helix

bendns' repos / helix

Unnamed repository; edit this file 'description' to name the repository.

lang-support.md

-rw-r--r--

3666

typable-cmd.md

-rw-r--r--

5741

ef='#n40'>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 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097

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::doc_formatter::TextFormat;
use helix_core::encoding::Encoding;
use helix_core::syntax::{Highlight, LanguageServerFeature};
use helix_core::text_annotations::{InlineAnnotation, TextAnnotations};
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::rc::Rc;
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::{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 text: Rope,
}

impl SavePoint {
    pub fn cursor(&self) -> usize {
        // we always create transactions with selections
        self.revert
            .lock()
            .selection()
            .unwrap()
            .primary()
            .cursor(self.text.slice(..))
    }
}

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>,
    /// 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: Rc<[InlineAnnotation]>,

    /// Inlay hints of `PARAMETER` kind, if any.
    pub parameter_inlay_hints: Rc<[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: Rc<[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: Rc<[InlineAnnotation]>,
    pub padding_after_inlay_hints: Rc<[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: Rc::new([]),
            parameter_inlay_hints: Rc::new([]),
            other_inlay_hints: Rc::new([]),
            padding_before_inlay_hints: Rc::new([]),
            padding_after_inlay_hints: Rc::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, 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,
        }
    }

    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<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((which::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_core::path::get_canonicalized_path(&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, fs::File};
            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 mut file = File::create(&path).await?;
            to_writer(&mut file, encoding_with_bom_info, &text).await?;

            let event = DocumentSavedEvent {
                revision: current_rev,
                doc_id,
                path,
                text: text.clone(),
            };

            for (_, language_server) in language_servers {
                if !language_server.is_initialized() {
                    return Ok(event);
                }
                if let Some(identifier) = &identifier {
                    if let Some(notification) =
                        language_server.text_document_did_save(identifier.clone(), &text)
                    {
                        notification.await?;
                    }
                }
            }

            Ok(event)
        };

        Ok(future)
    }

    /// Detect the programming language based on the file type.
    pub fn detect_language(&mut self, config_loader: Arc<syntax::Loader>) {
        self.set_language(
            self.detect_language_config(&config_loader),
            Some(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(..)))
    }

    /// Detect the indentation used in the file, or otherwise defaults to the language indentation
    /// configured in `languages.toml`, with a fallback to tabs if it isn't specified. Line ending
    /// is likewise auto-detected, and will remain unchanged if no line endings were detected.
    pub fn detect_indent_and_line_ending(&mut self) {
        self.indent_style = auto_detect_indent_style(&self.text).unwrap_or_else(|| {
            self.language_config()
                .and_then(|config| config.indent.as_ref())
                .map_or(DEFAULT_INDENT, |config| IndentStyle::from_str(&config.unit))
        });
        if let Some(line_ending) = auto_detect_line_ending(&self.text) {
            self.line_ending = line_ending;
        }
    }

    #[cfg(unix)]
    // Detect if the file is readonly and change the readonly field if necessary (unix only)
    pub fn detect_readonly(&mut self) {
        use rustix::fs::{access, Access};
        // Allows setting the flag for files the user cannot modify, like root files
        self.readonly = match &self.path {
            None => false,
            Some(p) => match access(p, Access::WRITE_OK) {
                Ok(_) => false,
                Err(err) if err.kind() == std::io::ErrorKind::NotFound => false,
                Err(_) => true,
            },
        };
    }

    #[cfg(not(unix))]
    // Detect if the file is readonly and change the readonly field if necessary (non-unix os)
    pub fn detect_readonly(&mut self) {
        // TODO Use the Windows' function `CreateFileW` to check if a file is readonly
        // Discussion: https://github.com/helix-editor/helix/pull/7740#issuecomment-1656806459
        // Vim implementation: https://github.com/vim/vim/blob/4c0089d696b8d1d5dc40568f25ea5738fa5bbffb/src/os_win32.c#L7665
        // Windows binding: https://microsoft.github.io/windows-docs-rs/doc/windows/Win32/Storage/FileSystem/fn.CreateFileW.html
        self.readonly = match &self.path {
            None => false,
            Some(p) => match std::fs::metadata(p) {
                Err(err) if err.kind() == std::io::ErrorKind::NotFound => false,
                Err(_) => false,
                Ok(metadata) => metadata.permissions().readonly(),
            },
        };
    }

    /// Reload the document from its path.
    pub fn reload(
        &mut self,
        view: &mut View,
        provider_registry: &DiffProviderRegistry,
    ) -> Result<(), Error> {
        let encoding = self.encoding;
        let path = self
            .path()
            .filter(|path| path.exists())
            .ok_or_else(|| anyhow!("can't find file to reload from {:?}", self.display_name()))?
            .to_owned();

        // Once we have a valid path we check if its readonly status has changed
        self.detect_readonly();

        let mut file = std::fs::File::open(&path)?;
        let (rope, ..) = from_reader(&mut file, Some(encoding))?;

