dfindexeddb 20240331a0__py3-none-any.whl → 20240417__py3-none-any.whl

dfindexeddb/indexeddb/chromium/definitions.py

@@ -141,10 +141,12 @@ class BlinkSerializationTag(IntEnum):
   ENCODED_AUDIO_CHUNK = ord('y')
   ENCODED_VIDEO_CHUNK = ord('z')
   CROP_TARGET = ord('c')
+  RESTRICTION_TARGET = ord('D')
   MEDIA_SOURCE_HANDLE = ord('S')
   DEPRECATED_DETECTED_BARCODE = ord('B')
   DEPRECATED_DETECTED_FACE = ord('F')
   DEPRECATED_DETECTED_TEXT = ord('t')
+  FENCED_FRAME_CONFIG = ord('C')
   DOM_EXCEPTION = ord('x')
   TRAILER_OFFSET = 0xFE
   VERSION = 0xFF
@@ -304,3 +306,67 @@ class V8ErrorTag(IntEnum):
   CAUSE = ord('c')
   STACK = ord('s')
   END = ord('.')
+
+
+class ImageSerializationTag(IntEnum):
+  """Image Serialization tags."""
+  END = 0
+  PREDEFINED_COLOR_SPACE = 1
+  CANVAS_PIXEL_FORMAT = 2
+  IMAGE_DATA_STORAGE_FORMAT = 3
+  ORIGIN_CLEAN = 4
+  IS_PREMULTIPLIED = 5
+  CANVAS_OPACITY_MODE = 6
+  PARAMETRIC_COLOR_SPACE = 7
+  IMAGE_ORIENTATION = 8
+  LAST = IMAGE_ORIENTATION
+
+
+class SerializedPredefinedColorSpace(IntEnum):
+  """Serialized Predefined Color Space enumeration."""
+  LEGACY_OBSOLETE = 0
+  SRGB = 1
+  REC2020 = 2
+  P3 = 3
+  REC2100HLG = 4
+  REC2100PQ = 5
+  SRGB_LINEAR = 6
+  LAST = SRGB_LINEAR
+
+
+class SerializedPixelFormat(IntEnum):
+  """Serialized Pixel Format enumeration."""
+  NATIVE8_LEGACY_OBSOLETE = 0
+  F16 = 1
+  RGBA8 = 2
+  BGRA8 = 3
+  RGBX8 = 4
+  LAST = RGBX8
+
+
+class SerializedImageDataStorageFormat(IntEnum):
+  """The Serialized Image Data Storage Format."""
+  UINT8CLAMPED = 0
+  UINT16 = 1
+  FLOAT32 = 2
+  LAST = FLOAT32
+
+
+class SerializedOpacityMode(IntEnum):
+  """The Serialized Opacity Mode."""
+  KNONOPAQUE = 0
+  KOPAQUE = 1
+  KLAST = KOPAQUE
+
+
+class SerializedImageOrientation(IntEnum):
+  """The Serialized Image Orientation."""
+  TOP_LEFT = 0
+  TOP_RIGHT = 1
+  BOTTOM_RIGHT = 2
+  BOTTOM_LEFT = 3
+  LEFT_TOP = 4
+  RIGHT_TOP = 5
+  RIGHT_BOTTOM = 6
+  LEFT_BOTTOM = 7
+  LAST = LEFT_BOTTOM

dfindexeddb/indexeddb/chromium/record.py

@@ -22,8 +22,7 @@ from typing import Any, BinaryIO, Optional, Tuple, Type, TypeVar, Union
 from dfindexeddb import errors
 from dfindexeddb.indexeddb.chromium import blink
 from dfindexeddb.indexeddb.chromium import definitions
-from dfindexeddb.leveldb import ldb
-from dfindexeddb.leveldb import log
+from dfindexeddb.leveldb import record
 from dfindexeddb.leveldb import utils
 
 
@@ -546,7 +545,7 @@ class EarliestSweepKey(BaseIndexedDBKey):
 
 
 @dataclass
-class EarlistCompactionTimeKey(BaseIndexedDBKey):
+class EarliestCompactionTimeKey(BaseIndexedDBKey):
   """An earliest compaction time IndexedDB key."""
 
