summaryrefslogblamecommitdiffstats
path: root/2023/captions/emacsconf-2023-test--what-i-learned-by-writing-test-cases-for-gnu-hyperbole--mats-lidell--answers.vtt
blob: bc1bfff5ea8506272c88f14bfe4646613c99ea4b (plain) (tree)
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
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406





























































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































                                                          
WEBVTT


00:00:08.740 --> 00:00:09.240
[Speaker 0]: Do we have any listeners?

00:00:13.340 --> 00:00:13.840
It's you and I. I have a question.

00:00:16.420 --> 00:00:16.640
How many tests do you have for hyperbole and

00:00:18.800 --> 00:00:19.279
How would you rate the test coverage compared

00:00:21.279 --> 00:00:21.500
to other packages? Well,

00:00:28.279 --> 00:00:28.700
that's a tricky 1. Shall I spell it out loud

00:00:31.100 --> 00:00:31.600
and then maybe type it at the same time?

00:00:36.420 --> 00:00:36.920
So, I believe it's around like more than 300

00:00:43.660 --> 00:00:44.059
test cases now. But I cannot compare the test

00:00:45.220 --> 00:00:45.720
coverage to any other

00:01:00.020 --> 00:01:00.520
other package. Maybe I can type that later.

00:01:01.560 --> 00:01:02.060
What do you say, Badal?

00:01:02.660 --> 00:01:02.900
[Speaker 1]: package. I have no knowledge of any Yeah,

00:01:03.840 --> 00:01:04.239
sure, yeah, that's totally fine.

00:01:05.660 --> 00:01:06.160
Feel free to just answer them with voice.

00:01:08.720 --> 00:01:09.220
[Speaker 0]: Yeah, yeah. There's another question.

00:01:10.520 --> 00:01:10.920
1 small suggestion to me,

00:01:11.440 --> 00:01:11.940
should means optional,

00:01:13.660 --> 00:01:14.160
where shall or must means required.

00:01:15.940 --> 00:01:16.220
Not sure if it is too late to make a major

00:01:17.220 --> 00:01:17.540
grammar change like that.

00:01:18.080 --> 00:01:18.580
Very nice presentation.

00:01:19.840 --> 00:01:20.340
So thanks for presentation,

00:01:24.380 --> 00:01:24.780
but the package ERT, well,

00:01:27.920 --> 00:01:28.080
it's not something that we have come up with.

00:01:28.920 --> 00:01:29.340
It's a standard package.

00:01:32.320 --> 00:01:32.560
So I believe it has been around for a long

00:01:37.760 --> 00:01:38.000
time. So, but please feel free to make

00:01:39.680 --> 00:01:40.180
suggestions and maybe you can,

00:01:43.660 --> 00:01:43.860
you know, like do a copy or like an alias for

00:01:46.080 --> 00:01:46.200
that. If you believe it makes more sense for

00:01:48.080 --> 00:01:48.580
your test cases to have that instead.

00:01:53.540 --> 00:01:53.720
And then we have another question here.

00:01:55.540 --> 00:01:55.680
For your info, you may find this helpful for

00:01:58.780 --> 00:01:59.020
running MX test lint both from a command line

00:02:01.220 --> 00:02:01.720
and from within MX with a transit menu.

00:02:03.600 --> 00:02:04.040
GitHub alpha papa make sure,

00:02:06.760 --> 00:02:07.100
yes. It also works on remote CI.

00:02:08.240 --> 00:02:08.740
Yeah, thank you, Alpha Papa.

00:02:10.580 --> 00:02:11.080
I think I've looked into that,

00:02:13.440 --> 00:02:13.940
but we haven't made any use of that.

00:02:17.920 --> 00:02:18.080
But maybe you'll inspire me to give it

00:02:18.400 --> 00:02:18.900
another look.

00:02:29.260 --> 00:02:29.760
[Speaker 2]: Hey guys.

00:02:34.120 --> 00:02:34.460
[Speaker 0]: I remember, I recognize that voice.

00:02:37.160 --> 00:02:37.660
Hi, Bob. Hey, how are you?

00:02:40.240 --> 00:02:40.580
Congratulations, man. Thanks,

00:02:43.020 --> 00:02:43.320
Hugh. Thank you. I have another question

00:02:45.520 --> 00:02:45.900
here. It is easy to run ad hoc tests inside

00:02:48.400 --> 00:02:48.600
an Emacs session given the command line

00:02:51.180 --> 00:02:51.560
scripts you need to run to get the batch test

00:02:54.960 --> 00:02:55.120
session running? You said it's to run an

00:03:05.680 --> 00:03:05.920
ad-hoc test. I'm not sure I understand that

00:03:14.440 --> 00:03:14.940
question. Yes, please.

