summaryrefslogblamecommitdiffstats
path: root/2020/subtitles/emacsconf-2020--23-incremental-parsing-with-emacs-tree-sitter--tuan-anh-nguyen-autogen.sbv
blob: 8761622385ce849b273a590ecb42f35a8f89f0fe (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
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
















































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































                                        
0:00:01.520,0:00:07.200
hello everyone my name is toniang

0:00:04.400,0:00:09.280
i've been using amax for about 10 years

0:00:07.200,0:00:11.519
today i'm going to talk about 360

0:00:09.280,0:00:13.759
a new imax package that allows ems to

0:00:11.519,0:00:17.840
pass multiple programming languages

0:00:13.759,0:00:21.840
in real time

0:00:17.840,0:00:23.359
so what is the problem statement

0:00:21.840,0:00:24.960
in order to support programming

0:00:23.359,0:00:25.760
functionalities for a particular

0:00:24.960,0:00:27.680
language

0:00:25.760,0:00:29.679
a text editor needs to have some degree

0:00:27.680,0:00:31.840
of language understanding

0:00:29.679,0:00:33.840
traditionally text editors have relied

0:00:31.840,0:00:34.960
very heavily on regular expressions for

0:00:33.840,0:00:38.320
this

0:00:34.960,0:00:39.280
e-max is no different most language

0:00:38.320,0:00:40.879
major modes use

0:00:39.280,0:00:42.960
regular expressions for syntax

0:00:40.879,0:00:46.239
highlighting code navigation

0:00:42.960,0:00:47.440
folding indexing and so on regular

0:00:46.239,0:00:50.559
expressions are

0:00:47.440,0:00:53.600
problematic for a couple of reasons

0:00:50.559,0:00:54.000
they're slow and inaccurate they also

0:00:53.600,0:00:56.800
make

0:00:54.000,0:00:57.440
the code hard to read and write

0:00:56.800,0:00:59.199
sometimes

0:00:57.440,0:01:01.199
it's because the regular expressions

0:00:59.199,0:01:04.000
themselves are very hairy

0:01:01.199,0:01:05.199
and sometimes because they are just not

0:01:04.000,0:01:07.840
powerful enough

0:01:05.199,0:01:11.200
some helper code is usually needed to

0:01:07.840,0:01:13.280
pass more intricate language features

0:01:11.200,0:01:16.159
that also illustrates the core problem

0:01:13.280,0:01:18.400
with regular expressions

0:01:16.159,0:01:21.119
in that they are not powerful enough to

0:01:18.400,0:01:22.640
pass programming languages

0:01:21.119,0:01:25.040
an example feature that regular

0:01:22.640,0:01:27.520
expressions cannot handle very well

0:01:25.040,0:01:28.320
is string interpolation which is a very

0:01:27.520,0:01:31.680
common feature

0:01:28.320,0:01:34.079
in many modern programming languages

0:01:31.680,0:01:35.840
it would be much nicer if image somehow

0:01:34.079,0:01:36.479
had structural understanding of source

0:01:35.840,0:01:39.439
code

0:01:36.479,0:01:39.439
like ides do

0:01:39.520,0:01:42.960
there have been multiple efforts to

0:01:41.119,0:01:45.280
bring this kind of programming language

0:01:42.960,0:01:47.119
understanding into emacs

0:01:45.280,0:01:48.640
there are language specific persons

0:01:47.119,0:01:50.240
written in elise

0:01:48.640,0:01:52.320
they can be thought of as the next

0:01:50.240,0:01:54.960
logical step of the glue code on top

0:01:52.320,0:01:56.000
of tribal expressions moving from

0:01:54.960,0:01:58.079
partial local

