039
Unfortunately, I could not find a a
suitable free one.

00:19:59.039 --> 00:20:03.067
The blocks... I can demonstrate some
Emacs Lisp.

00:20:03.067 --> 00:20:04.467
I can open up my Emacs, if people like,

00:20:04.467 --> 00:20:06.320
and we can try some things out.

00:20:06.320 --> 00:20:09.440
Happy to do that.

00:20:09.440 --> 00:20:10.133
You're welcome.

00:20:10.133 --> 00:20:15.520
Someone asks a side question about
org-reveal: "How do you get

00:20:15.520 --> 00:20:17.440
bespoke or multiple-column layouts

00:20:17.440 --> 00:20:19.120
without using HTML?"

00:20:19.120 --> 00:20:22.559
Excellent question. That's what we do.

00:20:22.559 --> 00:20:25.533
That's what this project is about.

00:20:25.533 --> 00:20:27.000
So it's not org-reveal,

00:20:27.000 --> 00:20:30.267
it's our fancy parallel block.

00:20:30.267 --> 00:20:33.440
So we have this thing. You say,
#+begin_parallel.

00:20:33.440 --> 00:20:35.679
You say how many columns you would like.

00:20:35.679 --> 00:20:37.967
Do you want a bar or not?

00:20:37.967 --> 00:20:39.679
And then you write some text,

00:20:39.679 --> 00:20:44.400
and then you get some text, and
according with the bar or not.

00:20:44.400 --> 00:20:47.520
That's how we achieve that in our slides.

00:20:47.520 --> 00:20:52.880
I'm not quite sure where this was.

00:20:52.880 --> 00:20:59.520
Somewhere here, I think.

00:20:59.520 --> 00:21:06.240
Let me try to find this for you.

00:21:06.240 --> 00:21:10.433
I can't seem to find where the parallel
blocks were. Apologies.

00:21:10.433 --> 00:21:15.039
Let's move on to the next question,
I suppose.

00:21:15.039 --> 00:21:18.400
I'm pretty sure they're here. Ah, there
they are.

00:21:18.400 --> 00:21:22.640
So these were just instances of using

00:21:22.640 --> 00:21:26.480
the parallel block, and it makes things
parallel.

00:21:26.480 --> 00:21:27.633
So that's quite nice.

00:21:27.633 --> 00:21:33.360
Another person asks,

00:21:33.360 --> 00:21:37.840
"How does this relate to pandoc,

00:21:37.840 --> 00:21:40.960
which is used for converting between
markup formats?"

00:21:40.960 --> 00:21:43.919
So all we're doing is we're saying,

00:21:43.919 --> 00:21:47.679
hey, please write Org because Org is
just fantastic,

00:21:47.679 --> 00:21:49.267
and we love it, and it's the dream,

00:21:49.267 --> 00:21:51.760
and if you would like to view things

00:21:51.760 --> 00:21:55.900
in HTML, or in org-reveal, or in PDF,

00:21:55.900 --> 00:21:58.559
that's up to the user.

00:22:02.320 --> 00:22:06.080
Made it too small now.
So here is an example.

00:22:06.080 --> 00:22:10.240
Here's how parallel is implemented,

00:22:10.240 --> 00:22:14.320
just as a quick example, not too long.

00:22:14.320 --> 00:22:17.800
About half of the implementation is
documentation,

00:22:17.800 --> 00:22:22.720
so, hopefully, that speaks for for how
useful this feature is.

00:22:22.720 --> 00:22:25.280
We decide if there's a rule or not.

00:22:25.280 --> 00:22:28.080
We look for the column break.

00:22:28.080 --> 00:22:30.600
Here we're looking at the backend.

00:22:30.600 --> 00:22:31.840
If the backend is LaTeX,

00:22:31.840 --> 00:22:34.133
please use this incantation

00:22:34.133 --> 00:22:37.679
with multicolumns, minipages,
what have you.

00:22:37.679 --> 00:22:41.600
If the backend is something else, please
do this:

00:22:41.600 --> 00:22:48.080
div, style and other gibberish that we
don't really want to look at.

00:22:48.080 --> 00:22:51.760
Pandoc works from Org,

00:22:51.760 --> 00:22:53.633
so it might not work directly,

00:22:53.633 --> 00:22:59.679
since our interface... The way we set it
up is: when you try to export,

00:22:59.679 --> 00:23:03.039
we hook in and we do a bunch of
pre-processing,

00:23:03.039 --> 00:23:07.440
so this defblock is a
string-valued function.