        // Calculate the difference between the buffer and source text, and apply it.
        // This is not considered a modification of the contents of the file regardless
        // of the encoding.
        let transaction = helix_core::diff::compare_ropes(self.text(), &rope);
        self.apply(&transaction, view.id);
        self.append_changes_to_history(view);
        self.reset_modified();

        self.last_saved_time = SystemTime::now();

        self.detect_indent_and_line_ending();

        match provider_registry.get_diff_base(&path) {
            Some(diff_base) => self.set_diff_base(diff_base),
            None => self.diff_handle = None,
        }

        self.version_control_head = provider_registry.get_current_head_name(&path);

        Ok(())
    }

    /// Sets the [`Document`]'s encoding with the encoding correspondent to `label`.
    pub fn set_encoding(&mut self, label: &str) -> Result<(), Error> {
        let encoding =
            Encoding::for_label(label.as_bytes()).ok_or_else(|| anyhow!("unknown encoding"))?;

        self.encoding = encoding;

        Ok(())
    }

    /// Returns the [`Document`]'s current encoding.
    pub fn encoding(&self) -> &'static Encoding {
        self.encoding
    }

    pub fn set_path(&mut self, path: Option<&Path>) {
        let path = path.map(helix_core::path::get_canonicalized_path);

        // if parent doesn't exist we still want to open the document
        // and error out when document is saved
        self.path = path;

        self.detect_readonly();
    }

    /// Set the programming language for the file and load associated data (e.g. highlighting)
    /// if it exists.
    pub fn set_language(
        &mut self,
        language_config: Option<Arc<helix_core::syntax::LanguageConfiguration>>,
        loader: Option<Arc<helix_core::syntax::Loader>>,
    ) {
        if let (Some(language_config), Some(loader)) = (language_config, loader) {
            if let Some(highlight_config) = language_config.highlight_config(&loader.scopes()) {
                self.syntax = Syntax::new(self.text.slice(..), highlight_config, loader);
            }

            self.language = Some(language_config);
        } else {
            self.syntax = None;
            self.language = None;
        };
    }

    /// Set the programming language for the file if you know the name (scope) but don't have the
    /// [`syntax::LanguageConfiguration`] for it.
    pub fn set_language2(&mut self, scope: &str, config_loader: Arc<syntax::Loader>) {
        let language_config = config_loader.language_config_for_scope(scope);

        self.set_language(language_config, Some(config_loader));
    }

    /// Set the programming language for the file if you know the language but don't have the
    /// [`syntax::LanguageConfiguration`] for it.
    pub fn set_language_by_language_id(
        &mut self,
        language_id: &str,
        config_loader: Arc<syntax::Loader>,
    ) -> anyhow::Result<()> {
        let language_config = config_loader
            .language_config_for_language_id(language_id)
            .ok_or_else(|| anyhow!("invalid language id: {}", language_id))?;
        self.set_language(Some(language_config), Some(config_loader));
        Ok(())
    }

    /// Select text within the [`Document`].
    pub fn set_selection(&mut self, view_id: ViewId, selection: Selection) {
        // TODO: use a transaction?
        self.selections
            .insert(view_id, selection.ensure_invariants(self.text().slice(..)));
    }

    /// Find the origin selection of the text in a document, i.e. where
    /// a single cursor would go if it were on the first grapheme. If
    /// the text is empty, returns (0, 0).
    pub fn origin(&self) -> Range {
        if self.text().len_chars() == 0 {
            return Range::new(0, 0);
        }

        Range::new(0, 1).grapheme_aligned(self.text().slice(..))
    }

    /// Reset the view's selection on this document to the
    /// [origin](Document::origin) cursor.
    pub fn reset_selection(&mut self, view_id: ViewId) {
        let origin = self.origin();
        self.set_selection(view_id, Selection::single(origin.anchor, origin.head));
    }

    /// Initializes a new selection for the given view if it does not
    /// already have one.
    pub fn ensure_view_init(&mut self, view_id: ViewId) {
        if self.selections.get(&view_id).is_none() {
            self.reset_selection(view_id);
        }
    }

    /// Mark document as recent used for MRU sorting
    pub fn mark_as_focused(&mut self) {
        self.focused_at = std::time::Instant::now();
    }

    /// Remove a view's selection and inlay hints from this document.
    pub fn remove_view(&mut self, view_id: ViewId) {
        self.selections.remove(&view_id);
        self.inlay_hints.remove(&view_id);
    }

    /// Apply a [`Transaction`] to the [`Document`] to change its text.
    fn apply_impl(
        &mut self,
        transaction: &Transaction,
        view_id: ViewId,
        emit_lsp_notification: bool,
    ) -> bool {
        use helix_core::Assoc;

        let old_doc = self.text().clone();

        let success = transaction.changes().apply(&mut self.text);

        if success {
            for selection in self.selections.values_mut() {
                *selection = selection
                    .clone()
                    // Map through changes
                    .map(transaction.changes())
                    // Ensure all selections across all views still adhere to invariants.
                    .ensure_invariants(self.text.slice(..));
            }