0:01:56.000,0:01:59.840
pattern recognition into a full-fledged

0:01:58.079,0:02:01.439
parser

0:01:59.840,0:02:03.040
the most prominent example of this

0:02:01.439,0:02:06.159
approach is probably the famous

0:02:03.040,0:02:06.159
js2 mode

0:02:06.479,0:02:12.959
however this approach has several issues

0:02:10.080,0:02:13.680
parsing is computationally expensive and

0:02:12.959,0:02:16.800
imagine

0:02:13.680,0:02:18.400
is not good at that kind of stuff

0:02:16.800,0:02:20.840
furthermore maintenance is very

0:02:18.400,0:02:22.160
troublesome in order to work on these

0:02:20.840,0:02:23.599
process

0:02:22.160,0:02:25.599
first you have to know at least well

0:02:23.599,0:02:27.760
enough and then you have to be

0:02:25.599,0:02:30.319
comfortable with writing a

0:02:27.760,0:02:32.080
recursive ascendant parser while

0:02:30.319,0:02:34.000
constantly keeping up with changes to

0:02:32.080,0:02:36.879
the language itself

0:02:34.000,0:02:39.360
which can be evolving very quickly like

0:02:36.879,0:02:41.599
javascript for example

0:02:39.360,0:02:45.680
together these constraints significantly

0:02:41.599,0:02:47.760
reduce the pull of potential maintenance

0:02:45.680,0:02:49.680
the biggest issue though in my opinion

0:02:47.760,0:02:52.879
is lack of the set of generic

0:02:49.680,0:02:54.319
and reusable apis this makes them very

0:02:52.879,0:02:55.920
hard to use

0:02:54.319,0:02:57.920
for minor modes that want to deal with

0:02:55.920,0:02:59.920
cross-cutting concerns across multiple

0:02:57.920,0:03:01.760
languages

0:02:59.920,0:03:03.599
the other approach which has been

0:03:01.760,0:03:04.319
gaining a lot of momentum in recent

0:03:03.599,0:03:06.560
years

0:03:04.319,0:03:08.159
is externalizing language understanding

0:03:06.560,0:03:12.239
to another process

0:03:08.159,0:03:14.480
also known as language server protocol

0:03:12.239,0:03:16.560
this second approach is actually a very

0:03:14.480,0:03:18.400
interesting one

0:03:16.560,0:03:21.280
my decoupling language understanding

0:03:18.400,0:03:23.760
from the editing facility itself

0:03:21.280,0:03:25.120
the usb servers can attract a lot more

0:03:23.760,0:03:28.959
contributors

0:03:25.120,0:03:32.400
which makes maintenance easier however

0:03:28.959,0:03:34.720
they also have several issues available

0:03:32.400,0:03:36.000
being a separate process they are

0:03:34.720,0:03:39.920
usually more resource

0:03:36.000,0:03:42.159
intensive and depending on the language

0:03:39.920,0:03:44.640
the usb server itself can bring with it

0:03:42.159,0:03:47.680
a host of additional dependencies

0:03:44.640,0:03:50.560
external to emacs which may message to

0:03:47.680,0:03:50.560
install and manage

0:03:50.640,0:03:55.120
furthermore json over rpc has pretty

0:03:53.760,0:03:57.840
high latency

0:03:55.120,0:04:00.879
for one-off tasks like jumping to source

0:03:57.840,0:04:03.040
or on-demand completion is great

0:04:00.879,0:04:06.000
but for things like code highlighting

0:04:03.040,0:04:08.319
the latency is just too much

0:04:06.000,0:04:10.480
i was using rust and i was following the

0:04:08.319,0:04:11.760
community effort to improve its id

0:04:10.480,0:04:13.680
support

0:04:11.760,0:04:15.760
hoping to integrate some of that into

0:04:13.680,0:04:17.600
emacs itself

0:04:15.760,0:04:19.759
then i heard someone from community

