Reduce calls to git cat-file -s

modules/git/parse_gogit.go

@@ -10,20 +10,21 @@ import (
 	"bytes"
 	"fmt"
 	"strconv"
+	"strings"
 
 	"github.com/go-git/go-git/v5/plumbing/filemode"
 	"github.com/go-git/go-git/v5/plumbing/object"
 )
 
-// ParseTreeEntries parses the output of a `git ls-tree` command.
+// ParseTreeEntries parses the output of a `git ls-tree -l` command.
 func ParseTreeEntries(data []byte) ([]*TreeEntry, error) {
 	return parseTreeEntries(data, nil)
 }
 
 func parseTreeEntries(data []byte, ptree *Tree) ([]*TreeEntry, error) {
 	entries := make([]*TreeEntry, 0, 10)
 	for pos := 0; pos < len(data); {
-		// expect line to be of the form "<mode> <type> <sha>\t<filename>"
+		// expect line to be of the form "<mode> <type> <sha> <space-padded-size>\t<filename>"
 		entry := new(TreeEntry)
 		entry.gogitTreeEntry = &object.TreeEntry{}
 		entry.ptree = ptree
@@ -61,7 +62,16 @@ func parseTreeEntries(data []byte, ptree *Tree) ([]*TreeEntry, error) {
 		entry.gogitTreeEntry.Hash = id
 		pos += 41 // skip over sha and trailing space
 
-		end := pos + bytes.IndexByte(data[pos:], '\n')
+		end := pos + bytes.IndexByte(data[pos:], '\t')
+		if end < pos {
+			return nil, fmt.Errorf("Invalid ls-tree -l output: %s", string(data))
+		}
+		entry.size, _ = strconv.ParseInt(strings.TrimSpace(string(data[pos:end])), 10, 64)
+		entry.sized = true
+
+		pos = end + 1
+
+		end = pos + bytes.IndexByte(data[pos:], '\n')
 		if end < pos {
 			return nil, fmt.Errorf("Invalid ls-tree output: %s", string(data))
 		}

modules/git/parse_gogit_test.go

@@ -24,7 +24,7 @@ func TestParseTreeEntries(t *testing.T) {
 			Expected: []*TreeEntry{},
 		},
 		{
-			Input: "100644 blob 61ab7345a1a3bbc590068ccae37b8515cfc5843c\texample/file2.txt\n",
+			Input: "100644 blob 61ab7345a1a3bbc590068ccae37b8515cfc5843c    1022\texample/file2.txt\n",
 			Expected: []*TreeEntry{
 				{
 					ID: MustIDFromString("61ab7345a1a3bbc590068ccae37b8515cfc5843c"),
@@ -33,12 +33,14 @@ func TestParseTreeEntries(t *testing.T) {
 						Name: "example/file2.txt",
 						Mode: filemode.Regular,
 					},
+					size:  1022,
+					sized: true,
 				},
 			},
 		},
 		{
-			Input: "120000 blob 61ab7345a1a3bbc590068ccae37b8515cfc5843c\t\"example/\\n.txt\"\n" +
-				"040000 tree 1d01fb729fb0db5881daaa6030f9f2d3cd3d5ae8\texample\n",
+			Input: "120000 blob 61ab7345a1a3bbc590068ccae37b8515cfc5843c  234131\t\"example/\\n.txt\"\n" +
+				"040000 tree 1d01fb729fb0db5881daaa6030f9f2d3cd3d5ae8       -\texample\n",
 			Expected: []*TreeEntry{
 				{
 					ID: MustIDFromString("61ab7345a1a3bbc590068ccae37b8515cfc5843c"),
@@ -47,9 +49,12 @@ func TestParseTreeEntries(t *testing.T) {
 						Name: "example/\n.txt",
 						Mode: filemode.Symlink,
 					},
+					size:  234131,
+					sized: true,
 				},
 				{
-					ID: MustIDFromString("1d01fb729fb0db5881daaa6030f9f2d3cd3d5ae8"),
+					ID:    MustIDFromString("1d01fb729fb0db5881daaa6030f9f2d3cd3d5ae8"),
+					sized: true,
 					gogitTreeEntry: &object.TreeEntry{
 						Hash: MustIDFromString("1d01fb729fb0db5881daaa6030f9f2d3cd3d5ae8"),
 						Name: "example",

modules/git/parse_nogogit.go

@@ -10,17 +10,18 @@ import (
 	"bytes"
 	"fmt"
 	"strconv"
+	"strings"
 )
 
