Use gix pipeline filter instead of manual crlf implementation (#9503)

Cargo.lock

@@ -448,6 +448,18 @@ dependencies = [
  "winapi",
 ]
 
+[[package]]
+name = "filetime"
+version = "0.2.23"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1ee447700ac8aa0b2f2bd7bc4462ad686ba06baa6727ac149a2d6277f0d240fd"
+dependencies = [
+ "cfg-if",
+ "libc",
+ "redox_syscall 0.4.1",
+ "windows-sys 0.52.0",
+]
+
 [[package]]
 name = "flate2"
 version = "1.0.27"
@@ -527,33 +539,41 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "31887c304d9a935f3e5494fb5d6a0106c34e965168ec0db9b457424eedd0c741"
 dependencies = [
  "gix-actor",
+ "gix-attributes",
+ "gix-command",
  "gix-commitgraph",
  "gix-config",
  "gix-date",
  "gix-diff",
  "gix-discover",
  "gix-features",
+ "gix-filter",
  "gix-fs",
  "gix-glob",
  "gix-hash",
  "gix-hashtable",
+ "gix-ignore",
+ "gix-index",
  "gix-lock",
  "gix-macros",
  "gix-object",
  "gix-odb",
  "gix-pack",
  "gix-path",
+ "gix-pathspec",
  "gix-ref",
  "gix-refspec",
  "gix-revision",
  "gix-revwalk",
  "gix-sec",
+ "gix-submodule",
  "gix-tempfile",
  "gix-trace",
  "gix-traverse",
  "gix-url",
  "gix-utils",
  "gix-validate",
+ "gix-worktree",
  "once_cell",
  "parking_lot",
  "smallvec",
@@ -574,6 +594,32 @@ dependencies = [
  "winnow 0.5.28",
 ]
 
+[[package]]
+name = "gix-attributes"
+version = "0.22.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "214ee3792e504ee1ce206b36dcafa4f328ca313d1e2ac0b41433d68ef4e14260"
+dependencies = [
+ "bstr",
+ "gix-glob",
+ "gix-path",
+ "gix-quote",
+ "gix-trace",
+ "kstring",
+ "smallvec",
+ "thiserror",
+ "unicode-bom",
+]
+
+[[package]]
+name = "gix-bitmap"
+version = "0.2.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "78b6cd0f246180034ddafac9b00a112f19178135b21eb031b3f79355891f7325"
+dependencies = [
+ "thiserror",
+]
+
 [[package]]
 name = "gix-chunk"
 version = "0.4.7"
@@ -583,6 +629,18 @@ dependencies = [
  "thiserror",
 ]
 
+[[package]]
+name = "gix-command"
+version = "0.3.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ce1ffc7db3fb50b7dae6ecd937a3527cb725f444614df2ad8988d81806f13f09"
+dependencies = [
+ "bstr",
+ "gix-path",
+ "gix-trace",
+ "shell-words",
+]
+
 [[package]]
 name = "gix-commitgraph"
 version = "0.24.0"
@@ -690,6 +748,27 @@ dependencies = [
  "walkdir",
 ]
 
+[[package]]
+name = "gix-filter"
+version = "0.9.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9240862840fb740d209422937195e129e4ed3da49af212383260134bea8f6c1a"
+dependencies = [
+ "bstr",
+ "encoding_rs",
+ "gix-attributes",
+ "gix-command",
+ "gix-hash",
+ "gix-object",
+ "gix-packetline-blocking",
+ "gix-path",
+ "gix-quote",
+ "gix-trace",
+ "gix-utils",
+ "smallvec",
+ "thiserror",
+]
+
 [[package]]
 name = "gix-fs"
 version = "0.10.0"
@@ -733,6 +812,44 @@ dependencies = [
  "parking_lot",
 ]
 
+[[package]]
+name = "gix-ignore"
+version = "0.11.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1f7069aaca4a05784c4cb44e392f0eaf627c6e57e05d3100c0e2386a37a682f0"
+dependencies = [
+ "bstr",
+ "gix-glob",
+ "gix-path",
+ "gix-trace",
+ "unicode-bom",
+]
+
+[[package]]
+name = "gix-index"
+version = "0.29.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1d7152181ba8f0a3addc5075dd612cea31fc3e252b29c8be8c45f4892bf87426"
+dependencies = [
+ "bitflags 2.4.2",
+ "bstr",
+ "btoi",
+ "filetime",
+ "gix-bitmap",
+ "gix-features",
+ "gix-fs",
+ "gix-hash",
+ "gix-lock",
+ "gix-object",
+ "gix-traverse",
+ "itoa",
+ "libc",
+ "memmap2",
+ "rustix",
+ "smallvec",
+ "thiserror",
+]
+
 [[package]]
 name = "gix-lock"
 version = "13.0.0"
@@ -814,6 +931,18 @@ dependencies = [
  "thiserror",
 ]
 
+[[package]]
+name = "gix-packetline-blocking"
+version = "0.17.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ca8ef6dd3ea50e26f3bf572e90c034d033c804d340cd1eb386392f184a9ba2f7"
+dependencies = [
+ "bstr",
+ "faster-hex",
+ "gix-trace",
+ "thiserror",
+]
+
 [[package]]
 name = "gix-path"
 version = "0.10.4"
@@ -827,6 +956,21 @@ dependencies = [
  "thiserror",
 ]
 