00:03:15.660 --> 00:03:16.160
[Speaker 1]: Maybe I can rephrase. Sure.

00:03:19.900 --> 00:03:20.400
So I think what I understand is that since

00:03:22.540 --> 00:03:23.040
you have to use some of these command lines

00:03:25.440 --> 00:03:25.940
scripts to get a batch test session running,

00:03:28.780 --> 00:03:29.180
is it easy to run ad hoc tests in an Emacs

00:03:30.700 --> 00:03:30.900
session or does that, like in your

00:03:32.040 --> 00:03:32.540
experience, has that been difficult?

00:03:36.820 --> 00:03:37.320
[Speaker 0]: Well, from the command line,

00:03:38.660 --> 00:03:38.940
if you look at the command line,

00:03:44.160 --> 00:03:44.340
you'll see that it's only like a few image

00:03:46.480 --> 00:03:46.980
functions to call to get that behavior to run

00:03:55.080 --> 00:03:55.240
the batch tests. So I think we made some

00:03:57.100 --> 00:03:57.600
support function for that in hyperbole.

00:04:02.800 --> 00:04:02.960
So it's not, I don't think it's possible out

00:04:05.540 --> 00:04:06.040
of the box to do it, but it's not complicated

00:04:08.060 --> 00:04:08.560
to do it.

00:04:12.190 --> 00:04:12.340
[Speaker 2]: You can define a test anytime,

00:04:14.780 --> 00:04:15.280
right? Just like a new function.

00:04:18.899 --> 00:04:19.240
So that's ad hoc. You just write your test

00:04:20.019 --> 00:04:20.519
and you can run it.

00:04:22.900 --> 00:04:23.400
[Speaker 0]: Yeah, yeah, I mean, of course,

00:04:25.900 --> 00:04:26.180
but I got the impression it was about running

00:04:28.620 --> 00:04:29.060
all your tests like we did with the command

00:04:35.740 --> 00:04:36.020
line. Well, so the question is more about how

00:04:38.260 --> 00:04:38.600
would you run all your test cases from within

00:04:44.860 --> 00:04:45.140
Emacs? And the easy answer to that is

00:04:48.420 --> 00:04:48.860
actually you load all your test case files,

00:04:51.760 --> 00:04:52.080
and then you run ERT with the T as the test

00:04:53.600 --> 00:04:53.880
selector and then it will run all your test

00:04:53.880 --> 00:04:54.380
cases.

00:05:01.780 --> 00:05:01.960
[Speaker 1]: Right. And I think they have expanded on

00:05:03.180 --> 00:05:03.520
their question a little bit as well,

00:05:04.960 --> 00:05:05.220
clarifying that. In other words,

00:05:07.200 --> 00:05:07.360
can you tweak tests in an Emacs session and

00:05:08.860 --> 00:05:09.360
run them right away? Which I believe,

00:05:11.400 --> 00:05:11.640
if I understand correctly what Bob was

00:05:13.820 --> 00:05:14.320
saying, you can basically define or redefine

00:05:15.920 --> 00:05:16.080
functions on the fly and then have them be

00:05:16.440 --> 00:05:16.940
run, right?

00:05:22.200 --> 00:05:22.360
[Speaker 0]: Yes, yes. You just go into that test case and

00:05:24.120 --> 00:05:24.620
you just change it and you run it again.

00:05:29.060 --> 00:05:29.200
And either you have to sort of load it or you

00:05:31.560 --> 00:05:32.060
can use like the commercial thing I did.

00:05:36.140 --> 00:05:36.340
You use hyperbole and just hit meta return on

00:05:38.560 --> 00:05:38.860
the test case and it will load it and run the

00:05:42.240 --> 00:05:42.360
test case again. So that's of course what you

00:05:44.220 --> 00:05:44.720
normally do when you're defining a test or

00:05:47.440 --> 00:05:47.940
debug a test case or develop a test case.

00:05:49.960 --> 00:05:50.460
Just start with something small,

00:05:52.700 --> 00:05:53.200
just make sure maybe you can prepare the test

00:05:55.320 --> 00:05:55.680
properly and run it again and again and again

00:05:56.720 --> 00:05:57.220
until you're ready with it.

00:05:59.760 --> 00:05:59.960
That's a good point. You can definitely do

00:06:02.800 --> 00:06:03.280
that and that's part of how I normally

00:06:06.420 --> 00:06:06.920
develop the test cases that I mean start with

00:06:09.160 --> 00:06:09.400
something small so I can see that I get there

00:06:12.180 --> 00:06:12.240
maybe the right input in the buffer that I

00:06:14.180 --> 00:06:14.340
want to test on or something and I expand on

00:06:18.160 --> 00:06:18.400
that more and more and add more and more more

00:06:18.460 --> 00:06:18.960
and more more

00:06:31.040 --> 00:06:31.540
[Speaker 2]: tests to it. You might tell them a bit about

00:06:33.280 --> 00:06:33.480
how many test cases you have.