-// ParseTreeEntries parses the output of a `git ls-tree` command.
+// ParseTreeEntries parses the output of a `git ls-tree -l` command.
 func ParseTreeEntries(data []byte) ([]*TreeEntry, error) {
 	return parseTreeEntries(data, nil)
 }
 
 func parseTreeEntries(data []byte, ptree *Tree) ([]*TreeEntry, error) {
 	entries := make([]*TreeEntry, 0, 10)
 	for pos := 0; pos < len(data); {
-		// expect line to be of the form "<mode> <type> <sha>\t<filename>"
+		// expect line to be of the form "<mode> <type> <sha> <space-padded-size>\t<filename>"
 		entry := new(TreeEntry)
 		entry.ptree = ptree
 		if pos+6 > len(data) {
@@ -56,7 +57,16 @@ func parseTreeEntries(data []byte, ptree *Tree) ([]*TreeEntry, error) {
 		entry.ID = id
 		pos += 41 // skip over sha and trailing space
 
-		end := pos + bytes.IndexByte(data[pos:], '\n')
+		end := pos + bytes.IndexByte(data[pos:], '\t')
+		if end < pos {
+			return nil, fmt.Errorf("Invalid ls-tree -l output: %s", string(data))
+		}
+		entry.size, _ = strconv.ParseInt(strings.TrimSpace(string(data[pos:end])), 10, 64)
+		entry.sized = true
+
+		pos = end + 1
+
+		end = pos + bytes.IndexByte(data[pos:], '\n')
 		if end < pos {
 			return nil, fmt.Errorf("Invalid ls-tree output: %s", string(data))
 		}

modules/git/parse_nogogit_test.go

@@ -0,0 +1,70 @@
+// Copyright 2021 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+// +build !gogit
+
+package git
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestParseTreeEntries(t *testing.T) {
+
+	testCases := []struct {
+		Input    string
+		Expected []*TreeEntry
+	}{
+		{
+			Input: `100644 blob ea0d83c9081af9500ac9f804101b3fd0a5c293af    8218	README.md
+100644 blob 037f27dc9d353ae4fd50f0474b2194c593914e35    4681	README_ZH.md
+100644 blob 9846a94f7e8350a916632929d0fda38c90dd2ca8     429	SECURITY.md
+040000 tree 84b90550547016f73c5dd3f50dea662389e67b6d       -	assets
+`,
+			Expected: []*TreeEntry{
+				{
+					ID:        MustIDFromString("ea0d83c9081af9500ac9f804101b3fd0a5c293af"),
+					name:      "README.md",
+					entryMode: EntryModeBlob,
+					size:      8218,
+					sized:     true,
+				},
+				{
+					ID:        MustIDFromString("037f27dc9d353ae4fd50f0474b2194c593914e35"),
+					name:      "README_ZH.md",
+					entryMode: EntryModeBlob,
+					size:      4681,
+					sized:     true,
+				},
+				{
+					ID:        MustIDFromString("9846a94f7e8350a916632929d0fda38c90dd2ca8"),
+					name:      "SECURITY.md",
+					entryMode: EntryModeBlob,
+					size:      429,
+					sized:     true,
+				},
+				{
+					ID:        MustIDFromString("84b90550547016f73c5dd3f50dea662389e67b6d"),
+					name:      "assets",
+					entryMode: EntryModeTree,
+					sized:     true,
+				},
+			},
+		},
+	}
+	for _, testCase := range testCases {
+		entries, err := ParseTreeEntries([]byte(testCase.Input))
+		assert.NoError(t, err)
+		assert.EqualValues(t, len(testCase.Expected), len(entries))
+		for i, entry := range entries {
+			assert.EqualValues(t, testCase.Expected[i].ID, entry.ID)
+			assert.EqualValues(t, testCase.Expected[i].name, entry.name)
+			assert.EqualValues(t, testCase.Expected[i].entryMode, entry.entryMode)
+			assert.EqualValues(t, testCase.Expected[i].sized, entry.sized)
+			assert.EqualValues(t, testCase.Expected[i].size, entry.size)
+		}
+	}
+}

modules/git/tree_entry_nogogit.go

@@ -87,5 +87,7 @@ func (te *TreeEntry) Blob() *Blob {
 		ID:       te.ID,
 		repoPath: te.ptree.repo.Path,
 		name:     te.Name(),
+		size:     te.size,
+		gotSize:  te.sized,
 	}
 }

modules/git/tree_nogogit.go

@@ -32,7 +32,7 @@ func (t *Tree) ListEntries() (Entries, error) {
 		return t.entries, nil
 	}
 
