From e52c902d3ad77e15bc32bc26503274f3b5ff9853 Mon Sep 17 00:00:00 2001
From: Martin von Zweigbergk <martinvonz@google.com>
Date: Sat, 9 Oct 2021 23:41:17 -0700
Subject: [PATCH] diff: compact adjacent unchanged regions also when using
 `for_tokenizer()`

I noticed while working on support for unified diffs (#33) that
`Diff::for_tokenizer(..., &find_line_ranges)` would return a
`DiffHunk::Matching` for each matching line instead of a single
`DiffHunk::Matching` for all the matching lines. That's different from
what you get from `Diff::default_refinement()` and seems less
convenient to work with.
---
 lib/src/diff.rs | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/lib/src/diff.rs b/lib/src/diff.rs
index 72abc13be..c4d59c4c3 100644
--- a/lib/src/diff.rs
+++ b/lib/src/diff.rs
@@ -392,11 +392,13 @@ impl<'input> Diff<'input> {
             .collect_vec();
         unchanged_regions.insert(BaseRange(base_input.len()..base_input.len()), offsets);
 
-        Self {
+        let mut diff = Self {
             base_input,
             other_inputs,
             unchanged_regions,
-        }
+        };
+        diff.compact_unchanged_regions();
+        diff
     }
 
     pub fn unrefined(inputs: &[&'input [u8]]) -> Self {
@@ -831,6 +833,23 @@ mod tests {
         );
     }
 
+    #[test]
+    fn test_diff_for_tokenizer_compacted() {
+        // Tests that unchanged regions are compacted when using for_tokenizer()
+        let diff = Diff::for_tokenizer(
+            &[b"a\nb\nc\nd\ne\nf\ng", b"a\nb\nc\nX\ne\nf\ng"],
+            &find_line_ranges,
+        );
+        assert_eq!(
+            diff.hunks().collect_vec(),
+            vec![
+                DiffHunk::Matching(b"a\nb\nc\n"),
+                DiffHunk::Different(vec![b"d\n", b"X\n"]),
+                DiffHunk::Matching(b"e\nf\ng"),
+            ]
+        );
+    }
+
     #[test]
     fn test_diff_nothing_in_common() {
         assert_eq!(