   def DecodeValue(self, decoder: utils.LevelDBDecoder) -> int:
@@ -558,11 +557,11 @@ class EarlistCompactionTimeKey(BaseIndexedDBKey):
   def FromDecoder(
       cls, decoder: utils.LevelDBDecoder, key_prefix: KeyPrefix,
       base_offset: int = 0
-  ) -> EarlistCompactionTimeKey:
+  ) -> EarliestCompactionTimeKey:
     """Decodes the earliest compaction time key."""
     offset, key_type = decoder.DecodeUint8()
     if key_type != definitions.GlobalMetadataKeyType.EARLIEST_COMPACTION_TIME:
-      raise errors.ParserError('Not a EarlistCompactionTimeKey')
+      raise errors.ParserError('Not a EarliestCompactionTimeKey')
     return cls(offset=base_offset + offset, key_prefix=key_prefix)
 
 
@@ -668,7 +667,7 @@ class GlobalMetaDataKey(BaseIndexedDBKey):
       definitions.GlobalMetadataKeyType
           .EARLIEST_SWEEP: EarliestSweepKey,
       definitions.GlobalMetadataKeyType
-          .EARLIEST_COMPACTION_TIME: EarlistCompactionTimeKey,
+          .EARLIEST_COMPACTION_TIME: EarliestCompactionTimeKey,
       definitions.GlobalMetadataKeyType
           .SCOPES_PREFIX: ScopesPrefixKey,
       definitions.GlobalMetadataKeyType
@@ -692,7 +691,7 @@ class GlobalMetaDataKey(BaseIndexedDBKey):
              Type[DatabaseFreeListKey],
              Type[DatabaseNameKey],
              Type[EarliestSweepKey],
-             Type[EarlistCompactionTimeKey],
+             Type[EarliestCompactionTimeKey],
              Type[MaxDatabaseIdKey],
              Type[RecoveryBlobJournalKey],
              Type[SchemaVersionKey],
@@ -972,7 +971,7 @@ class ObjectStoreDataValue:
     blob_offset: the blob offset, only valid if wrapped.
     value: the blink serialized value, only valid if not wrapped.
   """
-  unkown: int
+  unknown: int
   is_wrapped: bool
   blob_size: Optional[int]
   blob_offset: Optional[int]
@@ -1003,7 +1002,7 @@ class ObjectStoreDataKey(BaseIndexedDBKey):
       _, blob_size = decoder.DecodeVarint()
       _, blob_offset = decoder.DecodeVarint()
       return ObjectStoreDataValue(
-          unkown=unknown_integer,
+          unknown=unknown_integer,
           is_wrapped=True,
           blob_size=blob_size,
           blob_offset=blob_offset,
@@ -1011,7 +1010,7 @@ class ObjectStoreDataKey(BaseIndexedDBKey):
     _, blink_bytes = decoder.ReadBytes()
     blink_value = blink.V8ScriptValueDecoder.FromBytes(blink_bytes)
     return ObjectStoreDataValue(
-        unkown=unknown_integer,
+        unknown=unknown_integer,
         is_wrapped=False,
         blob_size=None,
         blob_offset=None,
@@ -1337,24 +1336,33 @@ class IndexedDBRecord:
     value: the value of the record.
     sequence_number: if available, the sequence number of the record.
     type: the type of the record.
+    level: the leveldb level, None indicates the record came from a log file.
+    recovered: True if the record is a recovered record.
   """
+  path: str
   offset: int
   key: Any
   value: Any
-  sequence_number: int
+  sequence_number: Optional[int]
   type: int
+  level: Optional[int]
+  recovered: Optional[bool]
 
   @classmethod
   def FromLevelDBRecord(
-      cls, record: Union[ldb.KeyValueRecord, log.ParsedInternalKey]
+      cls, db_record: record.LevelDBRecord
   ) -> IndexedDBRecord:
     """Returns an IndexedDBRecord from a ParsedInternalKey."""
-    idb_key = IndexedDbKey.FromBytes(record.key, base_offset=record.offset)
-    idb_value = idb_key.ParseValue(record.value)
+    idb_key = IndexedDbKey.FromBytes(
+        db_record.record.key, base_offset=db_record.record.offset)
+    idb_value = idb_key.ParseValue(db_record.record.value)
     return cls(
-      offset=record.offset,
-      key=idb_key,
-      value=idb_value,
-      sequence_number=record.sequence_number if hasattr(
-          record, 'sequence_number') else None,
-      type=record.record_type)
+        path=db_record.path,
+        offset=db_record.record.offset,
+        key=idb_key,
+        value=idb_value,
+        sequence_number=db_record.record.sequence_number if hasattr(
+            db_record.record, 'sequence_number') else None,
+        type=db_record.record.record_type,
+        level=db_record.level,
+        recovered=db_record.recovered)