00:06:36.020 --> 00:06:36.440
I guess you commented on that and like what

00:06:40.320 --> 00:06:40.820
happens, you know, with the CICD pipeline,

00:06:43.780 --> 00:06:44.020
every time we commit, you know,

00:06:46.360 --> 00:06:46.560
across all the versions and what you have set

00:06:48.760 --> 00:06:49.040
up there because you know I wish people could

00:06:53.940 --> 00:06:54.140
see it. You can go and check on GitHub and

00:06:57.440 --> 00:06:57.720
you can see the logs right of any of the

00:06:59.760 --> 00:06:59.960
builds and but tell them a bit about that

00:07:01.080 --> 00:07:01.320
Mats because I think that's pretty

00:07:01.320 --> 00:07:01.820
impressive.

00:07:07.280 --> 00:07:07.760
[Speaker 0]: Well, that's part of more the CI,

00:07:11.760 --> 00:07:12.160
CD, part of how we developed this using

00:07:15.460 --> 00:07:15.580
GitHub and workflows that you get out of the

00:07:20.740 --> 00:07:20.900
box from there. So this more than 300 test

00:07:23.440 --> 00:07:23.720
cases on our round for I think 5 different

00:07:26.480 --> 00:07:26.980
versions of Emacs when we do a pull request

00:07:33.900 --> 00:07:34.400
or a commit. So that's a good way to ensure

00:07:38.040 --> 00:07:38.540
that it works from version 27.2

00:07:42.240 --> 00:07:42.740
up to the latest master version because

00:07:45.860 --> 00:07:46.360
there's some changes in Emacs over different

00:07:48.940 --> 00:07:49.340
versions that can affect your functions or

00:07:49.600 --> 00:07:50.100
your code.

00:07:56.580 --> 00:07:56.720
[Speaker 2]: They all run in parallel and so typically in

00:08:00.580 --> 00:08:00.780
under 60 seconds I think you've got all of

00:08:03.960 --> 00:08:04.460
them run so you've got pretty extensive

00:08:08.860 --> 00:08:09.240
testing which does catch interesting bugs

00:08:09.760 --> 00:08:10.260
here and there, right?

00:08:13.320 --> 00:08:13.820
[Speaker 0]: Yes, of course it does.

00:08:18.060 --> 00:08:18.560
I mean, you normally develop with 1 version

00:08:20.280 --> 00:08:20.540
and then you think everything is okay.

00:08:21.720 --> 00:08:21.820
But then when you're tested with the

00:08:23.460 --> 00:08:23.960
different versions, you find out that there

00:08:26.080 --> 00:08:26.580
are some changes and there are things you

00:08:30.140 --> 00:08:30.400
might not sort of keep track of what's

00:08:34.340 --> 00:08:34.640
happening also. So that's a way to get

00:08:38.559 --> 00:08:38.940
noticed that the core developers of Emacs

00:08:41.120 --> 00:08:41.480
have changed something that you sort of based

00:08:44.380 --> 00:08:44.840
your code on. Now I got another question

00:08:47.900 --> 00:08:48.160
here. Did you have to change hyperbole code

00:08:50.580 --> 00:08:50.760
and design to be more readily testable as you

00:08:52.160 --> 00:08:52.660
were increasing your test coverage?

00:08:55.520 --> 00:08:56.020
Well, we haven't done that to a lot,

00:09:00.160 --> 00:09:00.320
to a big degree, although I believe that that

00:09:03.760 --> 00:09:04.260
is an important thing for sort of the future

00:09:06.020 --> 00:09:06.500
to do that because some of the hyperbolic

00:09:08.520 --> 00:09:08.720
functions are very complicated and long and

00:09:10.640 --> 00:09:11.140
that makes testing them rather difficult.

00:09:14.660 --> 00:09:14.900
So, at a few places we have sort of broken up

00:09:17.260 --> 00:09:17.720
functions in smaller pieces so it'd be easier

00:09:20.280 --> 00:09:20.660
to do like unit tests of the different parts

00:09:27.740 --> 00:09:27.980
of it. But there's a lot of more work that

00:09:28.680 --> 00:09:29.180
has to be done there.

00:09:33.820 --> 00:09:34.020
[Speaker 2]: 1 of the nice things is you know the great

00:09:36.760 --> 00:09:36.820
environment in Lisp where we're able to do a

00:09:40.520 --> 00:09:40.900
lot of interactive bottom-up testing before

00:09:42.840 --> 00:09:43.280
we even get to lighting tech pieces.