0:04:17.600,0:04:23.280
mention tree sitter

0:04:19.759,0:04:23.280
and i decided to check it out

0:04:23.360,0:04:28.720
basically trisita is an incremental

0:04:25.520,0:04:31.000
parsing library and a parser generator

0:04:28.720,0:04:33.040
it was introduced by the item editor in

0:04:31.000,0:04:35.680
2018

0:04:33.040,0:04:36.960
besides item is also being integrated

0:04:35.680,0:04:41.040
into the neo-vim

0:04:36.960,0:04:42.479
editor and github is using it to power

0:04:41.040,0:04:45.600
their source code analysis and

0:04:42.479,0:04:45.600
navigation features

0:04:45.840,0:04:49.199
it is written in c and can be compiled

0:04:48.639,0:04:53.120
for all

0:04:49.199,0:04:56.080
major platforms it can even be compiled

0:04:53.120,0:04:57.600
to web assembly to run on the web that's

0:04:56.080,0:05:00.400
how github is using it

0:04:57.600,0:05:00.400
on their website

0:05:00.800,0:05:05.840
so why is trisita an interesting

0:05:02.960,0:05:07.360
solution to this problem

0:05:05.840,0:05:10.000
there are multiple features that make it

0:05:07.360,0:05:12.400
an attractive option

0:05:10.000,0:05:13.680
it is designed to be fast by being

0:05:12.400,0:05:15.680
incremental

0:05:13.680,0:05:18.160
the initial parts of a typical big fight

0:05:15.680,0:05:20.240
can take tens of milliseconds

0:05:18.160,0:05:22.560
while subsequent incremental processes

0:05:20.240,0:05:24.720
are sub milliseconds

0:05:22.560,0:05:26.240
it achieves this by using structural

0:05:24.720,0:05:29.360
sharing

0:05:26.240,0:05:32.960
meaning replacing only affected nodes

0:05:29.360,0:05:36.000
in the old tree when it needs to

0:05:32.960,0:05:37.120
also unlike lsp being in the same

0:05:36.000,0:05:40.639
process

0:05:37.120,0:05:42.880
it has much lower latency

0:05:40.639,0:05:44.960
secondly it provides a uniform

0:05:42.880,0:05:47.039
programming interface

0:05:44.960,0:05:48.720
the same data structures and functions

0:05:47.039,0:05:50.400
work on parse trees of different

0:05:48.720,0:05:52.160
languages

0:05:50.400,0:05:54.160
syntax knows of different languages

0:05:52.160,0:05:57.360
differ only by their types

0:05:54.160,0:05:58.960
and their possible child nodes this

0:05:57.360,0:06:02.000
is a big advantage over language

0:05:58.960,0:06:02.000
specific parcels

0:06:02.240,0:06:06.880
thirdly it's written in self-contained

0:06:04.880,0:06:09.680
embeddable c

0:06:06.880,0:06:10.400
as i mentioned previously it can even be

0:06:09.680,0:06:13.759
compiled

0:06:10.400,0:06:15.199
to webassembly this makes integrating it

0:06:13.759,0:06:18.240
into various editors

0:06:15.199,0:06:21.840
quite easy without having to install

0:06:18.240,0:06:21.840
any external dependencies

0:06:22.880,0:06:28.000
one thing that is not mentioned here is

0:06:24.639,0:06:31.039
that being a parcel generator

0:06:28.000,0:06:34.880
scrummers are declarative

0:06:31.039,0:06:36.720
together with being editor independent

0:06:34.880,0:06:38.160
this makes the pool of potential

0:06:36.720,0:06:42.400
contributors

0:06:38.160,0:06:45.520
much larger so i was convinced

0:06:42.400,0:06:48.000
that trisito is a good fit for emacs

0:06:45.520,0:06:48.720
last year i started writing the bindings

0:06:48.000,0:06:50.960
using

0:06:48.720,0:06:53.280
dynamic model support introduced in imax

0:06:50.960,0:06:55.360
25.