dfindexeddb/indexeddb/chromium/v8.py

@@ -152,7 +152,12 @@ class ValueDeserializer:
       _, tag_value = self.decoder.PeekBytes(1)
     except errors.DecoderError:
       return None
-    return definitions.V8SerializationTag(tag_value[0])
+    try:
+      return definitions.V8SerializationTag(tag_value[0])
+    except ValueError as error:
+      raise errors.ParserError(
+          f'Invalid v8 tag value {tag_value} at offset'
+          f' {self.decoder.stream.tell()}') from error
 
   def _ReadTag(self) -> definitions.V8SerializationTag:
     """Returns the next non-padding serialization tag.
@@ -269,7 +274,7 @@ class ValueDeserializer:
             self.version >= 15):
       parsed_object = self.ReadSharedObject()
     elif self.version < 13:
-      self.decoder.stream.seek(-1)
+      self.decoder.stream.seek(-1, os.SEEK_CUR)
       parsed_object = self.ReadHostObject()
     else:
       parsed_object = None
@@ -492,7 +497,7 @@ class ValueDeserializer:
     return value
 
   def _ReadJSRegExp(self) -> RegExp:
-    """Reads a Javscript regular expression from the current position."""
+    """Reads a Javascript regular expression from the current position."""
     next_id = self._GetNextId()
     pattern = self.ReadString()
     _, flags = self.decoder.DecodeUint32Varint()  # TODO: verify flags

dfindexeddb/indexeddb/cli.py

@@ -15,6 +15,7 @@
 """A CLI tool for dfindexeddb."""
 import argparse
 import dataclasses
+import enum
 from datetime import datetime
 import json
 import pathlib
@@ -57,6 +58,8 @@ class Encoder(json.JSONEncoder):
       return list(o)
     if isinstance(o, v8.RegExp):
       return str(o)
+    if isinstance(o, enum.Enum):
+      return o.name
     return json.JSONEncoder.default(self, o)
 
 
@@ -70,22 +73,78 @@ def _Output(structure, output):
     print(structure)
 
 
-def IndexeddbCommand(args):
-  """The CLI for processing a log/ldb file as indexeddb."""
-  for db_record in leveldb_record.LevelDBRecord.FromDir(args.source):
+def DbCommand(args):
+  """The CLI for processing a directory as indexeddb."""
+  if args.use_manifest:
+    for db_record in leveldb_record.LevelDBRecord.FromManifest(args.source):
+      record = db_record.record
+      try:
+        idb_record = chromium_record.IndexedDBRecord.FromLevelDBRecord(
+            db_record)
+      except(
+          errors.ParserError,
+          errors.DecoderError,
+          NotImplementedError) as err:
+        print((
+            f'Error parsing Indexeddb record {record.__class__.__name__}: {err}'
+            f' at offset {record.offset} in {db_record.path}'), file=sys.stderr)
+        print(f'Traceback: {traceback.format_exc()}', file=sys.stderr)
+        continue
+      _Output(idb_record, output=args.output)
+  else:
+    for db_record in leveldb_record.LevelDBRecord.FromDir(args.source):
+      record = db_record.record
+      try:
+        idb_record = chromium_record.IndexedDBRecord.FromLevelDBRecord(
+            db_record)
+      except(
+          errors.ParserError,
+          errors.DecoderError,
+          NotImplementedError) as err:
+        print((
+            f'Error parsing Indexeddb record {record.__class__.__name__}: {err}'
+            f' at offset {record.offset} in {db_record.path}'), file=sys.stderr)
+        print(f'Traceback: {traceback.format_exc()}', file=sys.stderr)
+        continue
+      _Output(idb_record, output=args.output)
+
+
+def LdbCommand(args):
+  """The CLI for processing a leveldb table (.ldb) file as indexeddb."""
+  for db_record in leveldb_record.LevelDBRecord.FromFile(args.source):
+    record = db_record.record
+    try:
+      idb_record = chromium_record.IndexedDBRecord.FromLevelDBRecord(
+          db_record)
+    except(
+        errors.ParserError,
+        errors.DecoderError,
+        NotImplementedError) as err:
+      print(
+          (f'Error parsing Indexeddb record {record.__class__.__name__}: {err} '
+           f'at offset {record.offset} in {db_record.path}'), file=sys.stderr)
+      print(f'Traceback: {traceback.format_exc()}', file=sys.stderr)
+      continue
+    _Output(idb_record, output=args.output)
+
+
+def LogCommand(args):
+  """The CLI for processing a leveldb log file as indexeddb."""
+  for db_record in leveldb_record.LevelDBRecord.FromFile(args.source):
     record = db_record.record
     try:
-      db_record.record = chromium_record.IndexedDBRecord.FromLevelDBRecord(
-          record)
+      idb_record = chromium_record.IndexedDBRecord.FromLevelDBRecord(
+          db_record)
     except(
         errors.ParserError,
         errors.DecoderError,
         NotImplementedError) as err:
       print(
-          (f'Error parsing blink value: {err} for {record.__class__.__name__} '
+          (f'Error parsing Indexeddb record {record.__class__.__name__}: {err} '
            f'at offset {record.offset} in {db_record.path}'), file=sys.stderr)
       print(f'Traceback: {traceback.format_exc()}', file=sys.stderr)
-    _Output(db_record, output=args.output)
+      continue
+    _Output(idb_record, output=args.output)
 
