Switch to wu-diff for textual diffing

In #153 a user reported difftastic never terminated on a 140,000
file. This was due to the diff crate using a very large amount of time
and memory.

The diff crate does not use Myers' algorithm, which has a
divide-and-conquer approach using snakes:

https://blog.jcoglan.com/2017/03/22/myers-diff-in-linear-space-theory/

wu-diff does implement Myer's algorithm and performs much better on
these large files.

CHANGELOG.md

@@ -1,5 +1,9 @@
 ## 0.23 (unreleased)
 
+### Diffing
+
+Improved performance on very large files that are compared by text.
+
 ## 0.22 (release 10th March 2022)
 
 Difftastic now requires Rust 1.56 to build.

Cargo.lock

@@ -187,6 +187,7 @@ dependencies = [
  "tree-sitter",
  "typed-arena",
  "walkdir",
+ "wu-diff",
 ]
 
 [[package]]
@@ -574,3 +575,9 @@ name = "winapi-x86_64-pc-windows-gnu"
 version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
+
+[[package]]
+name = "wu-diff"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8e3e6735fcde06432870db8dc9d7e3ab1b93727c14eaef329969426299f28893"

Cargo.toml

@@ -41,6 +41,7 @@ term_size = "0.3.2"
 const_format = "0.2.22"
 owo-colors = "3.2.0"
 rpds = "0.11.0"
+wu-diff = "0.1.2"
 
 [dev-dependencies]
 pretty_assertions = "1.0.0"

src/line_parser.rs

@@ -109,15 +109,25 @@ fn diff_slice_by_hash<'a, T: Eq + Hash>(lhs: &'a [T], rhs: &'a [T]) -> Vec<diff:
         rhs_ids.push(id);
     }
 
-    diff::slice(&lhs_ids[..], &rhs_ids[..])
+    wu_diff::diff(&lhs_ids[..], &rhs_ids[..])
         .into_iter()
         .map(|result| match result {
-            diff::Result::Left(id) => diff::Result::Left(*id_values.get(id).unwrap()),
-            diff::Result::Both(lhs_id, rhs_id) => diff::Result::Both(
-                *id_values.get(lhs_id).unwrap(),
-                *id_values.get(rhs_id).unwrap(),
-            ),
-            diff::Result::Right(id) => diff::Result::Right(*id_values.get(id).unwrap()),
+            wu_diff::DiffResult::Removed(r) => {
+                let id = lhs_ids[r.old_index.unwrap()];
+                diff::Result::Left(*id_values.get(&id).unwrap())
+            }
+            wu_diff::DiffResult::Common(c) => {
+                let lhs_id = lhs_ids[c.old_index.unwrap()];
+                let rhs_id = rhs_ids[c.new_index.unwrap()];
+                diff::Result::Both(
+                    *id_values.get(&lhs_id).unwrap(),
+                    *id_values.get(&rhs_id).unwrap(),
+                )
+            }
+            wu_diff::DiffResult::Added(a) => {
+                let id = rhs_ids[a.new_index.unwrap()];
+                diff::Result::Right(*id_values.get(&id).unwrap())
+            }
         })
         .collect::<Vec<diff::Result<&'a T>>>()
 }