Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.11/site-packages/unblob/handlers/compression/lzip.py: 61%

1import io 

2 

3from structlog import get_logger 

4 

5from unblob.extractors import Command 

6 

7from ...file_utils import Endian, convert_int64 

8from ...models import ( 

9 File, 

10 Handler, 

11 HandlerDoc, 

12 HandlerType, 

13 HexString, 

14 Reference, 

15 ValidChunk, 

16) 

17 

18logger = get_logger() 

19 

20# magic (4 bytes) + VN (1 byte) + DS (1 byte) 

21HEADER_LEN = 4 + 1 + 1 

22# LZMA stream is 2 bytes aligned 

23LZMA_ALIGNMENT = 2 

24 

25 

26class LZipHandler(Handler): 

27 NAME = "lzip" 

28 

29 PATTERNS = [HexString("4C 5A 49 50 01")] 

30 

31 EXTRACTOR = Command( 

32 "lziprecover", "-k", "-D0", "-i", "{inpath}", "-o", "{outdir}/lz.uncompressed" 

33 ) 

34 

35 DOC = HandlerDoc( 

36 name="Lzip", 

37 description="Lzip is a lossless compressed file format based on the LZMA algorithm. It features a simple header, CRC-checked integrity, and efficient compression for large files.", 

38 handler_type=HandlerType.COMPRESSION, 

39 vendor=None, 

40 references=[ 

41 Reference( 

42 title="Lzip File Format Documentation", 

43 url="https://www.nongnu.org/lzip/manual/lzip_manual.html", 

44 ), 

45 Reference( 

46 title="Lzip Wikipedia", 

47 url="https://en.wikipedia.org/wiki/Lzip", 

48 ), 

49 ], 

50 limitations=[], 

51 ) 

52 

53 def calculate_chunk(self, file: File, start_offset: int) -> ValidChunk | None: 

54 file.seek(HEADER_LEN, io.SEEK_CUR) 

55 # quite the naive idea but it works 

56 # the idea is to read 8 bytes uint64 every 2 bytes alignment 

57 # until we end up reading the Member Size field which corresponds 

58 # to "the total size of the member, including header and trailer". 

59 # We either find it or reach EOF, which will be caught by finder. 

60 

61 while True: 

62 file.seek(LZMA_ALIGNMENT, io.SEEK_CUR) 

63 member_size = convert_int64(file.read(8), Endian.LITTLE) 

64 if member_size == (file.tell() - start_offset): 

65 end_offset = file.tell() 

66 break 

67 file.seek(-8, io.SEEK_CUR) 

68 

69 return ValidChunk(start_offset=start_offset, end_offset=end_offset)