00:09:48.740 --> 00:09:49.140
So it does tend to be more higher level bugs,

00:09:51.140 --> 00:09:51.640
I think, that get caught in cross-functional

00:09:55.940 --> 00:09:56.100
interaction. We had 1 recently that was an

00:09:58.100 --> 00:09:58.600
Emacs version change. It had been a function

00:10:01.100 --> 00:10:01.600
that had existed for a long time.

00:10:03.340 --> 00:10:03.840
It had an and rest in it,

00:10:05.740 --> 00:10:06.240
in its argument list, so it would assemble

00:10:08.600 --> 00:10:09.100
the list of arguments from individual

00:10:10.320 --> 00:10:10.820
arguments that you would give it,

00:10:13.100 --> 00:10:13.600
and they decided in a recent version,

00:10:15.200 --> 00:10:15.700
I think with Stefan's input,

00:10:19.400 --> 00:10:19.840
to change that to a list and allow the prior

00:10:22.740 --> 00:10:22.900
behavior, but it would issue a warning if you

00:10:23.620 --> 00:10:24.060
use the prior behavior.

00:10:25.560 --> 00:10:25.840
So all of a sudden, the way you were supposed

00:10:27.180 --> 00:10:27.680
to do it became semi-invalid.

00:10:30.440 --> 00:10:30.940
And so we started getting the warning,

00:10:32.760 --> 00:10:33.040
and we've tried to eliminate all those

00:10:35.600 --> 00:10:36.060
warnings in recent hyperbole developments.

00:10:37.120 --> 00:10:37.620
So we're like, what do we do?

00:10:39.020 --> 00:10:39.440
You know, because we wanted to be backward

00:10:42.140 --> 00:10:42.640
compatible to where you couldn't use a list.

00:10:44.620 --> 00:10:45.120
It required you to use individual arguments.

00:10:48.380 --> 00:10:48.560
And now it's sort of requiring you to do

00:10:51.660 --> 00:10:51.820
that. And all of that was caused by the

00:10:52.940 --> 00:10:53.440
automatic testing on it.

00:11:08.680 --> 00:11:08.860
So you said, Max, you were going to tell us

00:11:12.740 --> 00:11:13.220
what you learned. So what are the major

00:11:15.368 --> 00:11:15.396
things that you learned in doing all of this

00:11:15.680 --> 00:11:16.180
work? All of this work?

00:11:26.520 --> 00:11:26.740
[Speaker 0]: Well, I tried to cover some of it in the

00:11:29.380 --> 00:11:29.800
presentation, but as I was going along,

00:11:33.420 --> 00:11:33.920
the presentation became like twice as long as

00:11:36.180 --> 00:11:36.680
fitted into the time we had so I had to cut

00:11:42.380 --> 00:11:42.880
it out. But I think some of the core things

00:11:44.340 --> 00:11:44.840
still is in the presentation.

00:11:49.560 --> 00:11:50.060
From a personal perspective,

00:11:52.440 --> 00:11:52.940
And this might not be hard to realize,

00:11:56.960 --> 00:11:57.460
but forcing yourself to test functions,

00:12:02.900 --> 00:12:03.060
test code really forces you to understand the

00:12:05.080 --> 00:12:05.280
code a little bit better in a way that sort

00:12:07.300 --> 00:12:07.400
of makes it easier than just to read the

00:12:11.460 --> 00:12:11.960
code. I don't know how it is for the rest

00:12:13.780 --> 00:12:13.980
listening to this, but for me it works so

00:12:16.580 --> 00:12:17.080
that if I just read the code then I don't

00:12:20.140 --> 00:12:20.320
sort of become as sharp as I should be but if

00:12:22.500 --> 00:12:22.640
I try to write the test case for it then I

00:12:24.680 --> 00:12:24.880
really need to understand better of all the

00:12:27.660 --> 00:12:28.160
edge cases and all the sort of states and etc

00:12:30.060 --> 00:12:30.320
that is involved and I think that's That's

00:12:33.080 --> 00:12:33.200
what's sort of 1 of the learning things I

00:12:34.960 --> 00:12:35.280
wanted to communicate as well that I don't

00:12:38.940 --> 00:12:39.080
think I covered in detail in the

00:12:41.480 --> 00:12:41.980
presentation. Maybe all this,

00:12:48.060 --> 00:12:48.340
but try it. 1 other sort of more from the fun

00:12:50.740 --> 00:12:51.000
side is that I really think it's fun to write

00:12:55.080 --> 00:12:55.440
the test. So if you haven't tests in your

00:12:58.020 --> 00:12:58.520
package, you should start doing that because

00:13:05.740 --> 00:13:06.080
it is fun. It might feel like some extra

00:13:08.080 --> 00:13:08.580
work, but it really pays off in the long run,

00:13:10.320 --> 00:13:10.760
especially if you have it in like a pipeline

00:13:12.520 --> 00:13:12.980
and where you can run it regularly when you

00:13:13.940 --> 00:13:14.380
do new commits, et cetera.