0:06:53.280,0:06:58.479
dynamic module means there is platform

0:06:55.360,0:07:00.560
specific native code involved

0:06:58.479,0:07:02.880
but since they are pre-compiled binaries

0:07:00.560,0:07:06.319
for the three major platforms

0:07:02.880,0:07:08.319
it should work in most places currently

0:07:06.319,0:07:09.440
the core functionalities are in a pretty

0:07:08.319,0:07:12.560
good shape

0:07:09.440,0:07:14.840
syntax highlighting is working nicely

0:07:12.560,0:07:16.080
the whole thing is split into three

0:07:14.840,0:07:17.759
packages

0:07:16.080,0:07:20.319
tree sitter is the main package that

0:07:17.759,0:07:22.800
other packages should depend on

0:07:20.319,0:07:24.000
tree system lens is the language bundle

0:07:22.800,0:07:27.199
that includes support

0:07:24.000,0:07:30.080
for most common languages

0:07:27.199,0:07:32.160
and finally the core apis are in the

0:07:30.080,0:07:36.160
package tsc

0:07:32.160,0:07:38.800
which stands for trees the core

0:07:36.160,0:07:41.919
it is the implicit dependency of the

0:07:38.800,0:07:41.919
three-seater package

0:07:43.520,0:07:47.520
the main package includes the miner mode

0:07:46.000,0:07:49.840
3-seater mode

0:07:47.520,0:07:52.560
this provides the base for other major

0:07:49.840,0:07:55.280
or minor modes to build on

0:07:52.560,0:07:55.840
using image change tracking hooks it

0:07:55.280,0:07:58.080
enables

0:07:55.840,0:08:00.800
incremental parsing and provides a

0:07:58.080,0:08:04.080
syntax tree that is always up to date

0:08:00.800,0:08:06.560
after any edits in a buffer

0:08:04.080,0:08:10.080
there is also a basic debug mode that

0:08:06.560,0:08:13.360
shows the parse tree in another buffer

0:08:10.080,0:08:15.759
here is a quick demo

0:08:13.360,0:08:17.520
here i mean an empty python buffer with

0:08:15.759,0:08:19.440
three seater enabled

0:08:17.520,0:08:26.560
i'm going to turn on the debug mode to

0:08:19.440,0:08:28.720
see the parse tree

0:08:26.560,0:08:30.639
since the buffer is empty there is only

0:08:28.720,0:08:33.279
one node in the syntax tree the top

0:08:30.639,0:08:41.839
level module node

0:08:33.279,0:08:41.839
let's try typing some code

0:09:11.040,0:09:14.640
as you can see as i type into the python

0:09:13.600,0:09:19.120
buffer

0:09:14.640,0:09:21.120
the syntax tree updates in real time

0:09:19.120,0:09:23.279
the other minor mode included in the

0:09:21.120,0:09:26.640
main package is 3-seater

0:09:23.279,0:09:28.480
hl mode it overrides font-lock mode and

0:09:26.640,0:09:31.839
provides its own set of phases

0:09:28.480,0:09:32.800
and customization options it is query

0:09:31.839,0:09:35.200
driven

0:09:32.800,0:09:36.240
that means instead of regular

0:09:35.200,0:09:38.720
expressions

0:09:36.240,0:09:40.320
it uses a list like query language to

0:09:38.720,0:09:43.760
map syntax notes

0:09:40.320,0:09:45.760
to highlighting phrases i'm going to

0:09:43.760,0:09:51.839
open a python file with small snippets

0:09:45.760,0:09:51.839
that showcase syntax highlighting

0:09:54.320,0:09:59.279
so this is the default highlighting

0:09:55.920,0:09:59.279
provided by python mode

0:10:00.880,0:10:04.640
this is the highlighting enabled by tree

0:10:02.839,0:10:07.680
sitter

0:10:04.640,0:10:11.680
as you can see string interpolation

0:10:07.680,0:10:15.440
and decorators are highlighted correctly

