-
-
Notifications
You must be signed in to change notification settings - Fork 2.5k
/
document.rs
2264 lines (1995 loc) · 80.4 KB
/
document.rs
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
use anyhow::{anyhow, bail, Context, Error};
use arc_swap::access::DynAccess;
use arc_swap::ArcSwap;
use futures_util::future::BoxFuture;
use futures_util::FutureExt;
use helix_core::auto_pairs::AutoPairs;
use helix_core::chars::char_is_word;
use helix_core::doc_formatter::TextFormat;
use helix_core::encoding::Encoding;
use helix_core::syntax::{Highlight, LanguageServerFeature};
use helix_core::text_annotations::{InlineAnnotation, Overlay};
use helix_lsp::util::lsp_pos_to_pos;
use helix_stdx::faccess::{copy_metadata, readonly};
use helix_vcs::{DiffHandle, DiffProviderRegistry};
use ::parking_lot::Mutex;
use serde::de::{self, Deserialize, Deserializer};
use serde::Serialize;
use std::borrow::Cow;
use std::cell::Cell;
use std::collections::HashMap;
use std::fmt::Display;
use std::future::Future;
use std::path::{Path, PathBuf};
use std::str::FromStr;
use std::sync::{Arc, Weak};
use std::time::SystemTime;
use helix_core::{
encoding,
history::{History, State, UndoKind},
indent::{auto_detect_indent_style, IndentStyle},
line_ending::auto_detect_line_ending,
syntax::{self, LanguageConfiguration},
ChangeSet, Diagnostic, LineEnding, Range, Rope, RopeBuilder, Selection, Syntax, Transaction,
};
use crate::editor::Config;
use crate::events::{DocumentDidChange, SelectionDidChange};
use crate::{DocumentId, Editor, Theme, View, ViewId};
/// 8kB of buffer space for encoding and decoding `Rope`s.
const BUF_SIZE: usize = 8192;
const DEFAULT_INDENT: IndentStyle = IndentStyle::Tabs;
pub const DEFAULT_LANGUAGE_NAME: &str = "text";
pub const SCRATCH_BUFFER_NAME: &str = "[scratch]";
#[derive(Debug, Copy, Clone, PartialEq, Eq, Hash)]
pub enum Mode {
Normal = 0,
Select = 1,
Insert = 2,
}
impl Display for Mode {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
match self {
Mode::Normal => f.write_str("normal"),
Mode::Select => f.write_str("select"),
Mode::Insert => f.write_str("insert"),
}
}
}
impl FromStr for Mode {
type Err = Error;
fn from_str(s: &str) -> Result<Self, Self::Err> {
match s {
"normal" => Ok(Mode::Normal),
"select" => Ok(Mode::Select),
"insert" => Ok(Mode::Insert),
_ => bail!("Invalid mode '{}'", s),
}
}
}
// toml deserializer doesn't seem to recognize string as enum
impl<'de> Deserialize<'de> for Mode {
fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
where
D: Deserializer<'de>,
{
let s = String::deserialize(deserializer)?;
s.parse().map_err(de::Error::custom)
}
}
impl Serialize for Mode {
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
serializer.collect_str(self)
}
}
/// A snapshot of the text of a document that we want to write out to disk
#[derive(Debug, Clone)]
pub struct DocumentSavedEvent {
pub revision: usize,
pub doc_id: DocumentId,
pub path: PathBuf,
pub text: Rope,
}
pub type DocumentSavedEventResult = Result<DocumentSavedEvent, anyhow::Error>;
pub type DocumentSavedEventFuture = BoxFuture<'static, DocumentSavedEventResult>;
#[derive(Debug)]
pub struct SavePoint {
/// The view this savepoint is associated with
pub view: ViewId,
revert: Mutex<Transaction>,
}
pub struct Document {
pub(crate) id: DocumentId,
text: Rope,
selections: HashMap<ViewId, Selection>,
/// Inlay hints annotations for the document, by view.
