1
0
Fork 0
mirror of https://github.com/Luzifer/scs-extract.git synced 2024-12-20 21:41:16 +00:00
scs-extract/scs/scs.go

352 lines
8.6 KiB
Go

// Package scs contains a reader for SCS# archive files
package scs
import (
"bytes"
"compress/flate"
"compress/zlib"
"encoding/binary"
"errors"
"fmt"
"io"
"path"
"sort"
"strings"
"github.com/Luzifer/scs-extract/b0rkhash"
)
const (
flagIsDirectory = 0x10
supportedVersion = 0x2
zipHeaderSize = 0x2
)
type (
// File represents a file inside the SCS# archive
File struct {
Name string
CompressedSize uint32
Hash uint64
IsCompressed bool
IsDirectory bool
Size uint32
archiveReader io.ReaderAt
offset uint64
}
// Reader contains a parser for the archive and after creation will
// hold a list of files ready to be opened from the archive
Reader struct {
Files []*File
header fileHeader
entryTable []catalogEntry
metadataTable map[uint32]catalogMetaEntry
archiveReader io.ReaderAt
}
fileHeader struct {
Magic [4]byte
Version uint16
Salt uint16
HashMethod [4]byte
EntryCount uint32
EntryTableLength uint32
MetadataEntriesCount uint32
MetadataTableLength uint32
EntryTableStart uint64
MetadataTableStart uint64
SecurityDescriptorOffset uint32
Platform byte
}
catalogEntry struct {
Hash uint64
MetadataIndex uint32
MetadataCount uint16
Flags uint16
}
catalogMetaEntry struct {
Index uint32
Offset uint64
CompressedSize uint32
Size uint32
Flags byte
IsDirectory bool
IsCompressed bool
}
catalogMetaEntryType byte
)
const (
metaEntryTypeImage catalogMetaEntryType = 1
metaEntryTypeSample catalogMetaEntryType = 2
metaEntryTypeMipProxy catalogMetaEntryType = 3
metaEntryTypeInlineDirectory catalogMetaEntryType = 4
metaEntryTypePlain catalogMetaEntryType = 128
metaEntryTypeDirectory catalogMetaEntryType = 129
metaEntryTypeMip0 catalogMetaEntryType = 130
metaEntryTypeMip1 catalogMetaEntryType = 131
metaEntryTypeMipTail catalogMetaEntryType = 132
)
var (
scsMagic = []byte("SCS#")
scsHashMethod = []byte("CITY")
localeRootPathHash = b0rkhash.CityHash64([]byte("locale"))
rootPathHash = b0rkhash.CityHash64([]byte(""))
)
// NewReader opens the archive from the given io.ReaderAt and parses
// the header information
func NewReader(r io.ReaderAt) (out *Reader, err error) {
// Read the header
var header fileHeader
if err = binary.Read(
io.NewSectionReader(r, 0, int64(binary.Size(fileHeader{}))),
binary.LittleEndian,
&header,
); err != nil {
return nil, fmt.Errorf("reading header: %w", err)
}
// Sanity checks
if !bytes.Equal(header.Magic[:], scsMagic) {
return nil, fmt.Errorf("unexpected magic header")
}
if !bytes.Equal(header.HashMethod[:], scsHashMethod) {
return nil, fmt.Errorf("unexpected hash method")
}
if header.Version != supportedVersion {
return nil, fmt.Errorf("unsupported archive version: %d", header.Version)
}
// Do the real parsing
out = &Reader{
archiveReader: r,
header: header,
}
if err = out.parseEntryTable(); err != nil {
return nil, fmt.Errorf("parsing entry table: %w", err)
}
if err = out.parseMetadataTable(); err != nil {
return nil, fmt.Errorf("parsing metadata table: %w", err)
}
for _, e := range out.entryTable {
meta := out.metadataTable[e.MetadataIndex+uint32(e.MetadataCount)]
f := File{
CompressedSize: meta.CompressedSize,
Hash: e.Hash,
IsCompressed: meta.IsCompressed || (meta.Flags&flagIsDirectory) != 0,
IsDirectory: meta.IsDirectory,
Size: meta.Size,
archiveReader: r,
offset: meta.Offset,
}
out.Files = append(out.Files, &f)
}
return out, out.populateFileNames()
}
// Open opens the file for reading
func (f *File) Open() (io.ReadCloser, error) {
var rc io.ReadCloser
if f.IsCompressed {
r := io.NewSectionReader(f.archiveReader, int64(f.offset+zipHeaderSize), int64(f.CompressedSize)) //#nosec:G115 // int64 wraps at 9EB - We don't have to care for a LONG time
rc = flate.NewReader(r)
} else {
r := io.NewSectionReader(f.archiveReader, int64(f.offset), int64(f.Size)) //#nosec:G115 // int64 wraps at 9EB - We don't have to care for a LONG time
rc = io.NopCloser(r)
}
return rc, nil
}
func (r *Reader) parseEntryTable() error {