-	stdout, err := NewCommand("ls-tree", t.ID.String()).RunInDirBytes(t.repo.Path)
+	stdout, err := NewCommand("ls-tree", "-l", t.ID.String()).RunInDirBytes(t.repo.Path)
 	if err != nil {
 		if strings.Contains(err.Error(), "fatal: Not a valid object name") || strings.Contains(err.Error(), "fatal: not a tree object") {
 			return nil, ErrNotExist{
@@ -55,7 +55,7 @@ func (t *Tree) ListEntriesRecursive() (Entries, error) {
 	if t.entriesRecursiveParsed {
 		return t.entriesRecursive, nil
 	}
-	stdout, err := NewCommand("ls-tree", "-t", "-r", t.ID.String()).RunInDirBytes(t.repo.Path)
+	stdout, err := NewCommand("ls-tree", "-t", "-l", "-r", t.ID.String()).RunInDirBytes(t.repo.Path)
 	if err != nil {
 		return nil, err
 	}

modules/indexer/code/bleve.go

@@ -179,14 +179,20 @@ func (b *BleveIndexer) addUpdate(commitSha string, update fileUpdate, repo *mode
 		return nil
 	}
 
-	stdout, err := git.NewCommand("cat-file", "-s", update.BlobSha).
-		RunInDir(repo.RepoPath())
-	if err != nil {
-		return err
+	size := update.Size
+
+	if !update.Sized {
+		stdout, err := git.NewCommand("cat-file", "-s", update.BlobSha).
+			RunInDir(repo.RepoPath())
+		if err != nil {
+			return err
+		}
+		if size, err = strconv.ParseInt(strings.TrimSpace(stdout), 10, 64); err != nil {
+			return fmt.Errorf("Misformatted git cat-file output: %v", err)
+		}
 	}
-	if size, err := strconv.Atoi(strings.TrimSpace(stdout)); err != nil {
-		return fmt.Errorf("Misformatted git cat-file output: %v", err)
-	} else if int64(size) > setting.Indexer.MaxIndexerFileSize {
+
+	if size > setting.Indexer.MaxIndexerFileSize {
 		return b.addDelete(update.Filename, repo, batch)
 	}
 

modules/indexer/code/elastic_search.go

@@ -178,14 +178,20 @@ func (b *ElasticSearchIndexer) addUpdate(sha string, update fileUpdate, repo *mo
 		return nil, nil
 	}
 
-	stdout, err := git.NewCommand("cat-file", "-s", update.BlobSha).
-		RunInDir(repo.RepoPath())
-	if err != nil {
-		return nil, err
+	size := update.Size
+
+	if !update.Sized {
+		stdout, err := git.NewCommand("cat-file", "-s", update.BlobSha).
+			RunInDir(repo.RepoPath())
+		if err != nil {
+			return nil, err
+		}
+		if size, err = strconv.ParseInt(strings.TrimSpace(stdout), 10, 64); err != nil {
+			return nil, fmt.Errorf("Misformatted git cat-file output: %v", err)
+		}
 	}
-	if size, err := strconv.Atoi(strings.TrimSpace(stdout)); err != nil {
-		return nil, fmt.Errorf("Misformatted git cat-file output: %v", err)
-	} else if int64(size) > setting.Indexer.MaxIndexerFileSize {
+
+	if size > setting.Indexer.MaxIndexerFileSize {
 		return []elastic.BulkableRequest{b.addDelete(update.Filename, repo)}, nil
 	}
 

modules/indexer/code/git.go

@@ -17,6 +17,8 @@ import (
 type fileUpdate struct {
 	Filename string
 	BlobSha  string
+	Size     int64
+	Sized    bool
 }
 
 // repoChanges changes (file additions/updates/removals) to a repo
@@ -77,6 +79,8 @@ func parseGitLsTreeOutput(stdout []byte) ([]fileUpdate, error) {
 			updates[idxCount] = fileUpdate{
 				Filename: entry.Name(),
 				BlobSha:  entry.ID.String(),
+				Size:     entry.Size(),
+				Sized:    true,
 			}
 			idxCount++
 		}
@@ -87,7 +91,7 @@ func parseGitLsTreeOutput(stdout []byte) ([]fileUpdate, error) {
 // genesisChanges get changes to add repo to the indexer for the first time
 func genesisChanges(repo *models.Repository, revision string) (*repoChanges, error) {
 	var changes repoChanges
-	stdout, err := git.NewCommand("ls-tree", "--full-tree", "-r", revision).
+	stdout, err := git.NewCommand("ls-tree", "--full-tree", "-l", "-r", revision).
 		RunInDirBytes(repo.RepoPath())
 	if err != nil {
 		return nil, err
@@ -162,7 +166,7 @@ func nonGenesisChanges(repo *models.Repository, revision string) (*repoChanges,
 		}
 	}
 
-	cmd := git.NewCommand("ls-tree", "--full-tree", revision, "--")
+	cmd := git.NewCommand("ls-tree", "--full-tree", "-l", revision, "--")
 	cmd.AddArguments(updatedFilenames...)
 	lsTreeStdout, err := cmd.RunInDirBytes(repo.RepoPath())
 	if err != nil {