///
/// To know if they're up-to-date, check the `id` field in `DocumentInlayHints`.
pub(crate) inlay_hints: HashMap<ViewId, DocumentInlayHints>,
pub(crate) jump_labels: HashMap<ViewId, Vec<Overlay>>,
/// Set to `true` when the document is updated, reset to `false` on the next inlay hints
/// update from the LSP
pub inlay_hints_oudated: bool,
path: Option<PathBuf>,
encoding: &'static encoding::Encoding,
has_bom: bool,
pub restore_cursor: bool,
/// Current indent style.
pub indent_style: IndentStyle,
/// The document's default line ending.
pub line_ending: LineEnding,
pub syntax: Option<Syntax>,
/// Corresponding language scope name. Usually `source.<lang>`.
pub language: Option<Arc<LanguageConfiguration>>,
/// Pending changes since last history commit.
changes: ChangeSet,
/// State at last commit. Used for calculating reverts.
old_state: Option<State>,
/// Undo tree.
// It can be used as a cell where we will take it out to get some parts of the history and put
// it back as it separated from the edits. We could split out the parts manually but that will
// be more troublesome.
pub history: Cell<History>,
pub config: Arc<dyn DynAccess<Config>>,
savepoints: Vec<Weak<SavePoint>>,
// Last time we wrote to the file. This will carry the time the file was last opened if there
// were no saves.
last_saved_time: SystemTime,
last_saved_revision: usize,
version: i32, // should be usize?
pub(crate) modified_since_accessed: bool,
pub(crate) diagnostics: Vec<Diagnostic>,
pub(crate) language_servers: HashMap<LanguageServerName, Arc<Client>>,
diff_handle: Option<DiffHandle>,
version_control_head: Option<Arc<ArcSwap<Box<str>>>>,
// when document was used for most-recent-used buffer picker
pub focused_at: std::time::Instant,
pub readonly: bool,
}
/// Inlay hints for a single `(Document, View)` combo.
///
/// There are `*_inlay_hints` field for each kind of hints an LSP can send since we offer the
/// option to style theme differently in the theme according to the (currently supported) kinds
/// (`type`, `parameter` and the rest).
///
/// Inlay hints are always `InlineAnnotation`s, not overlays or line-ones: LSP may choose to place
/// them anywhere in the text and will sometime offer config options to move them where the user
/// wants them but it shouldn't be Helix who decides that so we use the most precise positioning.
///
/// The padding for inlay hints needs to be stored separately for before and after (the LSP spec
/// uses 'left' and 'right' but not all text is left to right so let's be correct) padding because
/// the 'before' padding must be added to a layer *before* the regular inlay hints and the 'after'
/// padding comes ... after.
#[derive(Debug, Clone)]
pub struct DocumentInlayHints {
/// Identifier for the inlay hints stored in this structure. To be checked to know if they have
/// to be recomputed on idle or not.
pub id: DocumentInlayHintsId,
/// Inlay hints of `TYPE` kind, if any.
pub type_inlay_hints: Vec<InlineAnnotation>,
/// Inlay hints of `PARAMETER` kind, if any.
pub parameter_inlay_hints: Vec<InlineAnnotation>,
/// Inlay hints that are neither `TYPE` nor `PARAMETER`.
///
/// LSPs are not required to associate a kind to their inlay hints, for example Rust-Analyzer
/// currently never does (February 2023) and the LSP spec may add new kinds in the future that
/// we want to display even if we don't have some special highlighting for them.
pub other_inlay_hints: Vec<InlineAnnotation>,
/// Inlay hint padding. When creating the final `TextAnnotations`, the `before` padding must be
/// added first, then the regular inlay hints, then the `after` padding.
pub padding_before_inlay_hints: Vec<InlineAnnotation>,
pub padding_after_inlay_hints: Vec<InlineAnnotation>,
}
impl DocumentInlayHints {
/// Generate an empty list of inlay hints with the given ID.
pub fn empty_with_id(id: DocumentInlayHintsId) -> Self {
Self {
id,
type_inlay_hints: Vec::new(),
parameter_inlay_hints: Vec::new(),
other_inlay_hints: Vec::new(),
padding_before_inlay_hints: Vec::new(),
padding_after_inlay_hints: Vec::new(),
}
}
}
/// Associated with a [`Document`] and [`ViewId`], uniquely identifies the state of inlay hints for
/// for that document and view: if this changed since the last save, the inlay hints for the view
/// should be recomputed.