 
 def App():
@@ -94,10 +153,51 @@ def App():
       prog='dfindexeddb',
       description='A cli tool for parsing indexeddb files',
       epilog=f'Version {version.GetVersion()}')
-  parser.add_argument(
+
+  subparsers = parser.add_subparsers()
+
+  parser_db = subparsers.add_parser(
+      'db', help='Parse a directory as indexeddb.')
+  parser_db.add_argument(
       '-s', '--source', required=True, type=pathlib.Path,
       help='The source leveldb folder')
-  parser.add_argument(
+  parser_db.add_argument(
+      '--use_manifest',
+      action='store_true',
+      help='Use manifest file to determine active/deleted records.')
+  parser_db.add_argument(
+      '-o',
+      '--output',
+      choices=[
+          'json',
+          'jsonl',
+          'repr'],
+      default='json',
+      help='Output format.  Default is json')
+  parser_db.set_defaults(func=DbCommand)
+
+  parser_ldb = subparsers.add_parser(
+      'ldb', help='Parse a ldb file as indexeddb.')
+  parser_ldb.add_argument(
+      '-s', '--source', required=True, type=pathlib.Path,
+      help='The source .ldb file.')
+  parser_ldb.add_argument(
+      '-o',
+      '--output',
+      choices=[
+          'json',
+          'jsonl',
+          'repr'],
+      default='json',
+      help='Output format.  Default is json')
+  parser_ldb.set_defaults(func=LdbCommand)
+
+  parser_log = subparsers.add_parser(
+      'log', help='Parse a log file as indexeddb.')
+  parser_log.add_argument(
+      '-s', '--source', required=True, type=pathlib.Path,
+      help='The source .log file.')
+  parser_log.add_argument(
       '-o',
       '--output',
       choices=[
@@ -106,7 +206,7 @@ def App():
           'repr'],
       default='json',
       help='Output format.  Default is json')
-  parser.set_defaults(func=IndexeddbCommand)
+  parser_log.set_defaults(func=LogCommand)
 
   args = parser.parse_args()
   args.func(args)

dfindexeddb/leveldb/cli.py

@@ -66,8 +66,12 @@ def _Output(structure, output):
 
 def DbCommand(args):
   """The CLI for processing leveldb folders."""
-  for rec in record.LevelDBRecord.FromDir(args.source):
-    _Output(rec, output=args.output)
+  if args.use_manifest:
+    for rec in record.LevelDBRecord.FromManifest(args.source):
+      _Output(rec, output=args.output)
+  else:
+    for rec in record.LevelDBRecord.FromDir(args.source):
+      _Output(rec, output=args.output)
 