0:10:11.680,0:10:15.440
function calls are also highlighted

0:10:17.440,0:10:21.839
you can also note that property

0:10:20.240,0:10:24.640
assessors

0:10:21.839,0:10:27.200
and property assignments are highlighted

0:10:24.640,0:10:27.200
differently

0:10:27.440,0:10:30.880
what i like the most about this is that

0:10:29.360,0:10:32.640
new bindings are consistently

0:10:30.880,0:10:36.320
highlighted

0:10:32.640,0:10:39.760
this included local variable

0:10:36.320,0:10:42.480
function parameters and property

0:10:39.760,0:10:42.480
mutations

0:10:45.760,0:10:49.279
before going through the three queries

0:10:48.000,0:10:51.680
and the syntax highlighting

0:10:49.279,0:10:53.760
customization options

0:10:51.680,0:10:55.040
let's take a brief look at the core data

0:10:53.760,0:10:58.079
structures and functions

0:10:55.040,0:10:59.839
that tree sitter provides

0:10:58.079,0:11:02.240
so parsing is done with the help of a

0:10:59.839,0:11:04.160
generic parser object

0:11:02.240,0:11:06.000
a single parser object can be used to

0:11:04.160,0:11:08.320
pass different languages

0:11:06.000,0:11:09.279
by sending different language objects to

0:11:08.320,0:11:10.880
it

0:11:09.279,0:11:14.079
the language objects themselves are

0:11:10.880,0:11:16.079
loaded from shared libraries

0:11:14.079,0:11:17.360
since three seater mode already handles

0:11:16.079,0:11:19.440
the parsing part

0:11:17.360,0:11:20.800
we will instead focus on the functions

0:11:19.440,0:11:24.720
that inspect nodes

0:11:20.800,0:11:24.720
and in the resulting path tree

0:11:25.279,0:11:43.839
we can ask tree sitter what is the

0:11:27.200,0:11:43.839
syntax node at point

0:11:44.240,0:11:48.480
uh is it an opaque object so this is not

0:11:47.200,0:11:57.839
very useful

0:11:48.480,0:11:57.839
we can instead ask what is its type

0:12:03.760,0:12:08.959
so his type is the symbol comparison

0:12:06.560,0:12:11.600
operator

0:12:08.959,0:12:13.680
trees there are two kinds of nodes

0:12:11.600,0:12:15.519
anonymous nodes and named nodes

0:12:13.680,0:12:17.040
anonymous nodes correspond to simple

0:12:15.519,0:12:19.839
grammar elements

0:12:17.040,0:12:21.279
like keywords operators punctuations and

0:12:19.839,0:12:24.160
so on

0:12:21.279,0:12:25.920
name nodes on the other hand grammar

0:12:24.160,0:12:26.639
elements that are interesting enough for

0:12:25.920,0:12:30.320
their own

0:12:26.639,0:12:31.839
to have a name like an identifier an

0:12:30.320,0:12:35.200
expression

0:12:31.839,0:12:35.200
or a function definition

0:12:35.440,0:12:41.519
name node types are symbols while

0:12:37.760,0:12:41.519
anonymous node types are strings

0:12:42.639,0:12:49.519
for example if we are on this

0:12:46.320,0:12:49.519
comparison operator

0:12:49.760,0:12:53.839
the node type should be a string

0:12:55.920,0:12:58.959
we can also get other information about

0:12:57.920,0:13:07.839
the node

0:12:58.959,0:13:07.839
for example what is this text

0:13:09.680,0:13:35.839
or where it is in the buffer

0:13:20.800,0:13:35.839
or what is its parent

0:13:43.199,0:13:46.839
there are many other apis to query or

0:13:46.160,0:13:49.839
not

0:13:46.839,0:13:49.839
properties

0:13:52.639,0:13:58.240
tree sitter allows searching for

0:13:54.399,0:14:01.440
structural patterns within a parse tree

0:13:58.240,0:14:03.519
it does so through a list like language