00:13:16.560 --> 00:13:17.060
So, I mean, that's maybe obvious from,

00:13:19.160 --> 00:13:19.440
if you look from the commercial side or your

00:13:21.080 --> 00:13:21.340
work side to do it like that.

00:13:22.260 --> 00:13:22.660
But even for your hobby project,

00:13:26.260 --> 00:13:26.760
it can be very sort of pay off really well.

00:13:32.900 --> 00:13:33.160
[Speaker 2]: It's worked really well when we're adding new

00:13:35.020 --> 00:13:35.180
functionality or we're changing some of the

00:13:36.560 --> 00:13:37.060
plumbing in the system.

00:13:40.400 --> 00:13:40.580
You know, you go and you do some surgery and

00:13:41.320 --> 00:13:41.820
then you run the tests.

00:13:45.400 --> 00:13:45.900
And sometimes 6 to 10 tests will fail.

00:13:48.260 --> 00:13:48.420
And you find there, you know,

00:13:50.460 --> 00:13:50.660
it tends to be they're all interconnected and

00:13:52.920 --> 00:13:53.320
it leads you back to the single source.

00:13:56.660 --> 00:13:56.980
You fix that and you know it could be an edge

00:14:00.560 --> 00:14:00.760
case and off by 1 or Sometimes it's an

00:14:03.520 --> 00:14:03.800
assumption about the way something is used

00:14:05.980 --> 00:14:06.480
and it's not actually always true.

00:14:09.520 --> 00:14:10.020
And so, Matt's just really good at

00:14:13.540 --> 00:14:14.040
identifying some of those scenarios and

00:14:17.480 --> 00:14:17.980
keeping us honest, I guess I would say.

00:14:22.900 --> 00:14:23.400
So I love, I run it as much as I before,

00:14:26.400 --> 00:14:26.900
you know, even before I commit something.

00:14:29.960 --> 00:14:30.060
So I get to see, you know,

00:14:30.940 --> 00:14:31.440
if anything has progressed.

00:14:39.480 --> 00:14:39.920
So yeah, I really recommend this process to

00:14:42.120 --> 00:14:42.620
people. I haven't seen it done.

00:14:45.720 --> 00:14:46.080
I don't think that, I don't know any other

00:14:47.800 --> 00:14:48.300
package that has done it to this level.

00:14:51.560 --> 00:14:51.820
And it's been working really great for us.

00:14:55.440 --> 00:14:55.640
And I think, well, we'll see too when we

00:14:56.780 --> 00:14:57.280
release to the general public.

00:15:04.380 --> 00:15:04.540
[Speaker 0]: But Bob, also, maybe the test part of

00:15:06.400 --> 00:15:06.560
different packages is not the first thing you

00:15:08.900 --> 00:15:09.100
look at. So I know there are packages that

00:15:10.960 --> 00:15:11.380
have testing, a lot of testing,

00:15:13.860 --> 00:15:14.160
but how much, much testing they have or not,

00:15:16.060 --> 00:15:16.220
I don't know. It's not what you normally look

00:15:17.900 --> 00:15:18.400
into when you look at someone's else code.

00:15:20.600 --> 00:15:20.820
You look maybe on the functionality side but

00:15:22.760 --> 00:15:23.000
not on how they've done the sort of the

00:15:26.540 --> 00:15:26.760
quality side. So there could be other

00:15:28.780 --> 00:15:29.280
packages out there that are well equipped.

00:15:31.800 --> 00:15:32.300
[Speaker 2]: I hope so. I hope so.

00:15:39.860 --> 00:15:40.180
[Speaker 0]: What's the craziest bug you found when

00:15:44.700 --> 00:15:45.200
writing these tests? Well,

00:15:50.760 --> 00:15:50.940
What springs to my mind just now is that we

00:15:52.760 --> 00:15:52.960
were doing some tests or I would do some

00:15:55.920 --> 00:15:56.420
tests for when you narrow,

00:15:57.940 --> 00:15:58.440
what do you say that? When you,

00:16:04.500 --> 00:16:05.000
in outlining, when you sort of compress

00:16:06.480 --> 00:16:06.980
things in an outline, so you just,

00:16:08.540 --> 00:16:09.040
sorry Bob, maybe you have it,

00:16:12.100 --> 00:16:12.600
[Speaker 2]: When you hide text.