etReader, err := zlib.NewReader(io.NewSectionReader(
r.archiveReader,
int64(r.header.EntryTableStart), //#nosec:G115 // int64 wraps at 9EB - We don't have to care for a LONG time
int64(r.header.EntryTableLength),
))
if err != nil {
return fmt.Errorf("opening entry-table reader: %w", err)
}
defer etReader.Close() //nolint:errcheck
for i := uint32(0); i < r.header.EntryCount; i++ {
var e catalogEntry
if err = binary.Read(etReader, binary.LittleEndian, &e); err != nil {
return fmt.Errorf("reading entry: %w", err)
}
r.entryTable = append(r.entryTable, e)
}
sort.Slice(r.entryTable, func(i, j int) bool {
return r.entryTable[i].MetadataIndex < r.entryTable[j].MetadataIndex
})
return nil
}
func (r *Reader) parseMetadataTable() error {
r.metadataTable = make(map[uint32]catalogMetaEntry)
mtReader, err := zlib.NewReader(io.NewSectionReader(
r.archiveReader,
int64(r.header.MetadataTableStart), //#nosec:G115 // int64 wraps at 9EB - We don't have to care for a LONG time
int64(r.header.MetadataTableLength),
))
if err != nil {
return fmt.Errorf("opening metadata-table reader: %w", err)
}
defer mtReader.Close() //nolint:errcheck
for {
var metaType metaEntryType
if err = binary.Read(mtReader, binary.LittleEndian, &metaType); err != nil {
if errors.Is(err, io.EOF) {
return nil
}
return fmt.Errorf("reading meta-type-header: %w", err)
}
var payload iMetaEntry
switch metaType.Type {
case metaEntryTypeDirectory:
var p metaEntryDir
if err = binary.Read(mtReader, binary.LittleEndian, &p); err != nil {
return fmt.Errorf("reading dir definition: %w", err)
}
payload = metaEntry{t: metaType, p: p}
case metaEntryTypePlain:
var p metaEntryFile
if err = binary.Read(mtReader, binary.LittleEndian, &p); err != nil {
return fmt.Errorf("reading file definition: %w", err)
}
payload = metaEntry{t: metaType, p: p}
case metaEntryTypeImage:
var p metaEntryImage
if err = binary.Read(mtReader, binary.LittleEndian, &p); err != nil {
return fmt.Errorf("reading image definition: %w", err)
}
payload = metaEntry{t: metaType, p: p}
default:
return fmt.Errorf("unhandled file type: %v", metaType.Type)
}
var e catalogMetaEntry
payload.Fill(&e)
r.metadataTable[e.Index] = e
}
}
func (r *Reader) populateFileNames() (err error) {
// first seek root entry, without the archive is not usable for us
var entry *File
for _, f := range r.Files {
if f.Hash == rootPathHash {
entry = f
entry.Name = ""
break
} else if f.Hash == localeRootPathHash {
entry = f
entry.Name = "locale"
break
}
}
if entry == nil {
// We found no suitable entrypoint
return fmt.Errorf("no root entry found")
}
if err = r.setFilenamesFromDir(entry); err != nil {
return fmt.Errorf("setting filenames: %w", err)
}
return nil
}
func (r *Reader) setFilenamesFromDir(node *File) error {
f, err := node.Open()
if err != nil {
return fmt.Errorf("opening file: %w", err)
}
defer f.Close() //nolint:errcheck
var entryCount uint32
if err = binary.Read(f, binary.LittleEndian, &entryCount); err != nil {
return fmt.Errorf("reading entry count: %w", err)
}
if entryCount == 0 {
// Listing without any files
return fmt.Errorf("no entries in directory listing")
}
stringLengths := make([]byte, entryCount)
if err = binary.Read(f, binary.LittleEndian, &stringLengths); err != nil {
return fmt.Errorf("reading string lengths: %w", err)
}
for i := uint32(0); i < entryCount; i++ {
var (
hash uint64
name = make([]byte, stringLengths[i])
recurse bool
)
if err = binary.Read(f, binary.LittleEndian, &name); err != nil {
return fmt.Errorf("reading name: %w", err)
}
if name[0] == '/' {
// Directory entry
recurse = true
name = name[1:]
}
hash = b0rkhash.CityHash64([]byte(strings.TrimPrefix(path.Join(node.Name, string(name)), "/")))
var next *File
for _, rf := range r.Files {
if rf.Hash == hash {
next = rf
break
}
}
if next == nil {
return fmt.Errorf("reference to void: %s", path.Join(node.Name, string(name)))
}
next.Name = strings.TrimPrefix(path.Join(node.Name, string(name)), "/")
if recurse {
if err = r.setFilenamesFromDir(next); err != nil {
return err
}
}
}
return nil
}