            // if specified, the current selection should instead be replaced by transaction.selection
            if let Some(selection) = transaction.selection() {
                self.selections.insert(
                    view_id,
                    selection.clone().ensure_invariants(self.text.slice(..)),
                );
            }

            self.modified_since_accessed = true;
        }

        if !transaction.changes().is_empty() {
            self.version += 1;
            // start computing the diff in parallel
            if let Some(diff_handle) = &self.diff_handle {
                diff_handle.update_document(self.text.clone(), false);
            }

            // generate revert to savepoint
            if !self.savepoints.is_empty() {
                let revert = transaction.invert(&old_doc);
                self.savepoints
                    .retain_mut(|save_point| match save_point.upgrade() {
                        Some(savepoint) => {
                            let mut revert_to_savepoint = savepoint.revert.lock();
                            *revert_to_savepoint =
                                revert.clone().compose(mem::take(&mut revert_to_savepoint));
                            true
                        }
                        None => false,
                    })
            }

            // update tree-sitter syntax tree
            if let Some(syntax) = &mut self.syntax {
                // TODO: no unwrap
                let res = syntax.update(
                    old_doc.slice(..),
                    self.text.slice(..),
                    transaction.changes(),
                );
                if res.is_err() {
                    log::error!("TS parser failed, disabeling TS for the current buffer: {res:?}");
                    self.syntax = None;
                }
            }

            let changes = transaction.changes();

            changes.update_positions(
                self.diagnostics
                    .iter_mut()
                    .map(|diagnostic| (&mut diagnostic.range.start, Assoc::After)),
            );
            changes.update_positions(
                self.diagnostics
                    .iter_mut()
                    .map(|diagnostic| (&mut diagnostic.range.end, Assoc::After)),
            );
            // map state.diagnostics over changes::map_pos too
            for diagnostic in &mut self.diagnostics {
                diagnostic.line = self.text.char_to_line(diagnostic.range.start);
            }

            self.diagnostics
                .sort_unstable_by_key(|diagnostic| diagnostic.range);

            // Update the inlay hint annotations' positions, helping ensure they are displayed in the proper place
            let apply_inlay_hint_changes = |annotations: &mut Rc<[InlineAnnotation]>| {
                if let Some(data) = Rc::get_mut(annotations) {
                    changes.update_positions(
                        data.iter_mut()
                            .map(|annotation| (&mut annotation.char_idx, Assoc::After)),
                    );
                }
            };

            self.inlay_hints_oudated = true;
            for text_annotation in self.inlay_hints.values_mut() {
                let DocumentInlayHints {
                    id: _,
                    type_inlay_hints,
                    parameter_inlay_hints,
                    other_inlay_hints,
                    padding_before_inlay_hints,
                    padding_after_inlay_hints,
                } = text_annotation;

                apply_inlay_hint_changes(padding_before_inlay_hints);
                apply_inlay_hint_changes(type_inlay_hints);
                apply_inlay_hint_changes(parameter_inlay_hints);
                apply_inlay_hint_changes(other_inlay_hints);
                apply_inlay_hint_changes(padding_after_inlay_hints);
            }

            if emit_lsp_notification {
                // emit lsp notification
                for language_server in self.language_servers() {
                    let notify = language_server.text_document_did_change(
                        self.versioned_identifier(),
                        &old_doc,
                        self.text(),
                        changes,
                    );

                    if let Some(notify) = notify {
                        tokio::spawn(notify);
                    }
                }
            }
        }
        success
    }

    fn apply_inner(
        &mut self,
        transaction: &Transaction,
        view_id: ViewId,
        emit_lsp_notification: bool,
    ) -> bool {
        // store the state just before any changes are made. This allows us to undo to the
        // state just before a transaction was applied.
        if self.changes.is_empty() && !transaction.changes().is_empty() {
            self.old_state = Some(State {
                doc: self.text.clone(),
                selection: self.selection(view_id).clone(),
            });
        }

        let success = self.apply_impl(transaction, view_id, emit_lsp_notification);

        if !transaction.changes().is_empty() {
            // Compose this transaction with the previous one
            take_with(&mut self.changes, |changes| {
                changes.compose(transaction.changes().clone())
            });
        }
        success
    }
    /// Apply a [`Transaction`] to the [`Document`] to change its text.
    pub fn apply(&mut self, transaction: &Transaction, view_id: ViewId) -> bool {
        self.apply_inner(transaction, view_id, true)
    }

    /// Apply a [`Transaction`] to the [`Document`] to change its text
    /// without notifying the language servers. This is useful for temporary transactions
    /// that must not influence the server.
    pub fn apply_temporary(&mut self, transaction: &Transaction, view_id: ViewId) -> bool {
        self.apply_inner(transaction, view_id, false)
    }

    fn undo_redo_impl(&mut self, view: &mut View, undo: bool) -> bool {
        let mut history = self.history.take();
        let txn = if undo { history.undo() } else { history.redo() };
        let success = if let Some(txn) = txn {
            self.apply_impl(txn, view.id, true)
        } else {
            false
        };
        self.history.set(history);

        if success {
            // reset changeset to fix len
            self.changes = ChangeSet::new(self.text().slice(..));
            // Sync with changes with the jumplist selections.
            view.sync_changes(self);
        }
        success
    }

