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
// Copyright 2020 The Jujutsu Authors
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// https://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#![allow(missing_docs)]
use std::collections::VecDeque;
use std::fmt::{Debug, Error, Formatter};
use crate::diff;
use crate::diff::{Diff, DiffHunk};
use crate::merge::{trivial_merge, Merge};
#[derive(PartialEq, Eq, Clone, Debug)]
pub struct DiffLine<'a> {
pub left_line_number: u32,
pub right_line_number: u32,
pub has_left_content: bool,
pub has_right_content: bool,
pub hunks: Vec<DiffHunk<'a>>,
}
impl DiffLine<'_> {
fn reset_line(&mut self) {
self.has_left_content = false;
self.has_right_content = false;
self.hunks.clear();
}
pub fn is_unmodified(&self) -> bool {
self.hunks
.iter()
.all(|hunk| matches!(hunk, DiffHunk::Matching(_)))
}
}
pub fn diff<'a>(left: &'a [u8], right: &'a [u8]) -> DiffLineIterator<'a> {
let diff_hunks = diff::diff(left, right);
DiffLineIterator::new(diff_hunks)
}
pub struct DiffLineIterator<'a> {
diff_hunks: Vec<DiffHunk<'a>>,
current_pos: usize,
current_line: DiffLine<'a>,
queued_lines: VecDeque<DiffLine<'a>>,
}
impl<'a> DiffLineIterator<'a> {
fn new(diff_hunks: Vec<DiffHunk<'a>>) -> Self {
let current_line = DiffLine {
left_line_number: 1,
right_line_number: 1,
has_left_content: false,
has_right_content: false,
hunks: vec![],
};
DiffLineIterator {
diff_hunks,
current_pos: 0,
current_line,
queued_lines: VecDeque::new(),
}
}
}
impl<'a> Iterator for DiffLineIterator<'a> {
type Item = DiffLine<'a>;
fn next(&mut self) -> Option<Self::Item> {
// TODO: Should we attempt to interpret as utf-8 and otherwise break only at
// newlines?
while self.current_pos < self.diff_hunks.len() && self.queued_lines.is_empty() {
let hunk = &self.diff_hunks[self.current_pos];
self.current_pos += 1;
match hunk {
diff::DiffHunk::Matching(text) => {
let lines = text.split_inclusive(|b| *b == b'\n');
for line in lines {
self.current_line.has_left_content = true;
self.current_line.has_right_content = true;
self.current_line.hunks.push(DiffHunk::matching(line));
if line.ends_with(b"\n") {
self.queued_lines.push_back(self.current_line.clone());
self.current_line.left_line_number += 1;
self.current_line.right_line_number += 1;
self.current_line.reset_line();
}
}
}
diff::DiffHunk::Different(contents) => {
let left_lines = contents[0].split_inclusive(|b| *b == b'\n');
for left_line in left_lines {
self.current_line.has_left_content = true;
self.current_line
.hunks
.push(DiffHunk::different([left_line, b""]));
if left_line.ends_with(b"\n") {
self.queued_lines.push_back(self.current_line.clone());
self.current_line.left_line_number += 1;
self.current_line.reset_line();
}
}
let right_lines = contents[1].split_inclusive(|b| *b == b'\n');
for right_line in right_lines {
self.current_line.has_right_content = true;
self.current_line
.hunks
.push(DiffHunk::different([b"", right_line]));
if right_line.ends_with(b"\n") {
self.queued_lines.push_back(self.current_line.clone());
self.current_line.right_line_number += 1;
self.current_line.reset_line();
}
}
}
}
}
if let Some(line) = self.queued_lines.pop_front() {
return Some(line);
}
if !self.current_line.hunks.is_empty() {
let line = self.current_line.clone();
self.current_line.reset_line();
return Some(line);
}
None
}
}
// TODO: Maybe ContentHunk can be replaced with BString?
