gix_imara_diff/postprocess.rs
1// Modified for gitoxide from the upstream imara-diff crate.
2// Upstream source: git cat-file -p 32d1e45d3df061e6ccba6db7fdce92db29e345d8:src/postprocess.rs
3
4use crate::intern::{InternedInput, Token};
5use crate::slider_heuristic::SliderHeuristic;
6use crate::util::{find_hunk_end, find_hunk_start};
7use crate::{Diff, Hunk};
8
9impl Diff {
10 /// Postprocesses the diff using explicit token sequences and a custom heuristic.
11 ///
12 /// This is a lower-level method that works directly with token sequences rather than
13 /// an `InternedInput`. Use [`postprocess_with_heuristic`](Self::postprocess_with_heuristic)
14 /// for a more convenient API.
15 ///
16 /// # Parameters
17 ///
18 /// * `before` - The token sequence from the first file, before changes
19 /// * `after` - The token sequence from the second file, after changes
20 /// * `heuristic` - The slider heuristic to use for positioning hunks
21 pub fn postprocess_with(&mut self, before: &[Token], after: &[Token], mut heuristic: impl SliderHeuristic) {
22 Postprocessor {
23 added: &mut self.added,
24 removed: &mut self.removed,
25 tokens: after,
26 hunk: Hunk {
27 before: 0..0,
28 after: 0..0,
29 },
30 heuristic: &mut heuristic,
31 }
32 .run();
33 Postprocessor {
34 added: &mut self.removed,
35 removed: &mut self.added,
36 tokens: before,
37 hunk: Hunk {
38 before: 0..0,
39 after: 0..0,
40 },
41 heuristic: &mut heuristic,
42 }
43 .run()
44 }
45
46 /// Postprocesses the diff using an `InternedInput` and a custom heuristic.
47 ///
48 /// This is a convenience wrapper around [`postprocess_with`](Self::postprocess_with)
49 /// that extracts the token sequences from the input automatically.
50 ///
51 /// # Parameters
52 ///
53 /// * `input` - The interned input containing the token sequences
54 /// * `heuristic` - The slider heuristic to use for positioning hunks
55 pub fn postprocess_with_heuristic<T>(&mut self, input: &InternedInput<T>, heuristic: impl SliderHeuristic) {
56 self.postprocess_with(&input.before, &input.after, heuristic);
57 }
58}
59
60/// Internal state for postprocessing a diff to improve readability.
61struct Postprocessor<'a, H> {
62 /// The mutable array tracking which tokens were added.
63 added: &'a mut [bool],
64 /// The immutable array tracking which tokens were removed.
65 removed: &'a [bool],
66 /// The token sequence being processed.
67 tokens: &'a [Token],
68 /// The current hunk being processed in the iteration.
69 hunk: Hunk,
70 /// The heuristic used to determine optimal hunk positions.
71 heuristic: &'a mut H,
72}
73
74impl<H: SliderHeuristic> Postprocessor<'_, H> {
75 fn run(mut self) {
76 loop {
77 // find a hunk
78 if !self.hunk.next_hunk(self.removed, self.added) {
79 break;
80 }
81
82 let mut earliest_end;
83 let mut is_modification;
84 loop {
85 // move hunk up as far as possible to possibly merge it with other hunks
86 // and discover if there are other possible positions
87 while self.slide_up() {}
88 earliest_end = self.hunk.after.end;
89 is_modification = self.hunk.before.start != self.hunk.before.end;
90
91 let hunk_size_unexpanded = self.hunk.after.len();
92 // move hunk down as far as possible (and merge with other hunks it if
93 // possible) sliding down is often the most preferred position
94 while self.slide_down() {
95 is_modification |= self.hunk.before.start != self.hunk.before.end;
96 }
97 // if this hunk was merged with another hunk while sliding down we might
98 // be able to slide up more otherwise we are done
99 if hunk_size_unexpanded == self.hunk.after.len() {
100 break;
101 }
102 }
103
104 if self.hunk.after.end == earliest_end {
105 continue;
106 }
107 if is_modification {
108 // hunk can be moved and there is a removed hunk in the same region
109 // move the hunk so it align with the other hunk to produce a single
110 // MODIFIED hunk instead of two separate ADDED/REMOVED hunks
111 while self.hunk.before.start == self.hunk.before.end {
112 let success = self.slide_up();
113 debug_assert!(success);
114 }
115 } else {
116 let slider_end = self
117 .heuristic
118 .best_slider_end(self.tokens, self.hunk.after.clone(), earliest_end);
119 for _ in slider_end..self.hunk.after.end {
120 let success = self.slide_up();
121 debug_assert!(success);
122 }
123 }
124 }
125 }
126
127 /// Slides a hunk down by one token/line, potentially merging it with a subsequent hunk.
128 fn slide_down(&mut self) -> bool {
129 let Some(&next_token) = self.tokens.get(self.hunk.after.end as usize) else {
130 return false;
131 };
132 if self.tokens[self.hunk.after.start as usize] != next_token {
133 return false;
134 }
135 self.added[self.hunk.after.start as usize] = false;
136 self.added[self.hunk.after.end as usize] = true;
137 self.hunk.after.start += 1;
138 self.hunk.after.end = find_hunk_end(self.added, self.hunk.after.end);
139 // move the end of the remove range one down to keep the unchanged lines aligned
140 self.hunk.before.start = self.hunk.before.end + 1;
141 self.hunk.before.end = find_hunk_end(self.removed, self.hunk.before.start);
142 true
143 }
144
145 /// Slides a hunk up by one token/line, potentially merging it with a previous hunk.
146 fn slide_up(&mut self) -> bool {
147 if self.hunk.after.start == 0 {
148 return false;
149 }
150 if self.tokens[self.hunk.after.start as usize - 1] != self.tokens[self.hunk.after.end as usize - 1] {
151 return false;
152 }
153 self.added[self.hunk.after.start as usize - 1] = true;
154 self.added[self.hunk.after.end as usize - 1] = false;
155 self.hunk.after.end -= 1;
156 self.hunk.after.start = find_hunk_start(self.added, self.hunk.after.start - 1);
157 // move the start of the remove range one up to keep the unchanged lines aligned
158 self.hunk.before.end = self.hunk.before.start - 1;
159 self.hunk.before.start = find_hunk_start(self.removed, self.hunk.before.start - 1);
160 true
161 }
162}