    /// Undo the last modification to the [`Document`]. Returns whether the undo was successful.
    pub fn undo(&mut self, view: &mut View) -> bool {
        self.undo_redo_impl(view, true)
    }

    /// Redo the last modification to the [`Document`]. Returns whether the redo was successful.
    pub fn redo(&mut self, view: &mut View) -> bool {
        self.undo_redo_impl(view, false)
    }

    /// Creates a reference counted snapshot (called savpepoint) of the document.
    ///
    /// The snapshot will remain valid (and updated) idenfinitly as long as ereferences to it exist.
    /// Restoring the snapshot will restore the selection and the contents of the document to
    /// the state it had when this function was called.
    pub fn savepoint(&mut self, view: &View) -> Arc<SavePoint> {
        let revert = Transaction::new(self.text()).with_selection(self.selection(view.id).clone());
        // check if there is already an existing (identical) savepoint around
        if let Some(savepoint) = self
            .savepoints
            .iter()
            .rev()
            .find_map(|savepoint| savepoint.upgrade())
        {
            let transaction = savepoint.revert.lock();
            if savepoint.view == view.id
                && transaction.changes().is_empty()
                && transaction.selection() == revert.selection()
            {
                drop(transaction);
                return savepoint;
            }
        }
        let savepoint = Arc::new(SavePoint {
            view: view.id,
            revert: Mutex::new(revert),
            text: self.text.clone(),
        });
        self.savepoints.push(Arc::downgrade(&savepoint));
        savepoint
    }

    pub fn restore(&mut self, view: &mut View, savepoint: &SavePoint, emit_lsp_notification: bool) {
        assert_eq!(
            savepoint.view, view.id,
            "Savepoint must not be used with a different view!"
        );
        // search and remove savepoint using a ptr comparison
        // this avoids a deadlock as we need to lock the mutex
        let savepoint_idx = self
            .savepoints
            .iter()
            .position(|savepoint_ref| savepoint_ref.as_ptr() == savepoint as *const _)
            .expect("Savepoint must belong to this document");

        let savepoint_ref = self.savepoints.remove(savepoint_idx);
        let mut revert = savepoint.revert.lock();
        self.apply_inner(&revert, view.id, emit_lsp_notification);
        *revert = Transaction::new(self.text()).with_selection(self.selection(view.id).clone());
        self.savepoints.push(savepoint_ref)
    }

    fn earlier_later_impl(&mut self, view: &mut View, uk: UndoKind, earlier: bool) -> bool {
        let txns = if earlier {
            self.history.get_mut().earlier(uk)
        } else {
            self.history.get_mut().later(uk)
        };
        let mut success = false;
        for txn in txns {
            if self.apply_impl(&txn, view.id, true) {
                success = true;
            }
        }
        if success {
            // reset changeset to fix len
            self.changes = ChangeSet::new(self.text().slice(..));
            // Sync with changes with the jumplist selections.
            view.sync_changes(self);
        }
        success
    }

    /// Undo modifications to the [`Document`] according to `uk`.
    pub fn earlier(&mut self, view: &mut View, uk: UndoKind) -> bool {
        self.earlier_later_impl(view, uk, true)
    }

    /// Redo modifications to the [`Document`] according to `uk`.
    pub fn later(&mut self, view: &mut View, uk: UndoKind) -> bool {
        self.earlier_later_impl(view, uk, false)
    }

    /// Commit pending changes to history
    pub fn append_changes_to_history(&mut self, view: &mut View) {
        if self.changes.is_empty() {
            return;
        }

        let new_changeset = ChangeSet::new(self.text().slice(..));
        let changes = std::mem::replace(&mut self.changes, new_changeset);
        // Instead of doing this messy merge we could always commit, and based on transaction
        // annotations either add a new layer or compose into the previous one.
        let transaction =
            Transaction::from(changes).with_selection(self.selection(view.id).clone());

        // HAXX: we need to reconstruct the state as it was before the changes..
        let old_state = self.old_state.take().expect("no old_state available");

        let mut history = self.history.take();
        history.commit_revision(&transaction, &old_state);
        self.history.set(history);

        // Update jumplist entries in the view.
        view.apply(&transaction, self);
    }

    pub fn id(&self) -> DocumentId {
        self.id
    }

    /// If there are unsaved modifications.
    pub fn is_modified(&self) -> bool {
        let history = self.history.take();
        let current_revision = history.current_revision();
        self.history.set(history);
        log::debug!(
            "id {} modified - last saved: {}, current: {}",
            self.id,
            self.last_saved_revision,
            current_revision
        );
        current_revision != self.last_saved_revision || !self.changes.is_empty()
    }

    /// Save modifications to history, and so [`Self::is_modified`] will return false.
    pub fn reset_modified(&mut self) {
        let history = self.history.take();
        let current_revision = history.current_revision();
        self.history.set(history);
        self.last_saved_revision = current_revision;
    }

