From 668709a33f3acd4aa63448ce512c98cb48d9792e Mon Sep 17 00:00:00 2001
From: Gusted <postmaster@gusted.xyz>
Date: Thu, 11 Apr 2024 13:34:53 +0200
Subject: [PATCH] [BUG] Handle bigger files in `git grep`

- The parser of `git grep`'s output uses `bufio.Scanner`, which is a good
choice overall, however it does have a limit that's usually not noticed,
it will not read more than `64 * 1024` bytes at once which can be hit in
practical scenarios.
- Use `bufio.Reader` instead which doesn't have this limitation, but is
a bit harder to work with as it's a more lower level primitive.
- Adds unit test.
- Resolves https://codeberg.org/forgejo/forgejo/issues/3149
---
 modules/git/grep.go      | 20 ++++++++++++++++----
 modules/git/grep_test.go | 27 +++++++++++++++++++++++++++
 2 files changed, 43 insertions(+), 4 deletions(-)

diff --git a/modules/git/grep.go b/modules/git/grep.go
index ee6a858f7..1de739107 100644
--- a/modules/git/grep.go
+++ b/modules/git/grep.go
@@ -10,6 +10,7 @@ import (
 	"context"
 	"errors"
 	"fmt"
+	"io"
 	"os"
 	"strconv"
 	"strings"
@@ -80,10 +81,21 @@ func GrepSearch(ctx context.Context, repo *Repository, search string, opts GrepO
 			defer stdoutReader.Close()
 
 			isInBlock := false
-			scanner := bufio.NewScanner(stdoutReader)
+			scanner := bufio.NewReader(stdoutReader)
 			var res *GrepResult
-			for scanner.Scan() {
-				line := scanner.Text()
+			for {
+				line, err := scanner.ReadString('\n')
+				if err != nil {
+					if err == io.EOF {
+						return nil
+					}
+					return err
+				}
+				// Remove delimiter.
+				if len(line) > 0 {
+					line = line[:len(line)-1]
+				}
+
 				if !isInBlock {
 					if _ /* ref */, filename, ok := strings.Cut(line, ":"); ok {
 						isInBlock = true
@@ -109,7 +121,7 @@ func GrepSearch(ctx context.Context, repo *Repository, search string, opts GrepO
 					res.LineCodes = append(res.LineCodes, lineCode)
 				}
 			}
-			return scanner.Err()
+			return nil
 		},
 	})
 	// git grep exits by cancel (killed), usually it is caused by the limit of results
diff --git a/modules/git/grep_test.go b/modules/git/grep_test.go
index b5fa437c5..a321b1145 100644
--- a/modules/git/grep_test.go
+++ b/modules/git/grep_test.go
@@ -4,7 +4,10 @@
 package git
 
 import (
+	"bytes"
 	"context"
+	"os"
+	"path"
 	"path/filepath"
 	"testing"
 
@@ -49,3 +52,27 @@ func TestGrepSearch(t *testing.T) {
 	assert.Error(t, err)
 	assert.Len(t, res, 0)
 }
+
+func TestGrepLongFiles(t *testing.T) {
+	tmpDir := t.TempDir()
+
+	err := InitRepository(DefaultContext, tmpDir, false, Sha1ObjectFormat.Name())
+	assert.NoError(t, err)
+
+	gitRepo, err := openRepositoryWithDefaultContext(tmpDir)
+	assert.NoError(t, err)
+	defer gitRepo.Close()
+
+	assert.NoError(t, os.WriteFile(path.Join(tmpDir, "README.md"), bytes.Repeat([]byte{'a'}, 65*1024), 0o666))
+
+	err = AddChanges(tmpDir, true)
+	assert.NoError(t, err)
+
+	err = CommitChanges(tmpDir, CommitChangesOptions{Message: "Long file"})
+	assert.NoError(t, err)
+
+	res, err := GrepSearch(context.Background(), gitRepo, "a", GrepOptions{})
+	assert.NoError(t, err)
+	assert.Len(t, res, 1)
+	assert.Len(t, res[0].LineCodes[0], 65*1024)
+}