Skip to content

Improve performance with very long diffs - #6121

Merged
stefanhaller merged 5 commits into
masterfrom
speed-up-diff-parsing
Oct 14, 2026
Merged

stefanhaller merged 5 commits into
masterfrom
speed-up-diff-parsing

Conversation

@stefanhaller

Copy link
Copy Markdown
Collaborator

Parsing long diffs was very slow, especially for commits that add or remove files with hundreds of thousands of lines. This caused very noticeable lags for certain operations such as scrolling down in such a diff using >, or bringing up the <ctrl-g> files menu.

stefanhaller and others added 5 commits October 14, 2026 08:41
With no diff renderer configured, opening the jump-to-file menu for the
diff of a commit that adds or deletes a file of 100,000 lines takes
more than 20 seconds, and so does picking a file from it. In such a
diff, moving the selection by change block takes as long for each key
press. All this time the UI is frozen.

To find out which line of which file each row of the main view shows,
lazygit parses the view's content as a diff. For every line of a file's
diff it asks the parsed patch for the line's number in the new and in
the old file, and each of these calls counts the lines of the hunk
before it. So parsing takes time quadratic in the length of a hunk, and
the diff of a new or deleted file is a single hunk. For a diff of
200,000 lines, one parse took about 10 seconds, and opening the menu
parses it twice. A diff renderer that states which line each row shows,
such as delta, doesn't go through this parse.

Number all lines of the patch in one pass instead. Parsing that diff
now takes about 40 milliseconds.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Several actions on the selected lines of a diff look up those lines in
the diff of their file by their line numbers: staging, unstaging and
discarding them, editing the hunk they are in, and adding them to a
custom patch. Showing which lines of a diff are in the custom patch
does the same lookup the other way round. For this, the change lines of
the file's diff are indexed by their line numbers.

The index asked the patch for the number of each change line separately,
and each of these calls counts the lines of the hunk before the line. So
for a new file of 100,000 lines, building the index took 2 seconds.
Number all lines of the patch in one pass instead. Building the index
for that file now takes 4 milliseconds.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Every query about the lines of the main view, such as moving the
selection to the next change block, finds out which line of which file
each row shows. For every row, this joined the path that the diff
states with the path of the worktree. In a diff of 300,000 lines, this
took more than half of the time of the query.

The lines of a diff come file by file, so remember the last path that
was made absolute and reuse it while the path stays the same. Do the
same in the other places that go through every line of a diff. A query
over that diff now takes about 27 milliseconds instead of 63.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Under a diff renderer that states which diff line each row shows, such
as delta, moving the selection by change block in a diff of 300,000
lines lags noticeably. One reason is that the queries about the lines
of the main view collect the records of every row from the row's
cells, going through every cell of every row. Under delta, a row has
about 130 cells.

Collect the records of a line once, when the write cursor moves on to
the next line, and take them from there. A line that is written to
again is read from its cells until it has been finished again.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Under a diff renderer that states which diff line each row shows, such
as delta, the queries about the lines of the main view read the rows by
those records alone. Still, they took the text of every row as well,
and for a diff of 300,000 lines this was most of the time each of them
took. Moving the selection by change block runs two of them for every
key press.

Take the records of the rows first, and their text only if no row
states a diff line.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
@stefanhaller
stefanhaller merged commit 883c893 into master Oct 14, 2026
14 checks passed
@stefanhaller
stefanhaller deleted the speed-up-diff-parsing branch October 14, 2026 07:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant