summaryrefslogtreecommitdiffstats
path: root/2022/captions/emacsconf-2022-detached--getting-detached-from-emacs--niklas-eklund--main.vtt
blob: 09c4a9df6d58446eea8fbf89a7a97c3a583fe98b (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
WEBVTT captioned by anush

00:00:00.000 --> 00:00:04.354
Hello, everyone! Welcome to my talk,

00:00:04.454 --> 00:00:07.221
"Getting detached from Emacs".

00:00:07.321 --> 00:00:09.720
When I started to use Emacs,

00:00:09.720 --> 00:00:12.920
I quickly gravitated towards using it

00:00:12.920 --> 00:00:15.000
as much as I could.

00:00:15.000 --> 00:00:17.880
Magit, Org, Dired,

00:00:17.880 --> 00:00:21.360
a lot of new possibilities opened up.

00:00:21.360 --> 00:00:23.280
However, there was a workflow

00:00:23.280 --> 00:00:27.120
that was difficult for me to replace.

00:00:27.120 --> 00:00:30.120
The problem for me was running shell commands

00:00:30.120 --> 00:00:32.560
in sub-processes of Emacs,

00:00:32.560 --> 00:00:34.880
which in some situations led me to stick to

00:00:34.880 --> 00:00:38.000
using an external terminal.

00:00:38.000 --> 00:00:39.588
These situations,

00:00:39.688 --> 00:00:43.254
often revolved around long-running shell commands,

00:00:43.354 --> 00:00:46.054
either on my local machine

00:00:46.154 --> 00:00:48.600
or on a remote host.

00:00:48.600 --> 00:00:50.240
When I was on a remote host,

00:00:50.240 --> 00:00:52.800
I would also rely on using the program tmux

00:00:52.800 --> 00:00:57.280
to be able to detach from the remote process.

00:00:57.280 --> 00:00:59.521
My main concern at the time

00:00:59.621 --> 00:01:01.854
was that I didn't want to having to avoid

00:01:01.954 --> 00:01:03.440
restarting Emacs,

00:01:03.440 --> 00:01:07.800
because I needed to wait for a process to complete.

00:01:07.800 --> 00:01:11.200
However, there was of course a lot of things

00:01:11.200 --> 00:01:15.800
I was missing out on by not using Emacs.

00:01:15.800 --> 00:01:19.760
Therefore, my solution to resolving

00:01:19.760 --> 00:01:22.840
the issue of occasionally having to leave Emacs

00:01:22.840 --> 00:01:26.188
led me down the path of developing the package

00:01:26.188 --> 00:01:28.221
Detached.

00:01:28.321 --> 00:01:31.160
The package allows Emacs to delegate

00:01:31.160 --> 00:01:33.640
the responsibility of creating processes

00:01:33.640 --> 00:01:36.840
to the program dtach.

00:01:36.840 --> 00:01:39.021
It also makes sure to write

00:01:39.121 --> 00:01:44.321
the output of the process to a file,

00:01:44.421 --> 00:01:48.154
which we will see later on how that is being used.

00:01:48.254 --> 00:01:50.800
The package makes Emacs capable of

00:01:50.800 --> 00:01:53.120
attaching to these processes

00:01:53.120 --> 00:01:55.280
as well as managing them.

00:01:55.280 --> 00:01:58.880
In the package, each process is called a session,

00:01:58.880 --> 00:02:01.040
and inside of Emacs that is just

00:02:01.040 --> 00:02:02.560
an object with properties

00:02:02.560 --> 00:02:04.588
such as what command is being run,

00:02:04.688 --> 00:02:06.720
what working directory is used,

00:02:06.720 --> 00:02:10.240
where the output is stored, etc.

00:02:10.240 --> 00:02:12.480
The important aspect is also that

00:02:12.480 --> 00:02:16.840
these objects are being persistent,

00:02:16.840 --> 00:02:20.121
so they are stored over time.

00:02:20.221 --> 00:02:22.920
Today, I'm going to walk you through

00:02:22.920 --> 00:02:26.054
how I use the package and what advantages

00:02:26.154 --> 00:02:30.840
there are of treating processes like text.

00:02:30.840 --> 00:02:35.720
I'm going to start by opening up M-x shell,

00:02:35.720 --> 00:02:40.288
and I will run a command

00:02:40.388 --> 00:02:46.920
to update my package manager.

00:02:46.920 --> 00:02:48.254
Instead of pressing return,

00:02:48.354 --> 00:02:50.240
I'll simply press shift return

00:02:50.240 --> 00:02:52.600
to let Emacs delegate the execution

00:02:52.600 --> 00:02:54.920
to the dtach program.

00:02:54.920 --> 00:02:57.600
Emacs will immediately attach itself to the process,

00:02:57.600 --> 00:03:00.960
and we therefore don't perceive any difference

00:03:00.960 --> 00:03:04.480
from when running the command as a subprocess.

00:03:04.480 --> 00:03:06.488
We now have the option though

00:03:06.588 --> 00:03:09.800
to detach from the session,

00:03:09.800 --> 00:03:12.080
and later on we can of course

00:03:12.080 --> 00:03:17.054
reattach Emacs to the session.

00:03:17.154 --> 00:03:20.421
For me, this addresses the core

00:03:20.521 --> 00:03:23.800
of the problem that I had.

00:03:23.800 --> 00:03:25.040
But let’s see what’s more

00:03:25.040 --> 00:03:30.800
the new workflow inside of Emacs can bring.

00:03:30.800 --> 00:03:35.720
The package supports multiple user interfaces

00:03:35.720 --> 00:03:39.840
such as Eshell and Compile.

00:03:39.840 --> 00:03:43.360
I will therefore switch to the Detached project,

00:03:43.360 --> 00:03:49.440
and I will run the build command that I use.

00:03:49.440 --> 00:03:54.040
I will run it with detached-compile

00:03:54.040 --> 00:03:55.800
with the difference that

00:03:55.800 --> 00:03:59.800
I can detach from the compilation.

00:03:59.800 --> 00:04:05.000
One benefit of this new workflow is that

00:04:05.000 --> 00:04:08.680
I can get a system notification shown up here

00:04:08.680 --> 00:04:10.760
once a session has finished.

00:04:10.760 --> 00:04:15.520
Previously, I was either forced to have

00:04:15.520 --> 00:04:19.040
the terminal open so I could see it or hiding it,

00:04:19.040 --> 00:04:27.754
but then risking having forgotten it.

00:04:27.854 --> 00:04:32.560
How do we then see the output of a session?

00:04:32.560 --> 00:04:36.788
Get the detached-list-sessions command,

00:04:36.888 --> 00:04:42.854
and here we see

00:04:42.954 --> 00:04:45.600
the command that we just built,

00:04:45.600 --> 00:04:48.440
and we see the guix pull with an asterisk

00:04:48.440 --> 00:04:54.480
indicating that it is continuously running.

00:04:54.480 --> 00:05:01.421
If I press enter,

00:05:01.521 --> 00:05:04.440
we will get the output of the session here.

00:05:04.440 --> 00:05:07.200
And since it was run using compile,

00:05:07.200 --> 00:05:11.720
we also have compilation mode enabled here,

00:05:11.720 --> 00:05:14.188
so we could navigate between

00:05:14.288 --> 00:05:17.160
potential warnings or errors.

00:05:17.160 --> 00:05:21.440
And we see that there is a warning here.

00:05:21.440 --> 00:05:26.621
One thing that

00:05:26.721 --> 00:05:28.920
I have prepared here is that

00:05:28.920 --> 00:05:32.240
if I open up the user interface,

00:05:32.240 --> 00:05:35.760
we only see two sessions,

00:05:35.760 --> 00:05:38.988
but that is because we applied a filter here.

00:05:39.088 --> 00:05:43.160
So, we have actually the only sessions

00:05:43.160 --> 00:05:46.000
that are created within the last 12 hours

00:05:46.000 --> 00:05:47.360
and that are considered unique.

00:05:47.360 --> 00:05:50.600
So, if I remove the uniqueness,

00:05:50.600 --> 00:05:57.040
we also see that we have a previous build

00:05:57.040 --> 00:05:59.280
running on the main branch.

00:05:59.280 --> 00:06:02.760
So, I think that's typically normal

00:06:02.760 --> 00:06:07.080
that you might have that.

00:06:07.080 --> 00:06:10.920
And since the sessions can be considered text,

00:06:10.920 --> 00:06:14.760
we can just mark these two and check,

00:06:14.760 --> 00:06:18.880
does this warning exist on the main branch or not?