0:14:01.440,0:14:04.639
this language supports by the matching

0:14:03.519,0:14:07.760
by node types

0:14:04.639,0:14:10.079
field names and predicates

0:14:07.760,0:14:12.639
it also allows capturing nodes for

0:14:10.079,0:14:17.839
further processing

0:14:12.639,0:14:17.839
let's try to see some examples

0:14:37.680,0:14:43.839
so in this very simple query we just

0:14:41.040,0:14:46.399
try to highlight all the identifiers in

0:14:43.839,0:14:46.399
the buffer

0:14:49.040,0:14:53.120
this s side tells trisito to capture a

0:14:51.920,0:14:55.839
node

0:14:53.120,0:14:57.360
in the context of the query builder it's

0:14:55.839,0:15:00.320
not very important

0:14:57.360,0:15:01.760
but in normal highlighting query this

0:15:00.320,0:15:05.920
will determine

0:15:01.760,0:15:05.920
the face used to highlight the note

0:15:06.639,0:15:10.320
suppose we want to capture all the

0:15:08.800,0:15:13.519
function names

0:15:10.320,0:15:27.839
instead of just any identifier

0:15:13.519,0:15:27.839
you can improve the query like this

0:15:29.440,0:15:32.639
uh this will highlight the whole

0:15:31.600,0:15:35.519
definition

0:15:32.639,0:15:36.399
but we only want to capture the function

0:15:35.519,0:15:39.600
name

0:15:36.399,0:15:42.800
which means the identifier

0:15:39.600,0:15:46.320
here so we

0:15:42.800,0:15:48.639
move the capture to after the identifier

0:15:46.320,0:15:48.639
node

0:15:49.600,0:15:52.959
if we want to capture the class names as

0:15:51.759,0:16:09.839
well

0:15:52.959,0:16:09.839
we just add another pattern

0:16:10.079,0:16:14.399
let's look at a more practical example

0:16:20.320,0:16:23.759
here we can see that single quotes

0:16:22.959,0:16:25.600
strings and

0:16:23.759,0:16:27.279
double quotes screens are highlighted

0:16:25.600,0:16:30.399
the same

0:16:27.279,0:16:33.440
but in some places

0:16:30.399,0:16:35.440
because of some coding conventions

0:16:33.440,0:16:37.279
it may be desirable to highlight them

0:16:35.440,0:16:39.680
differently for example if

0:16:37.279,0:16:40.880
the string is single quoted we may want

0:16:39.680,0:16:43.759
to highlight it

0:16:40.880,0:16:43.759
as a constant

0:16:44.399,0:16:47.600
let's try to see whether we can

0:16:46.160,0:16:51.839
distinguish these

0:16:47.600,0:16:51.839
two cases

0:16:56.240,0:17:00.160
so here we get all the strings

0:17:00.639,0:17:04.559
if we want to see if it's single quotes

0:17:04.079,0:17:07.520
or

0:17:04.559,0:17:07.520
double quote strings

0:17:08.799,0:17:12.480
we can try looking at the first

0:17:11.039,0:17:15.280
character

0:17:12.480,0:17:16.720
of the string i mean the first character

0:17:15.280,0:17:19.360
of the note

0:17:16.720,0:17:33.600
to check whether it's a single quote or

0:17:19.360,0:17:36.080
a double quote

0:17:33.600,0:17:36.799
yeah so for that we use the three

0:17:36.080,0:17:40.160
setters

0:17:36.799,0:17:43.360
support for predicate in this case

0:17:40.160,0:17:46.080
we use a match predicate

0:17:43.360,0:17:46.799
to check whether the string where the

0:17:46.080,0:17:50.320
note

0:17:46.799,0:17:51.280
starts with a single quote and with this

0:17:50.320,0:17:55.520
pattern

0:17:51.280,0:17:55.520
we only capture the single quotes

0:17:58.840,0:18:03.760
strings

0:18:00.400,0:18:07.760
let's try to give it a different face

0:18:03.760,0:18:07.760
