ok/jj
1
0
Fork 0
forked from mirrors/jj
Code Pull requests Activity

diff: extract helpers that process word-level diffs without interleaving

Browse source 
These functions will be reused by non-inline version of color-words diffs.
This commit is contained in:
Yuya Nishihara 2024-08-14 17:47:11 +09:00
parent decd913cea
commit 6b2e191b2b
1 changed files with 32 additions and 19 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

51
cli/src/diff_util.rs
View file

@ -12,6 +12,7 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::borrow::Borrow;
use std::cmp::max;
use std::collections::HashSet;
use std::ops::Range;
@ -1061,8 +1062,8 @@ fn unified_diff_hunks<'content>(
current_hunk.extend_context_lines(before_lines.into_iter().rev());
}
DiffHunk::Different(contents) => {
let [left, right] = contents.try_into().unwrap();
let (left_lines, right_lines) = inline_diff_hunks(left, right);
let (left_lines, right_lines) =
unzip_diff_hunks_to_lines(Diff::by_word(contents).hunks());
current_hunk.extend_removed_lines(left_lines);
current_hunk.extend_added_lines(right_lines);
}
@ -1074,19 +1075,21 @@ fn unified_diff_hunks<'content>(
hunks
}
/// Splits line-level hunks into word-level tokens. Returns lists of tokens per
/// line.
fn inline_diff_hunks<'content>(
left_content: &'content [u8],
right_content: &'content [u8],
) -> (Vec<DiffTokenVec<'content>>, Vec<DiffTokenVec<'content>>) {
/// Splits `(left, right)` hunk pairs into `(left_lines, right_lines)`.
fn unzip_diff_hunks_to_lines<'content, I>(
diff_hunks: I,
) -> (Vec<DiffTokenVec<'content>>, Vec<DiffTokenVec<'content>>)
where
I: IntoIterator,
I::Item: Borrow<DiffHunk<'content>>,
{
let mut left_lines: Vec<DiffTokenVec<'content>> = vec![];
let mut right_lines: Vec<DiffTokenVec<'content>> = vec![];
let mut left_tokens: DiffTokenVec<'content> = vec![];
let mut right_tokens: DiffTokenVec<'content> = vec![];
for hunk in Diff::by_word([left_content, right_content]).hunks() {
match hunk {
for hunk in diff_hunks {
match hunk.borrow() {
DiffHunk::Matching(content) => {
for token in content.split_inclusive(|b| *b == b'\n') {
left_tokens.push((DiffTokenType::Matching, token));
@ -1098,7 +1101,9 @@ fn inline_diff_hunks<'content>(
}
}
DiffHunk::Different(contents) => {
let [left, right] = contents.try_into().unwrap();
let [left, right] = contents[..]
.try_into()
.expect("hunk should have exactly two inputs");
for token in left.split_inclusive(|b| *b == b'\n') {
left_tokens.push((DiffTokenType::Different, token));
if token.ends_with(b"\n") {
@ -1147,14 +1152,7 @@ fn show_unified_diff_hunks(
};
formatter.with_label(label, |formatter| {
write!(formatter, "{sigil}")?;
for (token_type, content) in tokens {
match token_type {
DiffTokenType::Matching => formatter.write_all(content)?,
DiffTokenType::Different => formatter
.with_label("token", |formatter| formatter.write_all(content))?,
}
}
io::Result::Ok(())
show_diff_line_tokens(formatter, tokens)
})?;
let (_, content) = tokens.last().expect("hunk line must not be empty");
if !content.ends_with(b"\n") {
@ -1165,6 +1163,21 @@ fn show_unified_diff_hunks(
Ok(())
}
fn show_diff_line_tokens(
formatter: &mut dyn Formatter,
tokens: &[(DiffTokenType, &[u8])],
) -> io::Result<()> {
for (token_type, content) in tokens {
match token_type {
DiffTokenType::Matching => formatter.write_all(content)?,
DiffTokenType::Different => {
formatter.with_label("token", |formatter| formatter.write_all(content))?
}
}
}
Ok(())
}
pub fn show_git_diff(
formatter: &mut dyn Formatter,
store: &Store,