snapshot of https://github.com/mattermost/mattermost.git (master, 2026-09-16; blobs > 131072B omitted)

server/platform/services/docextractor/archive.go

@@ -0,0 +1,120 @@
+// Copyright (c) 2015-present Mattermost, Inc. All Rights Reserved.
+// See LICENSE.txt for license information.
+
+package docextractor
+
+import (
+	"bytes"
+	"context"
+	"errors"
+	"fmt"
+	"io"
+	"io/fs"
+	"os"
+	"path/filepath"
+	"strings"
+
+	"github.com/mholt/archives"
+
+	"github.com/mattermost/mattermost/server/v8/channels/utils"
+)
+
+type archiveExtractor struct {
+	SubExtractor Extractor
+}
+
+func (ae *archiveExtractor) Name() string {
+	return "archiveExtractor"
+}
+
+func (ae *archiveExtractor) Match(filename string) bool {
+	_, _, err := archives.Identify(context.Background(), filename, nil)
+	return err == nil
+}
+
+// getExtAlsoTarGz returns the extension of the given file name, special casing .tar.gz.
+func getExtAlsoTarGz(name string) string {
+	if strings.HasSuffix(name, ".tar.gz") {
+		return ".tar.gz"
+	}
+
+	return filepath.Ext(name)
+}
+
+func (ae *archiveExtractor) Extract(ctx context.Context, name string, r io.ReadSeeker, maxFileSize int64) (string, error) {
+	// MM-65701: Skip 7zip files due to OOM vulnerability in bodgit/sevenzip library
+	match, _ := (archives.SevenZip{}).Match(context.Background(), name, r)
+	_, _ = r.Seek(0, io.SeekStart) // Reset reader position after Match reads from stream
+	if match.ByName || match.ByStream {
+		return "", nil
+	}
+
+	ext := getExtAlsoTarGz(name)
+
+	// Create a temporary file, using `*` control the random component while preserving the extension.
+	f, err := os.CreateTemp("", "archiver-*"+ext)
+	if err != nil {
+		return "", fmt.Errorf("error creating temporary file: %v", err)
+	}
+	defer os.Remove(f.Name())
+
+	_, err = io.Copy(f, r)
+	f.Close()
+	if err != nil {
+		return "", fmt.Errorf("error copying data into temporary file: %v", err)
+	}
+
+	var text strings.Builder
+	fsys, err := archives.FileSystem(context.Background(), f.Name(), nil)
+	if err != nil {
+		return "", fmt.Errorf("error creating file system: %w", err)
+	}
+
+	err = fs.WalkDir(fsys, ".", func(path string, d fs.DirEntry, err error) error {
+		if err != nil {
+			return err
+		}
+		if d.IsDir() {
+			return nil
+		}
+
+		text.WriteString(path + " ")
+		if ae.SubExtractor != nil {
+			filename := filepath.Base(path)
+			filename = strings.ReplaceAll(filename, "-", " ")
+			filename = strings.ReplaceAll(filename, ".", " ")
+			filename = strings.ReplaceAll(filename, ",", " ")
+
+			file, err := fsys.Open(path)
+			if err != nil {
+				return err
+			}
+			defer file.Close()
+
+			// Limit the size of decompressed archive entries to prevent
+			// memory exhaustion from zip bombs or other malicious archives.
+			var reader io.Reader = file
+			if maxFileSize > 0 {
+				reader = utils.NewLimitedReaderWithError(file, maxFileSize)
+			}
+
+			data, err := io.ReadAll(reader)
+			if err != nil {
+				return fmt.Errorf("error reading archive entry: %w", err)
+			}
+
+			subtext, extractErr := ae.SubExtractor.Extract(ctx, filename, bytes.NewReader(data), maxFileSize)
+			if extractErr == nil {
+				text.WriteString(subtext + " ")
+			} else if errors.Is(extractErr, context.Canceled) || errors.Is(extractErr, context.DeadlineExceeded) {
+				return fmt.Errorf("error extracting archive entry: %w", extractErr)
+			}
+		}
+		return nil
+	})
+	if err != nil {
+		return "", err
+	}
+
+	return text.String(), nil
+}