00:23:07.440 --> 00:23:13.919
Whenever we see these #+begin_parallel
when you do an export,

00:23:13.919 --> 00:23:17.767
I tell Emacs, hold up, look for those
#+begin_parallels, please.

00:23:17.767 --> 00:23:20.320
Oh, you found them? Grab that text.

00:23:20.320 --> 00:23:21.533
You grabbed it. Great.

00:23:21.533 --> 00:23:24.080
Now please apply this person's function

00:23:24.080 --> 00:23:27.120
onto that text, and splice in the result.

00:23:27.120 --> 00:23:30.400
So when you export, we're performing

00:23:30.400 --> 00:23:35.120
arbitrary computations on your text.

00:23:35.120 --> 00:23:39.633
Some people might not find that
comforting,

00:23:39.633 --> 00:23:43.039
to have arbitrary computations happening.

00:23:43.039 --> 00:23:45.039
In this article, there's a few where

00:23:45.039 --> 00:23:47.167
we change your text upon export.

00:23:47.167 --> 00:23:51.760
We translate it, we do other things
to it.

00:23:51.760 --> 00:23:56.500
So someone says, "If you export to
LaTeX, to PDF,

00:23:56.500 --> 00:23:58.640
does that work well with Beamer as well

00:23:58.640 --> 00:24:00.320
to create slides with columns?"

00:24:05.200 --> 00:24:08.000
I made a bunch of these changes

00:24:08.000 --> 00:24:09.200
earlier this morning,

00:24:09.200 --> 00:24:12.320
and it just says LaTeX right here.

00:24:12.320 --> 00:24:14.400
So if you want to go to beamer,

00:24:14.400 --> 00:24:15.360
I think the back end for me,

00:24:15.360 --> 00:24:17.333
beamer is called, well, beamer,

00:24:17.333 --> 00:24:22.000
so instead of a pcase, what we would do
is, we would say,

00:24:22.000 --> 00:24:27.167
if it's a 'latex or it's a 'beamer, then
use this.

00:24:27.167 --> 00:24:30.267
Otherwise, it's not a LaTeX,

00:24:30.267 --> 00:24:31.867
it will simply default to this one,

00:24:31.867 --> 00:24:34.433
which could be dangerous
for your needs.

00:24:34.433 --> 00:24:39.167
I think it's a bad practice to put a
underscore,

00:24:39.167 --> 00:24:40.767
but I did it really quickly

00:24:40.767 --> 00:24:44.500
because I just wanted to show you that
it works fine in org-reveal

00:24:44.500 --> 00:24:47.440
Contributions are more than welcome.

00:24:47.440 --> 00:24:52.240
I happily would love any assistance.

00:24:52.240 --> 00:24:58.633
We have a Lisp reference
cheat sheet here

00:24:58.633 --> 00:25:01.000
to learn a little bit about Lisp, if
you're not comfortable,

00:25:01.000 --> 00:25:03.267
or to ask some questions.

00:25:03.267 --> 00:25:06.400
Lots of helpful people.

00:25:06.400 --> 00:25:09.440
So there's another question that says,

00:25:09.440 --> 00:25:13.120
"Does typing in a block mess up with
syntax highlighting?

00:25:13.120 --> 00:25:15.679
Usually, you use a single color inside an

00:25:15.679 --> 00:25:17.279
example block, for example.

00:25:17.279 --> 00:25:21.279
Ah, you found my crutch.

00:25:25.279 --> 00:25:27.333
Emacs is all encompassing,

00:25:27.333 --> 00:25:29.760
and I'm not quite sure how fonts work.

00:25:29.760 --> 00:25:32.559
I learned enough to get by.

00:25:37.440 --> 00:25:38.667
Here's how links work.

00:25:38.667 --> 00:25:40.799
They're a bit complicated.

00:25:40.799 --> 00:25:42.567
This is a bit scary.

00:25:42.567 --> 00:25:47.039
I don't recommend anyone read it.

00:25:47.039 --> 00:25:49.840
Actually, let me open up an email

00:25:50.559 --> 00:25:52.100
and you can see what I see.

00:25:52.100 --> 00:25:54.799
So here's an Emacs.

00:25:54.799 --> 00:25:56.799
Let's make that a bit bigger.

00:25:56.799 --> 00:25:59.133
Let's change this slightly.

00:25:59.133 --> 00:26:01.200
Nope, that's worse. There you go.