00:16:12.740 --> 00:16:13.240
[Speaker 0]: What I'm looking for? Yeah,

00:16:15.580 --> 00:16:15.920
when you hide. So I was doing some cursor

00:16:17.780 --> 00:16:17.980
movement over that. And I always assume that

00:16:22.540 --> 00:16:22.900
if you do like a prefix argument to like a

00:16:23.800 --> 00:16:24.240
simple cursor movement,

00:16:26.420 --> 00:16:26.920
like control F moving 1 character position,

00:16:28.340 --> 00:16:28.840
and you would give it the,

00:16:36.580 --> 00:16:37.080
and then the prefix, like you want to move

00:16:39.140 --> 00:16:39.640
like 2 or 3 positions,

00:16:43.040 --> 00:16:43.140
you would do like control U 3 and then

00:16:44.240 --> 00:16:44.740
control F and you move 3.

00:16:46.560 --> 00:16:46.960
I always assumed that that would be exactly

00:16:49.240 --> 00:16:49.440
the same as if you just hit the key control F

00:16:50.740 --> 00:16:51.240
3 times, but it's not.

00:16:53.160 --> 00:16:53.560
So it's not the bug, it's a feature,

00:16:54.620 --> 00:16:55.080
but that was the craziest thing.

00:16:58.180 --> 00:16:58.360
I spent the night trying to figure out why

00:17:00.720 --> 00:17:01.000
our code was wrong, but It turns out that's

00:17:03.560 --> 00:17:04.060
how Emacs behaves. Try it out yourself.

00:17:07.920 --> 00:17:08.300
Try to move over the 3 dots at the end of

00:17:09.140 --> 00:17:09.640
that and see what happens.

00:17:14.060 --> 00:17:14.240
Do it with cursor hitting the key or using a

00:17:16.260 --> 00:17:16.680
prefix argument and you see it behaves

00:17:18.720 --> 00:17:19.220
differently. That was the craziest thing.

00:17:21.960 --> 00:17:22.339
I think there was some other crazy thing or

00:17:24.280 --> 00:17:24.480
deep learning also, but I can't come up with

00:17:26.599 --> 00:17:26.760
it at the moment. So maybe I can write it in

00:17:27.900 --> 00:17:28.400
the Q&A later.

00:17:31.200 --> 00:17:31.440
[Speaker 1]: I think we're out of time on the stream,

00:17:33.360 --> 00:17:33.600
but people are welcome to join Mats and Bob

00:17:35.280 --> 00:17:35.640
here on BigBlueButton to further discuss

00:17:36.480 --> 00:17:36.980
this. Thank you both.

00:17:38.674 --> 00:17:38.792
[Speaker 0]: Okay, thank you. Thanks,

00:17:46.100 --> 00:17:46.600
Makaay. Thank you. I don't know,

00:17:48.740 --> 00:17:49.240
Is it only me and Bob here?

00:17:50.680 --> 00:17:51.180
So Bob, do you want to say something?

00:17:57.440 --> 00:17:57.940
[Speaker 2]: Well, I think it's been a great day.

00:18:00.720 --> 00:18:01.220
And I'm glad we did this.

00:18:02.280 --> 00:18:02.780
It takes a lot of energy.

00:18:15.140 --> 00:18:15.640
I'm just really excited about the progress

00:18:20.580 --> 00:18:20.740
that this, and we're actually doing a lot of

00:18:23.940 --> 00:18:24.160
QA at work and my professional software work

00:18:28.500 --> 00:18:28.840
and looking at you know how we can do more

00:18:32.980 --> 00:18:33.480
test driven development and so everybody's

00:18:35.980 --> 00:18:36.200
talking about this you know we've got AI over

00:18:37.540 --> 00:18:38.040
here that can generate test cases.

00:18:40.200 --> 00:18:40.700
But, you know, strangely enough,

00:18:43.100 --> 00:18:43.380
with the rapidity of development and web

00:18:46.720 --> 00:18:47.220
applications, I think the level of testing

00:18:50.140 --> 00:18:50.280
has gone down in recent years compared to

00:18:51.500 --> 00:18:51.780
where it used to be, right?

00:18:53.040 --> 00:18:53.540
Because the pace has gone up.

00:18:57.340 --> 00:18:57.840
And so I think it's starting to turn again

00:18:58.740 --> 00:18:59.240
where people are saying,

00:19:01.940 --> 00:19:02.440
we can't just release crap into the

00:19:08.120 --> 00:19:08.620
Webisphere and we have to better ourselves.

00:19:13.620 --> 00:19:13.820
And with all these advanced tool sets that

00:19:16.100 --> 00:19:16.600
you have, that you can do CICD testing,

00:19:19.860 --> 00:19:20.180
you know, I just, I just see it coming

00:19:21.900 --> 00:19:22.100
around, you know, as people develop new

00:19:24.000 --> 00:19:24.160
things. So That's kind of exciting to me

00:19:26.980 --> 00:19:27.480
because I came from a manufacturing culture

00:19:30.300 --> 00:19:30.780
originally where we, our company actually

00:19:33.800 --> 00:19:34.300
started a lot of the manufacturing quality

00:19:37.420 --> 00:19:37.920
efforts that you saw in Japan and elsewhere

00:19:40.600 --> 00:19:40.740
in America for a long time and that was you

00:19:42.040 --> 00:19:42.540
know entirely through testing.