 
 def LdbCommand(args):
@@ -159,6 +163,10 @@ def App():
       required=True,
       type=pathlib.Path,
       help='The source leveldb directory')
+  parser_db.add_argument(
+      '--use_manifest',
+      action='store_true',
+      help='Use manifest file to determine active/deleted records.')
   parser_db.add_argument(
       '-o',
       '--output',
@@ -168,6 +176,7 @@ def App():
           'repr'],
       default='json',
       help='Output format.  Default is json')
+  parser_db.set_defaults(func=DbCommand)
 
   parser_log = subparsers.add_parser(
       'log', help='Parse a leveldb log file.')

dfindexeddb/leveldb/definitions.py

@@ -25,6 +25,8 @@ PACKED_SEQUENCE_AND_TYPE_LENGTH = 8
 SEQUENCE_LENGTH = 7
 TYPE_LENGTH = 1
 
+MANIFEST_FILENAME_PATTERN = r'MANIFEST-[0-9]{6}'
+
 
 class BlockCompressionType(enum.IntEnum):
   """Block compression types."""

dfindexeddb/leveldb/descriptor.py

@@ -365,7 +365,8 @@ class FileReader:
     current_log = None
 
     for version_edit in self.GetVersionEdits():
-      current_log = f'{version_edit.log_number:06d}.log'
+      if version_edit.log_number:
+        current_log = f'{version_edit.log_number:06d}.log'
 
       for new_file in version_edit.new_files:
         active_files[new_file.level][f'{new_file.number:06d}.ldb'] = new_file
@@ -380,3 +381,8 @@ class FileReader:
           deleted_files=dict(deleted_files),
           version_edit_offset=version_edit.offset,
           last_sequence=version_edit.last_sequence)
+
+  def GetLatestVersion(self) -> LevelDBVersion:
+    """Returns the latest LevelDBVersion instance."""
+    *_, latest = self.GetVersions()
+    return latest

dfindexeddb/leveldb/record.py

@@ -16,9 +16,12 @@
 from __future__ import annotations
 import dataclasses
 import pathlib
+import re
 import sys
-from typing import Any, Generator, Union
+from typing import Any, Generator, Optional, Union
 
+from dfindexeddb import errors
+from dfindexeddb.leveldb import definitions
 from dfindexeddb.leveldb import descriptor
 from dfindexeddb.leveldb import ldb
 from dfindexeddb.leveldb import log
@@ -34,18 +37,20 @@ class LevelDBRecord:
   Attributes:
     path: the file path where the record was parsed from.
     record: the leveldb record.
+    level: the leveldb level, None indicates the record came from a log file.
+    recovered: True if the record is a recovered record.
   """
   path: str
   record: Union[
       ldb.KeyValueRecord,
-      log.ParsedInternalKey,
-      descriptor.VersionEdit]
+      log.ParsedInternalKey]
+  level: Optional[int] = None
+  recovered: Optional[bool] = None
 
   @classmethod
   def FromFile(
       cls,
-      file_path: pathlib.Path,
-      include_versionedit: bool = False
+      file_path: pathlib.Path
   ) -> Generator[LevelDBRecord, Any, Any]:
     """Yields leveldb records from the given path.
 
@@ -54,7 +59,6 @@ class LevelDBRecord:
 
     Args:
       file_path: the file path.
-      include_versionedit: include VersionEdit records from descriptor files.
     """
     if file_path.name.endswith('.log'):
       for record in log.FileReader(
@@ -64,12 +68,7 @@ class LevelDBRecord:
       for record in ldb.FileReader(file_path.as_posix()).GetKeyValueRecords():
         yield cls(path=file_path.as_posix(), record=record)
     elif file_path.name.startswith('MANIFEST'):
-      if not include_versionedit:
-        print(f'Ignoring {file_path.as_posix()}', file=sys.stderr)
-        return
-      for record in descriptor.FileReader(
-          file_path.as_posix()).GetVersionEdits():
-        yield cls(path=file_path.as_posix(), record=record)
+      print(f'Ignoring descriptor file {file_path.as_posix()}', file=sys.stderr)
     elif file_path.name in ('LOCK', 'CURRENT', 'LOG', 'LOG.old'):
       print(f'Ignoring {file_path.as_posix()}', file=sys.stderr)
     else:
@@ -78,25 +77,114 @@ class LevelDBRecord:
   @classmethod
   def FromDir(
       cls,
-      path: pathlib.Path,
-      include_versionedit: bool = False
+      path: pathlib.Path
   ) -> Generator[LevelDBRecord, Any, Any]:
     """Yields LevelDBRecords from the given directory.
 