    /// Set the document's latest saved revision to the given one.
    pub fn set_last_saved_revision(&mut self, rev: usize) {
        log::debug!(
            "doc {} revision updated {} -> {}",
            self.id,
            self.last_saved_revision,
            rev
        );
        self.last_saved_revision = rev;
        self.last_saved_time = SystemTime::now();
    }

    /// Get the document's latest saved revision.
    pub fn get_last_saved_revision(&mut self) -> usize {
        self.last_saved_revision
    }

    /// Get the current revision number
    pub fn get_current_revision(&mut self) -> usize {
        let history = self.history.take();
        let current_revision = history.current_revision();
        self.history.set(history);
        current_revision
    }

    /// Corresponding language scope name. Usually `source.<lang>`.
    pub fn language_scope(&self) -> Option<&str> {
        self.language
            .as_ref()
            .map(|language| language.scope.as_str())
    }

    /// Language name for the document. Corresponds to the `name` key in
    /// `languages.toml` configuration.
    pub fn language_name(&self) -> Option<&str> {
        self.language
            .as_ref()
            .map(|language| language.language_id.as_str())
    }

    /// Language ID for the document. Either the `language-id`,
    /// or the document language name if no `language-id` has been specified.
    pub fn language_id(&self) -> Option<&str> {
        self.language_config()?
            .language_server_language_id
            .as_deref()
            .or_else(|| self.language_name())
    }

    /// Corresponding [`LanguageConfiguration`].
    pub fn language_config(&self) -> Option<&LanguageConfiguration> {
        self.language.as_deref()
    }

    /// Current document version, incremented at each change.
    pub fn version(&self) -> i32 {
        self.version
    }

    /// maintains the order as configured in the language_servers TOML array
    pub fn language_servers(&self) -> impl Iterator<Item = &helix_lsp::Client> {
        self.language_config().into_iter().flat_map(move |config| {
            config.language_servers.iter().filter_map(move |features| {
                let ls = &**self.language_servers.get(&features.name)?;
                if ls.is_initialized() {
                    Some(ls)
                } else {
                    None
                }
            })
        })
    }

    pub fn remove_language_server_by_name(&mut self, name: &str) -> Option<Arc<Client>> {
        self.language_servers.remove(name)
    }

    pub fn language_servers_with_feature(
        &self,
        feature: LanguageServerFeature,
    ) -> impl Iterator<Item = &helix_lsp::Client> {
        self.language_config().into_iter().flat_map(move |config| {
            config.language_servers.iter().filter_map(move |features| {
                let ls = &**self.language_servers.get(&features.name)?;
                if ls.is_initialized()
                    && ls.supports_feature(feature)
                    && features.has_feature(feature)
                {
                    Some(ls)
                } else {
                    None
                }
            })
        })
    }

    pub fn supports_language_server(&self, id: usize) -> bool {
        self.language_servers().any(|l| l.id() == id)
    }

    pub fn diff_handle(&self) -> Option<&DiffHandle> {
        self.diff_handle.as_ref()
    }

    /// Intialize/updates the differ for this document with a new base.
    pub fn set_diff_base(&mut self, diff_base: Vec<u8>) {
        if let Ok((diff_base, ..)) = from_reader(&mut diff_base.as_slice(), Some(self.encoding)) {
            if let Some(differ) = &self.diff_handle {
                differ.update_diff_base(diff_base);
                return;
            }
            self.diff_handle = Some(DiffHandle::new(diff_base, self.text.clone()))
        } else {
            self.diff_handle = None;
        }
    }

    pub fn version_control_head(&self) -> Option<Arc<Box<str>>> {
        self.version_control_head.as_ref().map(|a| a.load_full())
    }

    pub fn set_version_control_head(
        &mut self,
        version_control_head: Option<Arc<ArcSwap<Box<str>>>>,
    ) {
        self.version_control_head = version_control_head;
    }

    #[inline]
    /// Tree-sitter AST tree
    pub fn syntax(&self) -> Option<&Syntax> {
        self.syntax.as_ref()
    }

    /// The width that the tab character is rendered at
    pub fn tab_width(&self) -> usize {
        self.language_config()
            .and_then(|config| config.indent.as_ref())
            .map_or(4, |config| config.tab_width) // fallback to 4 columns
    }

    // The width (in spaces) of a level of indentation.
    pub fn indent_width(&self) -> usize {
        self.indent_style.indent_width(self.tab_width())
    }

    pub fn changes(&self) -> &ChangeSet {
        &self.changes
    }

    #[inline]
    /// File path on disk.
    pub fn path(&self) -> Option<&PathBuf> {
        self.path.as_ref()
    }

    /// File path as a URL.
    pub fn url(&self) -> Option<Url> {
        Url::from_file_path(self.path()?).ok()
    }

    #[inline]
    pub fn text(&self) -> &Rope {
        &self.text
    }

    #[inline]
    pub fn selection(&self, view_id: ViewId) -> &Selection {
        &self.selections[&view_id]
    }

    #[inline]
    pub fn selections(&self) -> &HashMap<ViewId, Selection> {
        &self.selections
    }

    pub fn relative_path(&self) -> Option<PathBuf> {
        self.path
            .as_deref()
            .map(helix_core::path::get_relative_path)
    }

    pub fn display_name(&self) -> Cow<'static, str> {
        self.relative_path()
            .map(|path| path.to_string_lossy().to_string().into())
            .unwrap_or_else(|| SCRATCH_BUFFER_NAME.into())
    }

    // transact(Fn) ?