00:19:46.640 --> 00:19:47.020
We used to just build incredible test cases

00:19:49.120 --> 00:19:49.320
because we were combining software with

00:19:51.100 --> 00:19:51.380
hardware. And if, you know,

00:19:53.460 --> 00:19:53.600
the hardware doesn't work and you ship a

00:19:55.080 --> 00:19:55.520
million units, you're,

00:19:57.340 --> 00:19:57.840
you're in trouble. So,

00:20:00.260 --> 00:20:00.760
that was just something we had to do.

00:20:04.280 --> 00:20:04.780
And so it's nice to start to see that curve

00:20:07.020 --> 00:20:07.520
come around. And I think,

00:20:10.380 --> 00:20:10.880
you know, Matt Vance is very modest,

00:20:16.680 --> 00:20:16.920
but I think he's really the 1 that started us

00:20:20.400 --> 00:20:20.580
down this path and really made it into a

00:20:24.620 --> 00:20:24.840
reality. So everybody else just gets to

00:20:25.760 --> 00:20:26.260
benefit from that work.

00:20:27.540 --> 00:20:28.040
So thanks.

00:20:32.760 --> 00:20:33.260
[Speaker 1]: That's awesome.

00:20:39.960 --> 00:20:40.460
[Speaker 0]: Thanks. Okay. Yeah. So if there's nothing

00:20:43.200 --> 00:20:43.520
more here, then maybe we should just close

00:20:45.440 --> 00:20:45.940
this and I go over to write in the etherpad

00:20:47.960 --> 00:20:48.460
the replies we had.

00:20:51.900 --> 00:20:52.120
[Speaker 1]: Right, yeah, I think, let's see,

00:20:53.520 --> 00:20:53.760
I see 1 other person here,

00:20:55.080 --> 00:20:55.580
I believe Ihor just joined us.

00:20:58.780 --> 00:20:59.060
Yeah. Yeah, so if you do want to discuss with

00:21:00.220 --> 00:21:00.480
Mats and Bob, you're welcome to,

00:21:02.200 --> 00:21:02.700
otherwise, yeah, we can close the room now.

00:21:05.800 --> 00:21:06.020
[Speaker 3]: Well, I think I missed most of the talk

00:21:06.900 --> 00:21:07.400
because I had power outage,

00:21:12.180 --> 00:21:12.440
but the part I heard was about the mock

00:21:16.860 --> 00:21:17.220
library. And you mentioned that you don't

00:21:20.200 --> 00:21:20.700
like CL-let, but instead you use mock.

00:21:29.700 --> 00:21:29.800
[Speaker 0]: Yeah, I was more saying that you have to do a

00:21:31.560 --> 00:21:32.040
lot more work when you use the CL letdef.

00:21:34.540 --> 00:21:34.780
It's for more ambitious and maybe more

00:21:37.000 --> 00:21:37.500
complicated cases where you want to really

00:21:38.840 --> 00:21:39.340
make a new implementation,

00:21:41.940 --> 00:21:42.440
test implementation. If you use the mock,

00:21:44.380 --> 00:21:44.880
you get a lot of things out of the box,

00:21:47.440 --> 00:21:47.940
verifying that you actually,

00:21:50.820 --> 00:21:51.040
like the mock was actually called for

00:21:53.320 --> 00:21:53.820
instance, whereas if you do with the CLLatf,

00:21:56.520 --> 00:21:56.780
you would have to take correct track of that

00:22:02.020 --> 00:22:02.520
yourself. And so, so a lot of more work.

00:22:03.760 --> 00:22:04.260
Oh yeah.

00:22:07.940 --> 00:22:08.200
[Speaker 3]: I'm saying that most of the time CLLess is

00:22:09.720 --> 00:22:10.220
used for simple cases actually.

00:22:12.320 --> 00:22:12.820
Because, just for example,

00:22:15.100 --> 00:22:15.600
the function always returns the same.

00:22:17.980 --> 00:22:18.420
And it tends to be simple lambda that ignores

00:22:19.040 --> 00:22:19.540
all the input arguments.

00:22:23.000 --> 00:22:23.480
So that's really trivial most of the time but

00:22:25.520 --> 00:22:25.920
I actually thought the opposite that mock is

00:22:27.640 --> 00:22:28.140
supposed to be used for non-trivial cases.

00:22:32.280 --> 00:22:32.520
[Speaker 0]: Sorry, what was the question?

00:22:35.280 --> 00:22:35.780
Mock was supposed to be used for non-trivial.