     Args:
       path: the file path.
-      include_versionedit: include VersionEdit records from descriptor files.
+
+    Yields:
+      LevelDBRecords
+    """
+    if not path or not path.is_dir():
+      raise ValueError(f'{path} is not a directory')
+    for file_path in path.iterdir():
+      yield from cls.FromFile(file_path=file_path)
+
+  @classmethod
+  def FromManifest(
+      cls,
+      path: pathlib.Path
+  ) -> Generator[LevelDBRecord, Any, Any]:
+    """Yields LevelDBRecords from the given directory using the manifest.
+
+    Args:
+      path: the file path.
 
     Yields:
       LevelDBRecords
 
     Raises:
+      ParserError: if the CURRENT or MANIFEST-* file does not exist.
       ValueError: if path is not a directory.
     """
-    if path.is_dir():
-      for file_path in path.iterdir():
-        yield from cls.FromFile(
-            file_path=file_path,
-            include_versionedit=include_versionedit)
-    else:
+    if not path or not path.is_dir():
       raise ValueError(f'{path} is not a directory')
+
+    current_path = path / 'CURRENT'
+    if not current_path.exists():
+      raise errors.ParserError(f'{current_path!s} does not exist.')
+
+    current_manifest = current_path.read_text().strip()
+    manifest_regex = re.compile(definitions.MANIFEST_FILENAME_PATTERN)
+    if not manifest_regex.fullmatch(current_manifest):
+      raise errors.ParserError(
+          f'{current_path!s} does not contain the expected content')
+
+    manifest_path = path / current_manifest
+    if not manifest_path.exists():
+      raise errors.ParserError(f'{manifest_path!s} does not exist.')
+
+    latest_version = descriptor.FileReader(
+        str(manifest_path)).GetLatestVersion()
+    if not latest_version:
+      raise errors.ParserError(
+          f'Could not parse a leveldb version from {manifest_path!s}')
+
+    # read log records
+    log_records = []
+    if latest_version.current_log:
+      current_log = path / latest_version.current_log
+      if current_log.exists():
+        for log_record in cls.FromFile(file_path=current_log):
+          log_records.append(log_record)
+    else:
+      print('No current log file.', file=sys.stderr)
+
+    # read records from the "young" or 0-level
+    young_records = []
+    for active_file in latest_version.active_files.get(0, {}).keys():
+      current_young = path / active_file
+      if current_young.exists():
+        for young_record in cls.FromFile(current_young):
+          young_records.append(young_record)
+
+    active_records = {}
+    for record in sorted(
+        log_records,
+        key=lambda record: record.record.sequence_number,
+        reverse=True):
+      if record.record.key not in active_records:
+        record.recovered = False
+        active_records[record.record.key] = record
+      else:
+        record.recovered = True
+
+    for record in sorted(
+        young_records,
+        key=lambda record: record.record.sequence_number,
+        reverse=True):
+      if record.record.key not in active_records:
+        record.recovered = False
+        active_records[record.record.key] = record
+      else:
+        record.recovered = True
+      record.level = 0
+
+    yield from sorted(
+        log_records + young_records,
+        key=lambda record: record.record.sequence_number,
+        reverse=False)
+
+    if latest_version.active_files.keys():
+      for level in range(1, max(latest_version.active_files.keys()) + 1):
+        for filename in latest_version.active_files.get(level, []):
+          current_filename = path / filename
+          for record in cls.FromFile(file_path=current_filename):
+            if record.record.key in active_records:
+              record.recovered = True
+            else:
+              record.recovered = False
+            record.level = level
+            yield record

dfindexeddb/version.py

@@ -14,7 +14,8 @@
 # limitations under the License.
 """Version information for dfIndexeddb."""
 
-__version__ = "20240331a"
+
+__version__ = "20240417"
 
 
 def GetVersion():