    // -- LSP methods

    #[inline]
    pub fn identifier(&self) -> lsp::TextDocumentIdentifier {
        lsp::TextDocumentIdentifier::new(self.url().unwrap())
    }

    pub fn versioned_identifier(&self) -> lsp::VersionedTextDocumentIdentifier {
        lsp::VersionedTextDocumentIdentifier::new(self.url().unwrap(), self.version)
    }

    pub fn position(
        &self,
        view_id: ViewId,
        offset_encoding: helix_lsp::OffsetEncoding,
    ) -> lsp::Position {
        let text = self.text();

        helix_lsp::util::pos_to_lsp_pos(
            text,
            self.selection(view_id).primary().cursor(text.slice(..)),
            offset_encoding,
        )
    }

    #[inline]
    pub fn diagnostics(&self) -> &[Diagnostic] {
        &self.diagnostics
    }

    pub fn shown_diagnostics(&self) -> impl Iterator<Item = &Diagnostic> + DoubleEndedIterator {
        self.diagnostics.iter().filter(|d| {
            self.language_servers_with_feature(LanguageServerFeature::Diagnostics)
                .any(|ls| ls.id() == d.language_server_id)
        })
    }

    pub fn replace_diagnostics(
        &mut self,
        mut diagnostics: Vec<Diagnostic>,
        language_server_id: usize,
    ) {
        self.clear_diagnostics(language_server_id);
        self.diagnostics.append(&mut diagnostics);
        self.diagnostics
            .sort_unstable_by_key(|diagnostic| diagnostic.range);
    }

    pub fn clear_diagnostics(&mut self, language_server_id: usize) {
        self.diagnostics
            .retain(|d| d.language_server_id != language_server_id);
    }

    /// Get the document's auto pairs. If the document has a recognized
    /// language config with auto pairs configured, returns that;
    /// otherwise, falls back to the global auto pairs config. If the global
    /// config is false, then ignore language settings.
    pub fn auto_pairs<'a>(&'a self, editor: &'a Editor) -> Option<&'a AutoPairs> {
        let global_config = (editor.auto_pairs).as_ref();

        // NOTE: If the user specifies the global auto pairs config as false, then
        //       we want to disable it globally regardless of language settings
        #[allow(clippy::question_mark)]
        {
            if global_config.is_none() {
                return None;
            }
        }

        match &self.language {
            Some(lang) => lang.as_ref().auto_pairs.as_ref().or(global_config),
            None => global_config,
        }
    }

    pub fn text_format(&self, mut viewport_width: u16, theme: Option<&Theme>) -> TextFormat {
        let config = self.config.load();
        let text_width = self
            .language_config()
            .and_then(|config| config.text_width)
            .unwrap_or(config.text_width);
        let soft_wrap_at_text_width = self
            .language_config()
            .and_then(|config| {
                config
                    .soft_wrap
                    .as_ref()
                    .and_then(|soft_wrap| soft_wrap.wrap_at_text_width)
            })
            .or(config.soft_wrap.wrap_at_text_width)
            .unwrap_or(false);
        if soft_wrap_at_text_width {
            // We increase max_line_len by 1 because softwrap considers the newline character
            // as part of the line length while the "typical" expectation is that this is not the case.
            // In particular other commands like :reflow do not count the line terminator.
            // This is technically inconsistent for the last line as that line never has a line terminator
            // but having the last visual line exceed the width by 1 seems like a rare edge case.
            viewport_width = viewport_width.min(text_width as u16 + 1)
        }
        let config = self.config.load();
        let editor_soft_wrap = &config.soft_wrap;
        let language_soft_wrap = self
            .language
            .as_ref()
            .and_then(|config| config.soft_wrap.as_ref());
        let enable_soft_wrap = language_soft_wrap
            .and_then(|soft_wrap| soft_wrap.enable)
            .or(editor_soft_wrap.enable)
            .unwrap_or(false);
        let max_wrap = language_soft_wrap
            .and_then(|soft_wrap| soft_wrap.max_wrap)
            .or(config.soft_wrap.max_wrap)
            .unwrap_or(20);
        let max_indent_retain = language_soft_wrap
            .and_then(|soft_wrap| soft_wrap.max_indent_retain)
            .or(editor_soft_wrap.max_indent_retain)
            .unwrap_or(40);
        let wrap_indicator = language_soft_wrap
            .and_then(|soft_wrap| soft_wrap.wrap_indicator.clone())
            .or_else(|| config.soft_wrap.wrap_indicator.clone())
            .unwrap_or_else(|| "↪ ".into());
        let tab_width = self.tab_width() as u16;
        TextFormat {
            soft_wrap: enable_soft_wrap && viewport_width > 10,
            tab_width,
            max_wrap: max_wrap.min(viewport_width / 4),
            max_indent_retain: max_indent_retain.min(viewport_width * 2 / 5),
            // avoid spinning forever when the window manager
            // sets the size to something tiny
            viewport_width,
            wrap_indicator: wrap_indicator.into_boxed_str(),
            wrap_indicator_highlight: theme
                .and_then(|theme| theme.find_scope_index("ui.virtual.wrap"))
                .map(Highlight),
        }
    }