+[[package]]
+name = "gix-pathspec"
+version = "0.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9cbd49750edb26b0a691e5246fc635fa554d344da825cd20fa9ee0da9c1b761f"
+dependencies = [
+ "bitflags 2.4.2",
+ "bstr",
+ "gix-attributes",
+ "gix-config-value",
+ "gix-glob",
+ "gix-path",
+ "thiserror",
+]
+
 [[package]]
 name = "gix-quote"
 version = "0.4.10"
@@ -917,6 +1061,21 @@ dependencies = [
  "windows-sys 0.52.0",
 ]
 
+[[package]]
+name = "gix-submodule"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "73182f6c1f5ed1ed94ba16581ac62593d5e29cd1c028b2af618f836283b8f8d4"
+dependencies = [
+ "bstr",
+ "gix-config",
+ "gix-path",
+ "gix-pathspec",
+ "gix-refspec",
+ "gix-url",
+ "thiserror",
+]
+
 [[package]]
 name = "gix-tempfile"
 version = "13.0.0"
@@ -986,6 +1145,24 @@ dependencies = [
  "thiserror",
 ]
 
+[[package]]
+name = "gix-worktree"
+version = "0.30.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ca36bb3dc54038c66507dc75c4d8edbee2d6d5cc45227b4eb508ad13dd60a006"
+dependencies = [
+ "bstr",
+ "gix-attributes",
+ "gix-features",
+ "gix-fs",
+ "gix-glob",
+ "gix-hash",
+ "gix-ignore",
+ "gix-index",
+ "gix-object",
+ "gix-path",
+]
+
 [[package]]
 name = "globset"
 version = "0.4.14"
@@ -1408,6 +1585,15 @@ dependencies = [
  "wasm-bindgen",
 ]
 
+[[package]]
+name = "kstring"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ec3066350882a1cd6d950d055997f379ac37fd39f81cd4d8ed186032eb3c5747"
+dependencies = [
+ "static_assertions",
+]
+
 [[package]]
 name = "libc"
 version = "0.2.152"
@@ -1880,6 +2066,12 @@ version = "1.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ae1a47186c03a32177042e55dbc5fd5aee900b8e0069a8d70fba96a9375cd012"
 
+[[package]]
+name = "shell-words"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "24188a676b6ae68c3b2cb3a01be17fbf7240ce009799bb56d5b1409051e78fde"
+
 [[package]]
 name = "signal-hook"
 version = "0.3.17"

helix-vcs/Cargo.toml

@@ -19,7 +19,7 @@ tokio = { version = "1", features = ["rt", "rt-multi-thread", "time", "sync", "p
 parking_lot = "0.12"
 arc-swap = { version = "1.6.0" }
 
-gix = { version = "0.58.0", default-features = false , optional = true }
+gix = { version = "0.58.0", features = ["attributes"], default-features = false, optional = true }
 imara-diff = "0.1.5"
 anyhow = "1"
 

helix-vcs/src/git.rs

@@ -1,5 +1,7 @@
 use anyhow::{bail, Context, Result};
 use arc_swap::ArcSwap;
+use gix::filter::plumbing::driver::apply::Delay;
+use std::io::Read;
 use std::path::Path;
 use std::sync::Arc;
 
@@ -76,30 +78,22 @@ fn get_diff_base(&self, file: &Path) -> Result<Vec<u8>> {
         let file_oid = find_file_in_commit(&repo, &head, file)?;
 
         let file_object = repo.find_object(file_oid)?;
-        let mut data = file_object.detach().data;
+        let data = file_object.detach().data;
-        // convert LF to CRLF if configured to avoid showing every line as changed
+        // Get the actual data that git would make out of the git object.
-        if repo
+        // This will apply the user's git config or attributes like crlf conversions.
-            .config_snapshot()
+        if let Some(work_dir) = repo.work_dir() {
-            .boolean("core.autocrlf")
+            let rela_path = file.strip_prefix(work_dir)?;
-            .unwrap_or(false)
+            let rela_path = gix::path::try_into_bstr(rela_path)?;
-        {
+            let (mut pipeline, _) = repo.filter_pipeline(None)?;
-            let mut normalized_file = Vec::with_capacity(data.len());
+            let mut worktree_outcome =
-            let mut at_cr = false;
+                pipeline.convert_to_worktree(&data, rela_path.as_ref(), Delay::Forbid)?;
-            for &byte in &data {
+            let mut buf = Vec::with_capacity(data.len());
-                if byte == b'\n' {
+            worktree_outcome.read_to_end(&mut buf)?;
-                    // if this is a LF instead of a CRLF (last byte was not a CR)
+            Ok(buf)
-                    // insert a new CR to generate a CRLF
+        } else {
-                    if !at_cr {
-                        normalized_file.push(b'\r');
-                    }
-                }
-                at_cr = byte == b'\r';
-                normalized_file.push(byte)
-            }
-            data = normalized_file
-        }
             Ok(data)
         }
+    }
 
     fn get_current_head_name(&self, file: &Path) -> Result<Arc<ArcSwap<Box<str>>>> {
         debug_assert!(!file.exists() || file.is_file());