#[derive(PartialEq, Eq, Clone)]
pub struct ContentHunk(pub Vec<u8>);
impl AsRef<[u8]> for ContentHunk {
fn as_ref(&self) -> &[u8] {
self.0.as_ref()
}
}
impl Debug for ContentHunk {
fn fmt(&self, f: &mut Formatter<'_>) -> Result<(), Error> {
String::from_utf8_lossy(&self.0).fmt(f)
}
}
#[derive(PartialEq, Eq, Clone, Debug)]
pub enum MergeResult {
Resolved(ContentHunk),
Conflict(Vec<Merge<ContentHunk>>),
}
pub fn merge<T: AsRef<[u8]>>(slices: &Merge<T>) -> MergeResult {
// TODO: Using the first remove as base (first in the inputs) is how it's
// usually done for 3-way conflicts. Are there better heuristics when there are
// more than 3 parts?
let num_diffs = slices.removes().len();
let diff_inputs = slices.removes().chain(slices.adds());
merge_hunks(&Diff::by_line(diff_inputs), num_diffs)
}
fn merge_hunks(diff: &Diff, num_diffs: usize) -> MergeResult {
let mut resolved_hunk = ContentHunk(vec![]);
let mut merge_hunks: Vec<Merge<ContentHunk>> = vec![];
for diff_hunk in diff.hunks() {
match diff_hunk {
DiffHunk::Matching(content) => {
resolved_hunk.0.extend_from_slice(content);
}
DiffHunk::Different(parts) => {
if let Some(resolved) = trivial_merge(&parts[..num_diffs], &parts[num_diffs..]) {
resolved_hunk.0.extend_from_slice(resolved);
} else {
if !resolved_hunk.0.is_empty() {
merge_hunks.push(Merge::resolved(resolved_hunk));
resolved_hunk = ContentHunk(vec![]);
}
merge_hunks.push(Merge::from_removes_adds(
parts[..num_diffs]
.iter()
.map(|part| ContentHunk(part.to_vec())),
parts[num_diffs..]
.iter()
.map(|part| ContentHunk(part.to_vec())),
));
}
}
}
}
if merge_hunks.is_empty() {
MergeResult::Resolved(resolved_hunk)
} else {
if !resolved_hunk.0.is_empty() {
merge_hunks.push(Merge::resolved(resolved_hunk));
}
MergeResult::Conflict(merge_hunks)
}
}
#[cfg(test)]
mod tests {
use super::*;
fn hunk(data: &[u8]) -> ContentHunk {
ContentHunk(data.to_vec())
}
fn merge(removes: &[&[u8]], adds: &[&[u8]]) -> MergeResult {
super::merge(&Merge::from_removes_adds(removes, adds))
}
#[test]
fn test_merge_single_hunk() {
// Unchanged and empty on all sides
assert_eq!(merge(&[b""], &[b"", b""]), MergeResult::Resolved(hunk(b"")));
// Unchanged on all sides
assert_eq!(
merge(&[b"a"], &[b"a", b"a"]),
MergeResult::Resolved(hunk(b"a"))
);
// One side removed, one side unchanged
assert_eq!(
merge(&[b"a\n"], &[b"", b"a\n"]),
MergeResult::Resolved(hunk(b""))
);
// One side unchanged, one side removed
assert_eq!(
merge(&[b"a\n"], &[b"a\n", b""]),
MergeResult::Resolved(hunk(b""))
);
// Both sides removed same line
assert_eq!(
merge(&[b"a\n"], &[b"", b""]),
MergeResult::Resolved(hunk(b""))
);
// One side modified, one side unchanged
assert_eq!(
merge(&[b"a"], &[b"a b", b"a"]),
MergeResult::Resolved(hunk(b"a b"))
);
// One side unchanged, one side modified
assert_eq!(
merge(&[b"a"], &[b"a", b"a b"]),
MergeResult::Resolved(hunk(b"a b"))
);
// All sides added same content
assert_eq!(
merge(&[b"", b""], &[b"a\n", b"a\n", b"a\n"]),
MergeResult::Resolved(hunk(b"a\n"))
);
// One side modified, two sides added
assert_eq!(
merge(&[b"a", b""], &[b"b", b"b", b"b"]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a"), hunk(b"")],
vec![hunk(b"b"), hunk(b"b"), hunk(b"b")]
)])
);
// All sides removed same content
assert_eq!(
merge(&[b"a\n", b"a\n", b"a\n"], &[b"", b"", b"", b""]),
MergeResult::Resolved(hunk(b""))
);
// One side modified, two sides removed
assert_eq!(