///
/// We can't store the `ViewOffset` instead of the first and last asked-for lines because if
/// softwrapping changes, the `ViewOffset` may not change while the displayed lines will.
#[derive(Copy, Clone, PartialEq, Eq)]
pub struct DocumentInlayHintsId {
/// First line for which the inlay hints were requested.
pub first_line: usize,
/// Last line for which the inlay hints were requested.
pub last_line: usize,
}
use std::{fmt, mem};
impl fmt::Debug for Document {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_struct("Document")
.field("id", &self.id)
.field("text", &self.text)
.field("selections", &self.selections)
.field("inlay_hints_oudated", &self.inlay_hints_oudated)
.field("text_annotations", &self.inlay_hints)
.field("path", &self.path)
.field("encoding", &self.encoding)
.field("restore_cursor", &self.restore_cursor)
.field("syntax", &self.syntax)
.field("language", &self.language)
.field("changes", &self.changes)
.field("old_state", &self.old_state)
// .field("history", &self.history)
.field("last_saved_time", &self.last_saved_time)
.field("last_saved_revision", &self.last_saved_revision)
.field("version", &self.version)
.field("modified_since_accessed", &self.modified_since_accessed)
.field("diagnostics", &self.diagnostics)
// .field("language_server", &self.language_server)
.finish()
}
}
impl fmt::Debug for DocumentInlayHintsId {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
// Much more agreable to read when debugging
f.debug_struct("DocumentInlayHintsId")
.field("lines", &(self.first_line..self.last_line))
.finish()
}
}
enum Encoder {
Utf16Be,
Utf16Le,
EncodingRs(encoding::Encoder),
}
impl Encoder {
fn from_encoding(encoding: &'static encoding::Encoding) -> Self {
if encoding == encoding::UTF_16BE {
Self::Utf16Be
} else if encoding == encoding::UTF_16LE {
Self::Utf16Le
} else {
Self::EncodingRs(encoding.new_encoder())
}
}
fn encode_from_utf8(
&mut self,
src: &str,
dst: &mut [u8],
is_empty: bool,
) -> (encoding::CoderResult, usize, usize) {
if src.is_empty() {
return (encoding::CoderResult::InputEmpty, 0, 0);
}
let mut write_to_buf = |convert: fn(u16) -> [u8; 2]| {
let to_write = src.char_indices().map(|(indice, char)| {
let mut encoded: [u16; 2] = [0, 0];
(
indice,
char.encode_utf16(&mut encoded)
.iter_mut()
.flat_map(|char| convert(*char))
.collect::<Vec<u8>>(),
)
});
let mut total_written = 0usize;
for (indice, utf16_bytes) in to_write {
let character_size = utf16_bytes.len();
if dst.len() <= (total_written + character_size) {
return (encoding::CoderResult::OutputFull, indice, total_written);
}
for character in utf16_bytes {
dst[total_written] = character;
total_written += 1;
}
}
(encoding::CoderResult::InputEmpty, src.len(), total_written)
};
match self {
Self::Utf16Be => write_to_buf(u16::to_be_bytes),
Self::Utf16Le => write_to_buf(u16::to_le_bytes),
Self::EncodingRs(encoder) => {
let (code_result, read, written, ..) = encoder.encode_from_utf8(src, dst, is_empty);
(code_result, read, written)
}
}
}
}
// Apply BOM if encoding permit it, return the number of bytes written at the start of buf
fn apply_bom(encoding: &'static encoding::Encoding, buf: &mut [u8; BUF_SIZE]) -> usize {
if encoding == encoding::UTF_8 {
buf[0] = 0xef;
buf[1] = 0xbb;
buf[2] = 0xbf;
3
} else if encoding == encoding::UTF_16BE {
buf[0] = 0xfe;
buf[1] = 0xff;
2
} else if encoding == encoding::UTF_16LE {
buf[0] = 0xff;
buf[1] = 0xfe;
2
} else {
0
}
}
// The documentation and implementation of this function should be up-to-date with
// its sibling function, `to_writer()`.