so we copy the pattern

0:18:13.039,0:18:16.640
and we add this pattern

0:18:18.640,0:18:21.760
pop item only

0:18:25.120,0:18:31.440
but we also want to give the

0:18:28.400,0:18:36.320
capture a different name

0:18:31.440,0:18:36.320
let's say we want to highlight it as a

0:18:40.840,0:18:43.840
keyword

0:18:46.559,0:18:57.840
and now if we refresh the buffer

0:19:06.320,0:19:10.320
we see that single quote strings are

0:19:08.799,0:19:12.880
highlighted as

0:19:10.320,0:19:12.880
keywords

0:19:14.400,0:19:19.200
the highlighting patterns can also be

0:19:16.400,0:19:23.280
set for a single project

0:19:19.200,0:19:23.280
using directory local variable

0:19:23.440,0:19:30.000
for example let's take a look at

0:19:26.880,0:19:30.000
ems source code

0:19:35.760,0:19:43.760
so in image c source there are a lot of

0:19:40.400,0:19:47.679
uses of these different macros

0:19:43.760,0:19:50.400
to define functions

0:19:47.679,0:19:50.400
and you can see

0:19:51.200,0:19:55.760
this is actually the function name but

0:19:53.520,0:19:59.120
it's highlighted as the

0:19:55.760,0:20:03.679
string so what we want

0:19:59.120,0:20:07.600
is to somehow recognize this pattern

0:20:03.679,0:20:11.280
and highlight it

0:20:07.600,0:20:14.559
as highlight this part

0:20:11.280,0:20:17.679
with the function phase instead

0:20:14.559,0:20:20.240
in order to do that

0:20:17.679,0:20:21.760
we put a pattern in this project

0:20:20.240,0:20:24.880
directory local

0:20:21.760,0:20:24.880
settings file

0:20:31.760,0:20:37.760
so we can put this button in the c

0:20:34.799,0:20:37.760
mode section

0:20:40.159,0:20:50.480
and now if we enable tree sitter

0:20:48.000,0:20:52.720
you can see that this is the highlighted

0:20:50.480,0:20:52.720
uh

0:20:53.200,0:20:56.559
as a normal function definition so this

0:20:55.520,0:21:00.400
is the function

0:20:56.559,0:21:00.400
face like we wanted

0:21:01.200,0:21:06.080
the pattern for this is actually pretty

0:21:03.760,0:21:06.080
simple

0:21:07.200,0:21:09.919
it's only

0:21:10.720,0:21:17.440
only this part so

0:21:14.720,0:21:19.679
if it's a function call where the name

0:21:17.440,0:21:21.600
of the function is different

0:21:19.679,0:21:24.159
then we highlight the different as a

0:21:21.600,0:21:24.159
keyword

0:21:24.240,0:21:28.159
and then the first string element we

0:21:27.360,0:21:31.840
highlighted

0:21:28.159,0:21:31.840
as a function name

0:21:35.360,0:21:39.280
since the language objects are actually

0:21:37.679,0:21:40.799
native code

0:21:39.280,0:21:43.440
they have to be compiled for each

0:21:40.799,0:21:45.600
platform that we want to support

0:21:43.440,0:21:48.159
this will become a big obstacle for

0:21:45.600,0:21:50.240
3-seater adoption

0:21:48.159,0:21:52.960
therefore i've created a language window

0:21:50.240,0:21:54.960
package 3-seater length

0:21:52.960,0:21:56.320
that takes care of pre-compiling the

0:21:54.960,0:21:59.679
grammars the

0:21:56.320,0:22:01.600
most common grammars for all three major

0:21:59.679,0:22:04.080
platforms

0:22:01.600,0:22:05.360
it also takes care of distributing these

0:22:04.080,0:22:08.080
binaries

0:22:05.360,0:22:11.280
and provides some highlighting queries

0:22:08.080,0:22:11.280
for some of the languages

0:22:11.440,0:22:15.919
it should be noted that this package

0:22:13.760,0:22:19.520
should be treated as a temporary