merge(&[b"a\n", b"a\n"], &[b"b\n", b"", b""]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a\n"), hunk(b"a\n")],
vec![hunk(b"b\n"), hunk(b""), hunk(b"")]
)])
);
// Three sides made the same change
assert_eq!(
merge(&[b"a", b"a"], &[b"b", b"b", b"b"]),
MergeResult::Resolved(hunk(b"b"))
);
// One side removed, one side modified
assert_eq!(
merge(&[b"a\n"], &[b"", b"b\n"]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a\n")],
vec![hunk(b""), hunk(b"b\n")]
)])
);
// One side modified, one side removed
assert_eq!(
merge(&[b"a\n"], &[b"b\n", b""]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a\n")],
vec![hunk(b"b\n"), hunk(b"")]
)])
);
// Two sides modified in different ways
assert_eq!(
merge(&[b"a"], &[b"b", b"c"]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a")],
vec![hunk(b"b"), hunk(b"c")]
)])
);
// Two of three sides don't change, third side changes
assert_eq!(
merge(&[b"a", b"a"], &[b"a", b"", b"a"]),
MergeResult::Resolved(hunk(b""))
);
// One side unchanged, two other sides make the same change
assert_eq!(
merge(&[b"a", b"a"], &[b"b", b"a", b"b"]),
MergeResult::Resolved(hunk(b"b"))
);
// One side unchanged, two other sides make the different change
assert_eq!(
merge(&[b"a", b"a"], &[b"b", b"a", b"c"]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a"), hunk(b"a")],
vec![hunk(b"b"), hunk(b"a"), hunk(b"c")]
)])
);
// Merge of an unresolved conflict and another branch, where the other branch
// undid the change from one of the inputs to the unresolved conflict in the
// first.
assert_eq!(
merge(&[b"a", b"b"], &[b"b", b"a", b"c"]),
MergeResult::Resolved(hunk(b"c"))
);
// Merge of an unresolved conflict and another branch.
assert_eq!(
merge(&[b"a", b"b"], &[b"c", b"d", b"e"]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a"), hunk(b"b")],
vec![hunk(b"c"), hunk(b"d"), hunk(b"e")]
)])
);
// Two sides made the same change, third side made a different change
assert_eq!(
merge(&[b"a", b"b"], &[b"c", b"c", b"c"]),
MergeResult::Conflict(vec![Merge::from_removes_adds(
vec![hunk(b"a"), hunk(b"b")],
vec![hunk(b"c"), hunk(b"c"), hunk(b"c")]
)])
);
}
#[test]
fn test_merge_multi_hunk() {
// Two sides left one line unchanged, and added conflicting additional lines
assert_eq!(
merge(&[b"a\n"], &[b"a\nb\n", b"a\nc\n"]),
MergeResult::Conflict(vec![
Merge::resolved(hunk(b"a\n")),
Merge::from_removes_adds(vec![hunk(b"")], vec![hunk(b"b\n"), hunk(b"c\n")])
])
);
// Two sides changed different lines: no conflict
assert_eq!(
merge(&[b"a\nb\nc\n"], &[b"a2\nb\nc\n", b"a\nb\nc2\n"]),
MergeResult::Resolved(hunk(b"a2\nb\nc2\n"))
);
// Conflict with non-conflicting lines around
assert_eq!(
merge(&[b"a\nb\nc\n"], &[b"a\nb1\nc\n", b"a\nb2\nc\n"]),
MergeResult::Conflict(vec![
Merge::resolved(hunk(b"a\n")),
Merge::from_removes_adds(vec![hunk(b"b\n")], vec![hunk(b"b1\n"), hunk(b"b2\n")]),
Merge::resolved(hunk(b"c\n"))
])
);
// One side changes a line and adds a block after. The other side just adds the
// same block. You might expect the last block would be deduplicated. However,
// the changes in the first side can be parsed as follows:
// ```
// a {
// - p
// + q
// +}
// +
// +b {
// + x
// }
// ```
// Therefore, the first side modifies the block `a { .. }`, and the second side
// adds `b { .. }`. Git and Mercurial both duplicate the block in the result.
assert_eq!(
merge(
&[b"\
a {
p
}
"],
&[
b"\
a {
q
}
b {
x
}
",
b"\
a {
p
}
b {
x
}
"
],
),
MergeResult::Resolved(hunk(
b"\
a {
q
}
b {
x
}
b {
x
}
"
))
);
}
}