//
/// Decodes a stream of bytes into UTF-8, returning a `Rope` and the
/// encoding it was decoded as with BOM information. The optional `encoding`
/// parameter can be used to override encoding auto-detection.
pub fn from_reader<R: std::io::Read + ?Sized>(
reader: &mut R,
encoding: Option<&'static Encoding>,
) -> Result<(Rope, &'static Encoding, bool), Error> {
// These two buffers are 8192 bytes in size each and are used as
// intermediaries during the decoding process. Text read into `buf`
// from `reader` is decoded into `buf_out` as UTF-8. Once either
// `buf_out` is full or the end of the reader was reached, the
// contents are appended to `builder`.
let mut buf = [0u8; BUF_SIZE];
let mut buf_out = [0u8; BUF_SIZE];
let mut builder = RopeBuilder::new();
let (encoding, has_bom, mut decoder, read) =
read_and_detect_encoding(reader, encoding, &mut buf)?;
let mut slice = &buf[..read];
let mut is_empty = read == 0;
// `RopeBuilder::append()` expects a `&str`, so this is the "real"
// output buffer. When decoding, the number of bytes in the output
// buffer will often exceed the number of bytes in the input buffer.
// The `result` returned by `decode_to_str()` will state whether or
// not that happened. The contents of `buf_str` is appended to
// `builder` and it is reused for the next iteration of the decoding
// loop.
//
// As it is possible to read less than the buffer's maximum from `read()`
// even when the end of the reader has yet to be reached, the end of
// the reader is determined only when a `read()` call returns `0`.
//
// SAFETY: `buf_out` is a zero-initialized array, thus it will always
// contain valid UTF-8.
let buf_str = unsafe { std::str::from_utf8_unchecked_mut(&mut buf_out[..]) };
let mut total_written = 0usize;
loop {
let mut total_read = 0usize;
// An inner loop is necessary as it is possible that the input buffer
// may not be completely decoded on the first `decode_to_str()` call
// which would happen in cases where the output buffer is filled to
// capacity.
loop {
let (result, read, written, ..) = decoder.decode_to_str(
&slice[total_read..],
&mut buf_str[total_written..],
is_empty,
);
// These variables act as the read and write cursors of `buf` and `buf_str` respectively.
// They are necessary in case the output buffer fills before decoding of the entire input
// loop is complete. Otherwise, the loop would endlessly iterate over the same `buf` and
// the data inside the output buffer would be overwritten.
total_read += read;
total_written += written;
match result {
encoding::CoderResult::InputEmpty => {
debug_assert_eq!(slice.len(), total_read);
break;
}
encoding::CoderResult::OutputFull => {
debug_assert!(slice.len() > total_read);
builder.append(&buf_str[..total_written]);
total_written = 0;
}
}
}
// Once the end of the stream is reached, the output buffer is
// flushed and the loop terminates.
if is_empty {
debug_assert_eq!(reader.read(&mut buf)?, 0);
builder.append(&buf_str[..total_written]);
break;
}
// Once the previous input has been processed and decoded, the next set of
// data is fetched from the reader. The end of the reader is determined to
// be when exactly `0` bytes were read from the reader, as per the invariants
// of the `Read` trait.