0:22:15.919,0:22:19.520
distribution mechanism only

0:22:19.919,0:22:24.720
to help with bootstrapping three-seaters

0:22:22.240,0:22:27.760
adoption

0:22:24.720,0:22:29.760
the plan is that eventually these files

0:22:27.760,0:22:32.480
should be provided by the language major

0:22:29.760,0:22:35.120
modes themselves

0:22:32.480,0:22:36.320
but in order to do that we need better

0:22:35.120,0:22:40.240
tooling

0:22:36.320,0:22:42.559
so we're not there yet

0:22:40.240,0:22:43.280
since the call already works reasonably

0:22:42.559,0:22:44.640
well

0:22:43.280,0:22:46.320
there are several areas that would

0:22:44.640,0:22:48.960
benefit from the community's

0:22:46.320,0:22:48.960
contribution

0:22:49.120,0:22:52.640
so three seaters upstream language

0:22:51.520,0:22:54.400
prepositories

0:22:52.640,0:22:55.679
already contain highlighting queries on

0:22:54.400,0:22:58.480
their own

0:22:55.679,0:23:00.480
however they are pretty basic and they

0:22:58.480,0:23:02.559
may not fit well with existing emax

0:23:00.480,0:23:04.320
conventions

0:23:02.559,0:23:07.120
therefore the language bundle has its

0:23:04.320,0:23:10.559
own set of highlighting queries

0:23:07.120,0:23:11.600
this requires maintenance until language

0:23:10.559,0:23:13.760
measurements adopt

0:23:11.600,0:23:16.240
three sitter and maintain the queries on

0:23:13.760,0:23:16.240
their own

0:23:16.640,0:23:22.000
the queries are actually quite easy to

0:23:18.480,0:23:24.240
write as you've already seen

0:23:22.000,0:23:25.360
you just need to be familiar with the

0:23:24.240,0:23:30.000
language

0:23:25.360,0:23:32.880
familiar enough to come up with sensible

0:23:30.000,0:23:32.880
highlighting patterns

0:23:35.200,0:23:39.679
and if you are a maintainer of a

0:23:37.600,0:23:42.320
language major mode

0:23:39.679,0:23:43.360
you may want to consider integrating

0:23:42.320,0:23:46.960
tree sitter into

0:23:43.360,0:23:50.080
your mode initially maybe as an

0:23:46.960,0:23:53.279
optional feature the integration is

0:23:50.080,0:23:56.640
actually pretty straightforward

0:23:53.279,0:24:00.880
especially for syntax highlighting

0:23:56.640,0:24:00.880
or alternatively

0:24:01.520,0:24:04.640
you can also try writing a new major

0:24:03.760,0:24:08.000
mode

0:24:04.640,0:24:11.360
from scratch that relies on tree sitter

0:24:08.000,0:24:11.360
from the very beginning

0:24:12.559,0:24:19.679
the code for such a major mode is

0:24:16.320,0:24:23.200
quite simple for example

0:24:19.679,0:24:26.240
this is the proposed

0:24:23.200,0:24:30.720
what mode for web assembly

0:24:26.240,0:24:30.720
the code is just

0:24:31.039,0:24:37.120
like one page of code not

0:24:34.559,0:24:37.120
not a lot

0:24:39.520,0:24:46.559
you can also try writing new minor modes

0:24:42.720,0:24:50.080
or writing integration packages

0:24:46.559,0:24:50.880
for example a lot of package a lot of

0:24:50.080,0:24:54.559
packages

0:24:50.880,0:24:58.840
may benefit from tree sitter integration

0:24:54.559,0:25:01.840
but no one has written the integration

0:24:58.840,0:25:01.840
yet

0:25:02.960,0:25:06.720
if you are interested in 3-seater you

0:25:05.039,0:25:10.320
can use these links to

0:25:06.720,0:25:11.440
learn more about it i think that's it

0:25:10.320,0:25:18.159
for me today

0:25:11.440,0:25:18.159
i'm happy to answer any questions