Initial orientator

2026-05-13 19:52:51 +03:00
commit 49fb60f831
24 changed files with 1919 additions and 0 deletions
--- a/internal/extract/archive.go
+++ b/internal/extract/archive.go
@@ -0,0 +1,160 @@
+// Package extract iterates over the contents of common archive formats
+// (zip, tar, tar.gz, 7z, rar) without extracting to disk.
+package extract
+
+import (
+	"archive/tar"
+	"archive/zip"
+	"bytes"
+	"compress/gzip"
+	"errors"
+	"fmt"
+	"io"
+	"path/filepath"
+	"strings"
+
+	"github.com/bodgit/sevenzip"
+	"github.com/nwaples/rardecode/v2"
+)
+
+type Entry struct {
+	Name string
+	Data []byte
+}
+
+type WalkFn func(Entry) error
+
+// Walk dispatches by extension and calls fn for each file entry.
+func Walk(name string, data []byte, fn WalkFn) error {
+	ext := strings.ToLower(filepath.Ext(name))
+	switch ext {
+	case ".zip":
+		return walkZip(data, fn)
+	case ".tar":
+		return walkTar(bytes.NewReader(data), fn)
+	case ".gz", ".tgz":
+		return walkTarGz(data, fn)
+	case ".7z":
+		return walk7z(data, fn)
+	case ".rar":
+		return walkRar(data, fn)
+	}
+	return fmt.Errorf("unsupported archive: %s", ext)
+}
+
+// IsArchive reports whether name has an extension we know how to extract.
+func IsArchive(name string) bool {
+	switch strings.ToLower(filepath.Ext(name)) {
+	case ".zip", ".tar", ".gz", ".tgz", ".7z", ".rar":
+		return true
+	}
+	return false
+}
+
+func walkZip(data []byte, fn WalkFn) error {
+	r, err := zip.NewReader(bytes.NewReader(data), int64(len(data)))
+	if err != nil {
+		return err
+	}
+	for _, f := range r.File {
+		if f.FileInfo().IsDir() {
+			continue
+		}
+		rc, err := f.Open()
+		if err != nil {
+			return err
+		}
+		buf, err := io.ReadAll(rc)
+		rc.Close()
+		if err != nil {
+			return err
+		}
+		if err := fn(Entry{Name: f.Name, Data: buf}); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+func walkTar(r io.Reader, fn WalkFn) error {
+	tr := tar.NewReader(r)
+	for {
+		hdr, err := tr.Next()
+		if errors.Is(err, io.EOF) {
+			return nil
+		}
+		if err != nil {
+			return err
+		}
+		if hdr.Typeflag != tar.TypeReg {
+			continue
+		}
+		buf, err := io.ReadAll(tr)
+		if err != nil {
+			return err
+		}
+		if err := fn(Entry{Name: hdr.Name, Data: buf}); err != nil {
+			return err
+		}
+	}
+}
+
+func walkTarGz(data []byte, fn WalkFn) error {
+	gz, err := gzip.NewReader(bytes.NewReader(data))
+	if err != nil {
+		return err
+	}
+	defer gz.Close()
+	return walkTar(gz, fn)
+}
+
+func walk7z(data []byte, fn WalkFn) error {
+	r, err := sevenzip.NewReader(bytes.NewReader(data), int64(len(data)))
+	if err != nil {
+		return err
+	}
+	for _, f := range r.File {
+		if f.FileInfo().IsDir() {
+			continue
+		}
+		rc, err := f.Open()
+		if err != nil {
+			return err
+		}
+		buf, err := io.ReadAll(rc)
+		rc.Close()
+		if err != nil {
+			return err
+		}
+		if err := fn(Entry{Name: f.Name, Data: buf}); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+func walkRar(data []byte, fn WalkFn) error {
+	r, err := rardecode.NewReader(bytes.NewReader(data))
+	if err != nil {
+		return err
+	}
+	for {
+		hdr, err := r.Next()
+		if errors.Is(err, io.EOF) {
+			return nil
+		}
+		if err != nil {
+			return err
+		}
+		if hdr.IsDir {
+			continue
+		}
+		buf, err := io.ReadAll(r)
+		if err != nil {
+			return err
+		}
+		if err := fn(Entry{Name: hdr.Name, Data: buf}); err != nil {
+			return err
+		}
+	}
+}