let read = reader.read(&mut buf)?;
slice = &buf[..read];
is_empty = read == 0;
}
let rope = builder.finish();
Ok((rope, encoding, has_bom))
}
pub fn read_to_string<R: std::io::Read + ?Sized>(
reader: &mut R,
encoding: Option<&'static Encoding>,
) -> Result<(String, &'static Encoding, bool), Error> {
let mut buf = [0u8; BUF_SIZE];
let (encoding, has_bom, mut decoder, read) =
read_and_detect_encoding(reader, encoding, &mut buf)?;
let mut slice = &buf[..read];
let mut is_empty = read == 0;
let mut buf_string = String::with_capacity(buf.len());
loop {
let mut total_read = 0usize;
loop {
let (result, read, ..) =
decoder.decode_to_string(&slice[total_read..], &mut buf_string, is_empty);
total_read += read;
match result {
encoding::CoderResult::InputEmpty => {
debug_assert_eq!(slice.len(), total_read);
break;
}
encoding::CoderResult::OutputFull => {
debug_assert!(slice.len() > total_read);
buf_string.reserve(buf.len())
}
}
}
if is_empty {
debug_assert_eq!(reader.read(&mut buf)?, 0);
break;
}
let read = reader.read(&mut buf)?;
slice = &buf[..read];
is_empty = read == 0;
}
Ok((buf_string, encoding, has_bom))
}
/// Reads the first chunk from a Reader into the given buffer
/// and detects the encoding.
///
/// By default, the encoding of the text is auto-detected by
/// `encoding_rs` for_bom, and if it fails, from `chardetng`
/// crate which requires sample data from the reader.
/// As a manual override to this auto-detection is possible, the
/// same data is read into `buf` to ensure symmetry in the upcoming
/// loop.
fn read_and_detect_encoding<R: std::io::Read + ?Sized>(
reader: &mut R,
encoding: Option<&'static Encoding>,
buf: &mut [u8],
) -> Result<(&'static Encoding, bool, encoding::Decoder, usize), Error> {
let read = reader.read(buf)?;
let is_empty = read == 0;
let (encoding, has_bom) = encoding
.map(|encoding| (encoding, false))
.or_else(|| encoding::Encoding::for_bom(buf).map(|(encoding, _bom_size)| (encoding, true)))
.unwrap_or_else(|| {
let mut encoding_detector = chardetng::EncodingDetector::new();
encoding_detector.feed(buf, is_empty);
(encoding_detector.guess(None, true), false)
});
let decoder = encoding.new_decoder();
Ok((encoding, has_bom, decoder, read))
}
// The documentation and implementation of this function should be up-to-date with
// its sibling function, `from_reader()`.
//
/// Encodes the text inside `rope` into the given `encoding` and writes the
/// encoded output into `writer.` As a `Rope` can only contain valid UTF-8,
/// replacement characters may appear in the encoded text.
pub async fn to_writer<'a, W: tokio::io::AsyncWriteExt + Unpin + ?Sized>(
writer: &'a mut W,
encoding_with_bom_info: (&'static Encoding, bool),
rope: &'a Rope,
) -> Result<(), Error> {
// Text inside a `Rope` is stored as non-contiguous blocks of data called
// chunks. The absolute size of each chunk is unknown, thus it is impossible
// to predict the end of the chunk iterator ahead of time. Instead, it is
// determined by filtering the iterator to remove all empty chunks and then
// appending an empty chunk to it. This is valuable for detecting when all
// chunks in the `Rope` have been iterated over in the subsequent loop.
let (encoding, has_bom) = encoding_with_bom_info;
let iter = rope
.chunks()
.filter(|c| !c.is_empty())
.chain(std::iter::once(""));
let mut buf = [0u8; BUF_SIZE];
let mut total_written = if has_bom {
apply_bom(encoding, &mut buf)
} else {
0
};
let mut encoder = Encoder::from_encoding(encoding);
for chunk in iter {
let is_empty = chunk.is_empty();
let mut total_read = 0usize;
// An inner loop is necessary as it is possible that the input buffer
// may not be completely encoded on the first `encode_from_utf8()` call
// which would happen in cases where the output buffer is filled to
// capacity.
loop {
let (result, read, written, ..) =
encoder.encode_from_utf8(&chunk[total_read..], &mut buf[total_written..], is_empty);
// These variables act as the read and write cursors of `chunk` and `buf` respectively.