    /// Get the text annotations that apply to the whole document, those that do not apply to any
    /// specific view.
    pub fn text_annotations(&self, _theme: Option<&Theme>) -> TextAnnotations {
        TextAnnotations::default()
    }

    /// Set the inlay hints for this document and `view_id`.
    pub fn set_inlay_hints(&mut self, view_id: ViewId, inlay_hints: DocumentInlayHints) {
        self.inlay_hints.insert(view_id, inlay_hints);
    }

    /// Get the inlay hints for this document and `view_id`.
    pub fn inlay_hints(&self, view_id: ViewId) -> Option<&DocumentInlayHints> {
        self.inlay_hints.get(&view_id)
    }

    /// Completely removes all the inlay hints saved for the document, dropping them to free memory
    /// (since it often means inlay hints have been fully deactivated).
    pub fn reset_all_inlay_hints(&mut self) {
        self.inlay_hints = Default::default();
    }
}

#[derive(Clone, Debug)]
pub enum FormatterError {
    SpawningFailed {
        command: String,
        error: std::io::ErrorKind,
    },
    BrokenStdin,
    WaitForOutputFailed,
    InvalidUtf8Output,
    DiskReloadError(String),
    NonZeroExitStatus(Option<String>),
}

impl std::error::Error for FormatterError {}

impl Display for FormatterError {
    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        match self {
            Self::SpawningFailed { command, error } => {
                write!(f, "Failed to spawn formatter {}: {:?}", command, error)
            }
            Self::BrokenStdin => write!(f, "Could not write to formatter stdin"),
            Self::WaitForOutputFailed => write!(f, "Waiting for formatter output failed"),
            Self::InvalidUtf8Output => write!(f, "Invalid UTF-8 formatter output"),
            Self::DiskReloadError(error) => write!(f, "Error reloading file from disk: {}", error),
            Self::NonZeroExitStatus(Some(output)) => write!(f, "Formatter error: {}", output),
            Self::NonZeroExitStatus(None) => {
                write!(f, "Formatter exited with non zero exit status")
            }
        }
    }
}

#[cfg(test)]
mod test {
    use arc_swap::ArcSwap;

    use super::*;

    #[test]
    fn changeset_to_changes_ignore_line_endings() {
        use helix_lsp::{lsp, Client, OffsetEncoding};
        let text = Rope::from("hello\r\nworld");
        let mut doc = Document::from(
            text,
            None,
            Arc::new(ArcSwap::new(Arc::new(Config::default()))),
        );
        let view = ViewId::default();
        doc.set_selection(view, Selection::single(0, 0));

        let transaction =
            Transaction::change(doc.text(), vec![(5, 7, Some("\n".into()))].into_iter());
        let old_doc = doc.text().clone();
        doc.apply(&transaction, view);
        let changes = Client::changeset_to_changes(
            &old_doc,
            doc.text(),
            transaction.changes(),
            OffsetEncoding::Utf8,
        );

        assert_eq!(doc.text(), "hello\nworld");

        assert_eq!(
            changes,
            &[lsp::TextDocumentContentChangeEvent {
                range: Some(lsp::Range::new(
                    lsp::Position::new(0, 5),
                    lsp::Position::new(1, 0)
                )),
                text: "\n".into(),
                range_length: None,
            }]
        );
    }

    #[test]
    fn changeset_to_changes() {
        use helix_lsp::{lsp, Client, OffsetEncoding};
        let text = Rope::from("hello");
        let mut doc = Document::from(
            text,
            None,
            Arc::new(ArcSwap::new(Arc::new(Config::default()))),
        );
        let view = ViewId::default();
        doc.set_selection(view, Selection::single(5, 5));

        // insert

        let transaction = Transaction::insert(doc.text(), doc.selection(view), " world".into());
        let old_doc = doc.text().clone();
        doc.apply(&transaction, view);
        let changes = Client::changeset_to_changes(
            &old_doc,
            doc.text(),
            transaction.changes(),
            OffsetEncoding::Utf8,
        );

        assert_eq!(
            changes,
            &[lsp::TextDocumentContentChangeEvent {
                range: Some(lsp::Range::new(
                    lsp::Position::new(0, 5),
                    lsp::Position::new(0, 5)
                )),
                text: " world".into(),
                range_length: None,
            }]
        );