00:06:18.880 --> 00:06:22.720
So, we can just diff these ones

00:06:22.720 --> 00:06:27.454
and we see that the warning is only present

00:06:27.554 --> 00:06:34.040
on the emacsconf branch.

00:06:34.040 --> 00:06:37.520
Now, another benefit, in my opinion,

00:06:37.520 --> 00:06:41.840
of the new way of working is that

00:06:41.840 --> 00:06:43.240
I have these properties being displayed

00:06:43.240 --> 00:06:45.321
in the user interface.

00:06:45.421 --> 00:06:48.960
I can quickly see which commands are still running,

00:06:48.960 --> 00:06:51.360
what hosts they are running on,

00:06:51.360 --> 00:06:53.388
where they are running,

00:06:53.488 --> 00:06:58.121
and for how long they have been running.

00:06:58.221 --> 00:07:00.321
And if they have run,

00:07:00.421 --> 00:07:04.421
how long did it take?

00:07:04.521 --> 00:07:06.880
Occasionally, though, there might be

00:07:06.880 --> 00:07:16.880
even more input needed to distinguish sessions.

00:07:16.880 --> 00:07:19.600
So, what I typically do then is

00:07:19.600 --> 00:07:22.200
press A to annotate the session,

00:07:22.200 --> 00:07:28.988
I would add a "Warning found at emacsconf"

00:07:29.088 --> 00:07:32.854
And then it will show up this annotation

00:07:32.954 --> 00:07:44.988
in the echo area when I select the session.

00:07:45.088 --> 00:07:48.080
Another great improvement of using these sessions

00:07:48.080 --> 00:07:52.640
and consider them being text is,

00:07:52.640 --> 00:07:57.321
now we also see the guix pull completed here.

00:07:57.421 --> 00:08:00.921
Then we can also select and see that, okay,

00:08:01.021 --> 00:08:08.600
there is a lot of updates in this command,

00:08:08.600 --> 00:08:10.160
but let's not look at it now.

00:08:10.160 --> 00:08:17.554
Let's instead remember that

00:08:17.654 --> 00:08:19.840
previously last week

00:08:19.840 --> 00:08:21.640
when I ran a guix pull,

00:08:21.640 --> 00:08:27.680
I saw an Emacs package that looks interesting.

00:08:27.680 --> 00:08:30.040
I rather don't remember its full name,

00:08:30.040 --> 00:08:32.720
but it has something to do with collection.

00:08:32.720 --> 00:08:38.840
So, what I then can do is

00:08:38.840 --> 00:08:42.280
remove the 12 hour narrowing criteria,

00:08:42.280 --> 00:08:47.680
and we can see here I got sessions ranging back

00:08:47.680 --> 00:08:51.040
even to 28th of October.

00:08:51.040 --> 00:08:55.600
Since these are just to be considered text,

00:08:55.600 --> 00:08:58.654
I can--

00:08:58.754 --> 00:09:00.320
now let's first narrow the sessions

00:09:00.320 --> 00:09:05.188
to only show the ones that run guix pull.

00:09:05.288 --> 00:09:08.440
Then I would narrow based on the output

00:09:08.440 --> 00:09:11.280
containing a regular expression.

00:09:11.280 --> 00:09:14.400
So, I remember it was something with collection.

00:09:14.400 --> 00:09:19.988
And we got one hit.

00:09:20.088 --> 00:09:24.088
Here it should be something with collection.

00:09:24.188 --> 00:09:29.521
It was emacs-flymake-collection.

00:09:29.621 --> 00:09:34.721
This is something that is a

00:09:34.821 --> 00:09:38.400
very nice feature to have.

00:09:38.400 --> 00:09:42.160
It doesn't create any overhead

00:09:42.160 --> 00:09:45.280
of having these old sessions lying around

00:09:45.280 --> 00:09:47.400
and occasionally, it can be interesting

00:09:47.400 --> 00:09:49.680
to search through them as well.

00:09:49.680 --> 00:09:55.000
So, for me, this is another example of

00:09:55.000 --> 00:09:59.954
when bringing workflows into Emacs,

00:10:00.054 --> 00:10:05.021
it often opens up new exciting possibilities.

00:10:05.121 --> 00:10:10.760
Thanks a lot for listening.