// They are necessary in case the output buffer fills before encoding of the entire input
// loop is complete. Otherwise, the loop would endlessly iterate over the same `chunk` and
// the data inside the output buffer would be overwritten.
total_read += read;
total_written += written;
match result {
encoding::CoderResult::InputEmpty => {
debug_assert_eq!(chunk.len(), total_read);
debug_assert!(buf.len() >= total_written);
break;
}
encoding::CoderResult::OutputFull => {
debug_assert!(chunk.len() > total_read);
writer.write_all(&buf[..total_written]).await?;
total_written = 0;
}
}
}
// Once the end of the iterator is reached, the output buffer is
// flushed and the outer loop terminates.
if is_empty {
writer.write_all(&buf[..total_written]).await?;
writer.flush().await?;
break;
}
}
Ok(())
}
fn take_with<T, F>(mut_ref: &mut T, f: F)
where
T: Default,
F: FnOnce(T) -> T,
{
*mut_ref = f(mem::take(mut_ref));
}
use helix_lsp::{lsp, Client, LanguageServerId, LanguageServerName};
use url::Url;
impl Document {
pub fn from(
text: Rope,
encoding_with_bom_info: Option<(&'static Encoding, bool)>,
config: Arc<dyn DynAccess<Config>>,
) -> Self {
let (encoding, has_bom) = encoding_with_bom_info.unwrap_or((encoding::UTF_8, false));
let line_ending = config.load().default_line_ending.into();
let changes = ChangeSet::new(text.slice(..));
let old_state = None;
Self {
id: DocumentId::default(),
path: None,
encoding,
has_bom,
text,
selections: HashMap::default(),
inlay_hints: HashMap::default(),
inlay_hints_oudated: false,
indent_style: DEFAULT_INDENT,
line_ending,
restore_cursor: false,
syntax: None,
language: None,
changes,
old_state,
diagnostics: Vec::new(),
version: 0,
history: Cell::new(History::default()),
savepoints: Vec::new(),
last_saved_time: SystemTime::now(),
last_saved_revision: 0,
modified_since_accessed: false,
language_servers: HashMap::new(),
diff_handle: None,
config,
version_control_head: None,
focused_at: std::time::Instant::now(),
readonly: false,
jump_labels: HashMap::new(),
}
}
pub fn default(config: Arc<dyn DynAccess<Config>>) -> Self {
let line_ending: LineEnding = config.load().default_line_ending.into();
let text = Rope::from(line_ending.as_str());
Self::from(text, None, config)
}
// TODO: async fn?
/// Create a new document from `path`. Encoding is auto-detected, but it can be manually
/// overwritten with the `encoding` parameter.
pub fn open(
path: &Path,
encoding: Option<&'static Encoding>,
config_loader: Option<Arc<ArcSwap<syntax::Loader>>>,
config: Arc<dyn DynAccess<Config>>,
) -> Result<Self, Error> {
// Open the file if it exists, otherwise assume it is a new file (and thus empty).
let (rope, encoding, has_bom) = if path.exists() {
let mut file =
std::fs::File::open(path).context(format!("unable to open {:?}", path))?;
from_reader(&mut file, encoding)?
} else {
let line_ending: LineEnding = config.load().default_line_ending.into();
let encoding = encoding.unwrap_or(encoding::UTF_8);
(Rope::from(line_ending.as_str()), encoding, false)
};
let mut doc = Self::from(rope, Some((encoding, has_bom)), config);
// set the path and try detecting the language
doc.set_path(Some(path));
if let Some(loader) = config_loader {
doc.detect_language(loader);
}
doc.detect_indent_and_line_ending();
Ok(doc)
}
/// The same as [`format`], but only returns formatting changes if auto-formatting
/// is configured.
pub fn auto_format(&self) -> Option<BoxFuture<'static, Result<Transaction, FormatterError>>> {
if self.language_config()?.auto_format {
self.format()
} else {
None
}
}
/// If supported, returns the changes that should be applied to this document in order
/// to format it nicely.