        // delete

        let transaction = transaction.invert(&old_doc);
        let old_doc = doc.text().clone();
        doc.apply(&transaction, view);
        let changes = Client::changeset_to_changes(
            &old_doc,
            doc.text(),
            transaction.changes(),
            OffsetEncoding::Utf8,
        );

        // line: 0-based.
        // col: 0-based, gaps between chars.
        // 0 1 2 3 4 5 6 7 8 9 0 1
        // |h|e|l|l|o| |w|o|r|l|d|
        //           -------------
        // (0, 5)-(0, 11)
        assert_eq!(
            changes,
            &[lsp::TextDocumentContentChangeEvent {
                range: Some(lsp::Range::new(
                    lsp::Position::new(0, 5),
                    lsp::Position::new(0, 11)
                )),
                text: "".into(),
                range_length: None,
            }]
        );

        // replace

        // also tests that changes are layered, positions depend on previous changes.

        doc.set_selection(view, Selection::single(0, 5));
        let transaction = Transaction::change(
            doc.text(),
            vec![(0, 2, Some("aei".into())), (3, 5, Some("ou".into()))].into_iter(),
        );
        // aeilou
        let old_doc = doc.text().clone();
        doc.apply(&transaction, view);
        let changes = Client::changeset_to_changes(
            &old_doc,
            doc.text(),
            transaction.changes(),
            OffsetEncoding::Utf8,
        );

        assert_eq!(
            changes,
            &[
                // 0 1 2 3 4 5
                // |h|e|l|l|o|
                // ----
                //
                // aeillo
                lsp::TextDocumentContentChangeEvent {
                    range: Some(lsp::Range::new(
                        lsp::Position::new(0, 0),
                        lsp::Position::new(0, 2)
                    )),
                    text: "aei".into(),
                    range_length: None,
                },
                // 0 1 2 3 4 5 6
                // |a|e|i|l|l|o|
                //         -----
                //
                // aeilou
                lsp::TextDocumentContentChangeEvent {
                    range: Some(lsp::Range::new(
                        lsp::Position::new(0, 4),
                        lsp::Position::new(0, 6)
                    )),
                    text: "ou".into(),
                    range_length: None,
                }
            ]
        );
    }

    #[test]
    fn test_line_ending() {
        assert_eq!(
            Document::default(Arc::new(ArcSwap::new(Arc::new(Config::default()))))
                .text()
                .to_string(),
            helix_core::NATIVE_LINE_ENDING.as_str()
        );
    }

    macro_rules! decode {
        ($name:ident, $label:expr, $label_override:expr) => {
            #[test]
            fn $name() {
                let encoding = encoding::Encoding::for_label($label_override.as_bytes()).unwrap();
                let base_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")).join("tests/encoding");
                let path = base_path.join(format!("{}_in.txt", $label));
                let ref_path = base_path.join(format!("{}_in_ref.txt", $label));
                assert!(path.exists());
                assert!(ref_path.exists());

                let mut file = std::fs::File::open(path).unwrap();
                let text = from_reader(&mut file, Some(encoding.into()))
                    .unwrap()
                    .0
                    .to_string();
                let expectation = std::fs::read_to_string(ref_path).unwrap();
                assert_eq!(text[..], expectation[..]);
            }
        };
        ($name:ident, $label:expr) => {
            decode!($name, $label, $label);
        };
    }

    macro_rules! encode {
        ($name:ident, $label:expr, $label_override:expr) => {
            #[test]
            fn $name() {
                let encoding = encoding::Encoding::for_label($label_override.as_bytes()).unwrap();
                let base_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")).join("tests/encoding");
                let path = base_path.join(format!("{}_out.txt", $label));
                let ref_path = base_path.join(format!("{}_out_ref.txt", $label));
                assert!(path.exists());
                assert!(ref_path.exists());

                let text = Rope::from_str(&std::fs::read_to_string(path).unwrap());
                let mut buf: Vec<u8> = Vec::new();
                helix_lsp::block_on(to_writer(&mut buf, (encoding, false), &text)).unwrap();

                let expectation = std::fs::read(ref_path).unwrap();
                assert_eq!(buf, expectation);
            }
        };
        ($name:ident, $label:expr) => {
            encode!($name, $label, $label);
        };
    }

    decode!(big5_decode, "big5");
    encode!(big5_encode, "big5");
    decode!(euc_kr_decode, "euc_kr", "EUC-KR");
    encode!(euc_kr_encode, "euc_kr", "EUC-KR");
    decode!(gb18030_decode, "gb18030");
    encode!(gb18030_encode, "gb18030");
    decode!(iso_2022_jp_decode, "iso_2022_jp", "ISO-2022-JP");
    encode!(iso_2022_jp_encode, "iso_2022_jp", "ISO-2022-JP");
    decode!(jis0208_decode, "jis0208", "EUC-JP");
    encode!(jis0208_encode, "jis0208", "EUC-JP");
    decode!(jis0212_decode, "jis0212", "EUC-JP");
    decode!(shift_jis_decode, "shift_jis");
    encode!(shift_jis_encode, "shift_jis");
}