From f4a7bf6d2a75a4d5932bdaa6a9a3932bf1687ba5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rados=C5=82aw=20Piliszek?= <radek@piliszek.it>
Date: Sun, 11 Aug 2024 14:24:40 +0200
Subject: [PATCH] git-grep: skip binary files

It is a waste of resources to scan them looking for matches
because they are never returned back - they appear as empty
lines in the current format.

Notably, even if they were returned, it is unlikely that matching
in binary files makes sense when the goal is "code search".
---
 modules/git/grep.go      |  3 ++-
 modules/git/grep_test.go | 25 +++++++++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletion(-)

diff --git a/modules/git/grep.go b/modules/git/grep.go
index ba870e0541..6449b465b9 100644
--- a/modules/git/grep.go
+++ b/modules/git/grep.go
@@ -63,8 +63,9 @@ func GrepSearch(ctx context.Context, repo *Repository, search string, opts GrepO
 	 2^@10^@repo: go-gitea/gitea
 	*/
 	var results []*GrepResult
+	// -I skips binary files
 	cmd := NewCommand(ctx, "grep",
-		"--null", "--break", "--heading", "--column",
+		"-I", "--null", "--break", "--heading", "--column",
 		"--fixed-strings", "--line-number", "--ignore-case", "--full-name")
 	cmd.AddOptionValues("--context", fmt.Sprint(opts.ContextLineNumber))
 	if opts.MatchesPerFile > 0 {
diff --git a/modules/git/grep_test.go b/modules/git/grep_test.go
index bb7db7d58d..5d0d343ac4 100644
--- a/modules/git/grep_test.go
+++ b/modules/git/grep_test.go
@@ -98,6 +98,31 @@ func TestGrepSearch(t *testing.T) {
 	assert.Empty(t, res)
 }
 
+func TestGrepNoBinary(t *testing.T) {
+	tmpDir := t.TempDir()
+
+	err := InitRepository(DefaultContext, tmpDir, false, Sha1ObjectFormat.Name())
+	require.NoError(t, err)
+
+	gitRepo, err := openRepositoryWithDefaultContext(tmpDir)
+	require.NoError(t, err)
+	defer gitRepo.Close()
+
+	require.NoError(t, os.WriteFile(path.Join(tmpDir, "BINARY"), []byte("I AM BINARY\n\x00\nYOU WON'T SEE ME"), 0o666))
+	require.NoError(t, os.WriteFile(path.Join(tmpDir, "TEXT"), []byte("I AM NOT BINARY\nYOU WILL SEE ME"), 0o666))
+
+	err = AddChanges(tmpDir, true)
+	require.NoError(t, err)
+
+	err = CommitChanges(tmpDir, CommitChangesOptions{Message: "Binary and text files"})
+	require.NoError(t, err)
+
+	res, err := GrepSearch(context.Background(), gitRepo, "BINARY", GrepOptions{})
+	require.NoError(t, err)
+	assert.Len(t, res, 1)
+	assert.Equal(t, "TEXT", res[0].Filename)
+}
+
 func TestGrepLongFiles(t *testing.T) {
 	tmpDir := t.TempDir()