// We can't use anyhow::Result here since the output of the future has to be
// clonable to be used as shared future. So use a custom error type.
pub fn format(&self) -> Option<BoxFuture<'static, Result<Transaction, FormatterError>>> {
if let Some((fmt_cmd, fmt_args)) = self
.language_config()
.and_then(|c| c.formatter.as_ref())
.and_then(|formatter| {
Some((
helix_stdx::env::which(&formatter.command).ok()?,
&formatter.args,
))
})
{
use std::process::Stdio;
let text = self.text().clone();
let mut process = tokio::process::Command::new(&fmt_cmd);
process
.args(fmt_args)
.stdin(Stdio::piped())
.stdout(Stdio::piped())
.stderr(Stdio::piped());
let formatting_future = async move {
let mut process = process
.spawn()
.map_err(|e| FormatterError::SpawningFailed {
command: fmt_cmd.to_string_lossy().into(),
error: e.kind(),
})?;
{
let mut stdin = process.stdin.take().ok_or(FormatterError::BrokenStdin)?;
to_writer(&mut stdin, (encoding::UTF_8, false), &text)
.await
.map_err(|_| FormatterError::BrokenStdin)?;
}
let output = process
.wait_with_output()
.await
.map_err(|_| FormatterError::WaitForOutputFailed)?;
if !output.status.success() {
if !output.stderr.is_empty() {
let err = String::from_utf8_lossy(&output.stderr).to_string();
log::error!("Formatter error: {}", err);
return Err(FormatterError::NonZeroExitStatus(Some(err)));
}
return Err(FormatterError::NonZeroExitStatus(None));
} else if !output.stderr.is_empty() {
log::debug!(
"Formatter printed to stderr: {}",
String::from_utf8_lossy(&output.stderr).to_string()
);
}
let str = std::str::from_utf8(&output.stdout)
.map_err(|_| FormatterError::InvalidUtf8Output)?;
Ok(helix_core::diff::compare_ropes(&text, &Rope::from(str)))
};
return Some(formatting_future.boxed());
};
let text = self.text.clone();
// finds first language server that supports formatting and then formats
let language_server = self
.language_servers_with_feature(LanguageServerFeature::Format)
.next()?;
let offset_encoding = language_server.offset_encoding();
let request = language_server.text_document_formatting(
self.identifier(),
lsp::FormattingOptions {
tab_size: self.tab_width() as u32,
insert_spaces: matches!(self.indent_style, IndentStyle::Spaces(_)),
..Default::default()
},
None,
)?;
let fut = async move {
let edits = request.await.unwrap_or_else(|e| {
log::warn!("LSP formatting failed: {}", e);
Default::default()
});
Ok(helix_lsp::util::generate_transaction_from_edits(
&text,
edits,
offset_encoding,
))
};
Some(fut.boxed())
}
pub fn save<P: Into<PathBuf>>(
&mut self,
path: Option<P>,
force: bool,
) -> Result<
impl Future<Output = Result<DocumentSavedEvent, anyhow::Error>> + 'static + Send,
anyhow::Error,
> {
let path = path.map(|path| path.into());
self.save_impl(path, force)
// futures_util::future::Ready<_>,
}
/// The `Document`'s text is encoded according to its encoding and written to the file located
/// at its `path()`.
fn save_impl(
&mut self,
path: Option<PathBuf>,
force: bool,
) -> Result<
impl Future<Output = Result<DocumentSavedEvent, anyhow::Error>> + 'static + Send,
anyhow::Error,
> {
log::debug!(
"submitting save of doc '{:?}'",
self.path().map(|path| path.to_string_lossy())
);
// we clone and move text + path into the future so that we asynchronously save the current
// state without blocking any further edits.