00:26:01.919 --> 00:26:09.360
Here's some words. Here's red hello.

00:26:09.360 --> 00:26:15.679
But you're worried about preserving
fontification.

00:26:15.679 --> 00:26:18.480
Let's make an emacs-lisp block.

00:26:18.880 --> 00:26:22.840
Let's say, (+ 1 2).

00:26:22.840 --> 00:26:31.133
Ah, where's the fun? Hello. Bye.

00:26:31.133 --> 00:26:32.080
Okay. Where's the coloring?

00:26:32.080 --> 00:26:36.000
If we zoom in on this #+begin_src block,

00:26:36.000 --> 00:26:39.200
you can see down here

00:26:39.200 --> 00:26:42.159
we have our our coloring

00:26:42.159 --> 00:26:43.279
when we zoom in.

00:26:43.279 --> 00:26:46.960
If we zoom out, no coloring.

00:26:46.960 --> 00:26:50.880
Zoom in, coloring.
Zoom out, aah, no coloring.

00:26:50.880 --> 00:26:55.679
Let's take off these bad boys,
and oh, look, my coloring's back.

00:26:55.679 --> 00:27:03.760
In a previous iteration of the system,
I was able to maintain coloring.

00:27:03.760 --> 00:27:06.400
In this new iteration, I am not.

00:27:06.400 --> 00:27:07.400
I don't know how to do it.

00:27:07.400 --> 00:27:10.333
I haven't had the time to implement it.

00:27:10.333 --> 00:27:17.279
I spent a lot of time writing this
48-page documentation

00:27:17.279 --> 00:27:21.133
with some fun examples to try to help
people learn.

00:27:21.133 --> 00:27:23.200
But I would appreciate any help or
guidance

00:27:23.200 --> 00:27:26.240
on how to maintain the fontification.

00:27:26.240 --> 00:27:29.200
I really would like to keep those
colors in.

00:27:29.200 --> 00:27:32.640
[Amin]: Musa, we have time for maybe one
more question,

00:27:32.640 --> 00:27:34.500
one or two more questions,

00:27:34.500 --> 00:27:37.039
and then we have to move on to the
next talk.

00:27:37.039 --> 00:27:39.120
You're more than welcome to

00:27:39.120 --> 00:27:42.559
continue taking the questions via
IRC or the pad.

00:27:42.559 --> 00:27:45.760
[Musa]: Okay. Thank you.

00:27:45.760 --> 00:27:48.880
The final question we'll take is,

00:27:48.880 --> 00:27:52.320
"Should packages implement

00:27:52.320 --> 00:27:53.967
interface to one specific format,

00:27:53.967 --> 00:27:55.600
or attempt to be inclusive

00:27:55.600 --> 00:27:57.279
to all the potential output targets?"

00:27:57.279 --> 00:27:59.300
I think you should just make them
as you go,

00:27:59.300 --> 00:28:01.500
and add them as you need them.

00:28:01.500 --> 00:28:05.600
We'll make Github requests for things.

00:28:05.600 --> 00:28:08.533
We can share recipes in this document,

00:28:08.533 --> 00:28:12.333
and then try to add other techniques,

00:28:12.333 --> 00:28:19.200
and then we can use these blocks as a
common interface

00:28:19.200 --> 00:28:22.240
for exporting to PDF and other things.

00:28:22.240 --> 00:28:26.000
Since someone asked,

00:28:26.000 --> 00:28:28.033
here what a PDF looks like.

00:28:28.033 --> 00:28:31.667
This is the same PDF rendered.

00:28:31.667 --> 00:28:34.960
I made no effort to make it look good,

00:28:34.960 --> 00:28:37.840
but it surprisingly does look good.

00:28:38.559 --> 00:28:40.067
That was nice.

00:28:40.067 --> 00:28:44.320
That was a terrible magenta,
but that is life.

00:28:44.320 --> 00:28:47.100
Anyhow, I hope you all enjoyed this talk.

00:28:47.100 --> 00:28:51.033
I hope you will find
defblock useful to you.

00:28:51.033 --> 00:28:52.799
It is available on MELPA.

00:28:52.799 --> 00:28:56.367
In a rush to make it available for
EmacsConf 2020,

00:28:56.367 --> 00:29:00.159
some MELPA guidelines may not have been
adhered to.

00:29:00.159 --> 00:29:01.600
Please do not hit me.

00:29:01.600 --> 00:29:08.559
I hope everyone enjoys the rest of the
EmacsConf 2020. Thank you!