00:22:47.680 --> 00:22:48.180
Yeah I mean I don't know how to explain this.

00:22:50.140 --> 00:22:50.640
I mean, CLF can be used for non-trivial

00:22:54.400 --> 00:22:54.840
definitely. You can define then any behavior

00:22:56.180 --> 00:22:56.680
you want. You can write your own function,

00:22:58.440 --> 00:22:58.660
but you need to keep track of whether that

00:22:59.620 --> 00:23:00.100
function is called or not,

00:23:06.260 --> 00:23:06.380
for instance. So you have to make note of

00:23:08.440 --> 00:23:08.940
that the function was called so you can fire

00:23:12.440 --> 00:23:12.800
sort of an error in case your function wasn't

00:23:16.960 --> 00:23:17.440
called because that would be 1 error case.

00:23:20.660 --> 00:23:20.860
[Speaker 3]: So you mean the mock fires an error if the

00:23:22.580 --> 00:23:23.080
mocked function was actually not called?

00:23:30.060 --> 00:23:30.560
[Speaker 0]: Yes, it does. Yes. So if your assumptions,

00:23:33.900 --> 00:23:34.120
you sort of document with the mock also your

00:23:37.080 --> 00:23:37.220
assumptions how your code is going to be

00:23:40.020 --> 00:23:40.380
called. And if those are wrong,

00:23:41.120 --> 00:23:41.540
you will get an error.

00:23:43.680 --> 00:23:44.060
So you would, so if the implementation would

00:23:44.840 --> 00:23:45.100
maybe change, for instance,

00:23:46.640 --> 00:23:47.140
and not call the thing you're mocking,

00:23:50.460 --> 00:23:50.960
then you will notice that.

00:23:53.100 --> 00:23:53.560
But if you see a letdef,

00:23:54.840 --> 00:23:55.040
then you will have to keep track of that

00:23:57.560 --> 00:23:58.060
yourself. Okay, I see.

00:23:58.260 --> 00:23:58.760
I see.

00:24:01.240 --> 00:24:01.740
[Speaker 3]: And you know, our mode also uses a lot of

00:24:09.340 --> 00:24:09.620
test. In our mode, we have a lot of tests

00:24:13.940 --> 00:24:14.440
[Speaker 0]: Ah, okay. Yeah. Yeah. I'm sure I have.

00:24:15.900 --> 00:24:16.400
[Speaker 3]: also. We rely on CLLatF for,

00:24:19.220 --> 00:24:19.720
we don't use third-party libraries at all.

00:24:22.140 --> 00:24:22.640
[Speaker 0]: Oh, you use CLLatF, okay.

00:24:26.680 --> 00:24:27.180
Yeah. Yeah. Yeah. At First I found it very

00:24:29.480 --> 00:24:29.700
powerful to use that, but then I sort of,

00:24:32.120 --> 00:24:32.320
I learned more about how we can use the

00:24:34.340 --> 00:24:34.840
mocking library for what I needed.

00:24:36.900 --> 00:24:37.400
And I prefer that at the moment.

00:24:40.560 --> 00:24:41.060
[Speaker 3]: I see, that is interesting.

00:24:42.500 --> 00:24:42.700
Because I had seen it,

00:24:45.440 --> 00:24:45.600
but I didn't consider that it's gonna be

00:24:46.800 --> 00:24:47.300
useful even in simple cases.

00:24:52.640 --> 00:24:53.140
[Speaker 0]: It has its limitations.

00:24:58.260 --> 00:24:58.760
So it's like life, how you turn depends.

00:25:03.740 --> 00:25:04.020
But maybe I should look more into the org

00:25:05.880 --> 00:25:06.100
mode and the test case to learn more about

00:25:07.480 --> 00:25:07.980
that. So thanks for pointing that out.

00:25:14.620 --> 00:25:15.120
[Speaker 3]: We are trying to cover as much as we can.

00:25:17.520 --> 00:25:17.740
It's almost impossible for org.

00:25:20.500 --> 00:25:21.000
But yeah, we keep adding more tests.

00:25:22.780 --> 00:25:23.280
[Speaker 0]: That's great.

00:25:52.720 --> 00:25:53.200
Someone's typing. I don't know.

00:25:54.340 --> 00:25:54.840
Any more questions? No?

00:26:01.060 --> 00:26:01.560
Okay, then I'll go back and try to document

00:26:05.200 --> 00:26:05.360
this in the etherpad. Thank you everybody for

00:26:08.860 --> 00:26:09.160
[Speaker 1]: Thank you guys. Great work.

00:26:09.400 --> 00:26:09.900
[Speaker 0]: joining. Great. Thank you.

00:26:11.100 --> 00:26:11.600
Take care. Bye-bye.

00:26:15.060 --> 00:26:15.560
[Speaker 1]: Take care. Bye. Silence.