let text = self.text().clone();
let path = match path {
Some(path) => helix_stdx::path::canonicalize(path),
None => {
if self.path.is_none() {
bail!("Can't save with no path set!");
}
self.path.as_ref().unwrap().clone()
}
};
let identifier = self.path().map(|_| self.identifier());
let language_servers = self.language_servers.clone();
// mark changes up to now as saved
let current_rev = self.get_current_revision();
let doc_id = self.id();
let encoding_with_bom_info = (self.encoding, self.has_bom);
let last_saved_time = self.last_saved_time;
// We encode the file according to the `Document`'s encoding.
let future = async move {
use tokio::fs;
if let Some(parent) = path.parent() {
// TODO: display a prompt asking the user if the directories should be created
if !parent.exists() {
if force {
std::fs::DirBuilder::new().recursive(true).create(parent)?;
} else {
bail!("can't save file, parent directory does not exist (use :w! to create it)");
}
}
}
// Protect against overwriting changes made externally
if !force {
if let Ok(metadata) = fs::metadata(&path).await {
if let Ok(mtime) = metadata.modified() {
if last_saved_time < mtime {
bail!("file modified by an external process, use :w! to overwrite");
}
}
}
}
let write_path = tokio::fs::read_link(&path)
.await
.unwrap_or_else(|_| path.clone());
if readonly(&write_path) {
bail!(std::io::Error::new(
std::io::ErrorKind::PermissionDenied,
"Path is read only"
));
}
let backup = if path.exists() {
let path_ = write_path.clone();
// hacks: we use tempfile to handle the complex task of creating
// non clobbered temporary path for us we don't want
// the whole automatically delete path on drop thing
// since the path doesn't exist yet, we just want
// the path
tokio::task::spawn_blocking(move || -> Option<PathBuf> {
tempfile::Builder::new()
.prefix(path_.file_name()?)
.suffix(".bck")
.make_in(path_.parent()?, |backup| std::fs::rename(&path_, backup))
.ok()?
.into_temp_path()
.keep()
.ok()
})
.await
.ok()
.flatten()
} else {
None
};
let write_result: anyhow::Result<_> = async {
let mut dst = tokio::fs::File::create(&write_path).await?;
to_writer(&mut dst, encoding_with_bom_info, &text).await?;
Ok(())
}
.await;
if let Some(backup) = backup {
if write_result.is_err() {
// restore backup
let _ = tokio::fs::rename(&backup, &write_path)
.await
.map_err(|e| log::error!("Failed to restore backup on write failure: {e}"));
} else {
// copy metadata and delete backup
let _ = tokio::task::spawn_blocking(move || {
let _ = copy_metadata(&backup, &write_path)
.map_err(|e| log::error!("Failed to copy metadata on write: {e}"));
let _ = std::fs::remove_file(backup)
.map_err(|e| log::error!("Failed to remove backup file on write: {e}"));
})
.await;
}
}
write_result?;
let event = DocumentSavedEvent {
revision: current_rev,
doc_id,
path,
text: text.clone(),
};
for (_, language_server) in language_servers {
if !language_server.is_initialized() {
continue;
}
if let Some(notification) = identifier
.clone()
.and_then(|id| language_server.text_document_did_save(id, &text))
{
if let Err(err) = notification.await {
log::error!("Failed to send textDocument/didSave: {err}");
}
}
}
Ok(event)
};
Ok(future)
}
/// Detect the programming language based on the file type.
pub fn detect_language(&mut self, config_loader: Arc<ArcSwap<syntax::Loader>>) {
let loader = config_loader.load();
self.set_language(
self.detect_language_config(&loader),
Some(Arc::clone(&config_loader)),
);
}
/// Detect the programming language based on the file type.
pub fn detect_language_config(
&self,
config_loader: &syntax::Loader,
) -> Option<Arc<helix_core::syntax::LanguageConfiguration>> {
config_loader
.language_config_for_file_name(self.path.as_ref()?)
.or_else(|| config_loader.language_config_for_shebang(self.text().slice(..)))