git Service of IN-Ulm e.V. Explore Help
Sign In
cbiedl
/
file
1
0
Fork 0
Files Pull Requests 0 Wiki
Tree: 83e702bf6f
Branches Tags
file
/
Magdir
/
jpeg

jpeg 2.3 KB
History Raw

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061 
  1. 

  2. #------------------------------------------------------------------------------
    3. 

  3. # JPEG images
    4. 

  4. # SunOS 5.5.1 had
    5. 

  5. #
    6. 

  6. #	0	string		\377\330\377\340	JPEG file
    7. 

  7. #	0	string		\377\330\377\356	JPG file
    8. 

  8. #
    9. 

  9. # both of which turn into "JPEG image data" here.
    10. 

  10. #
    11. 

  11. 0	beshort		0xffd8		JPEG image data
    12. 

  12. >6	string		JFIF		\b, JFIF standard
    13. 

  13. 

  14. # The following added by Erik Rossen <rossen@freesurf.ch> 1999-09-06
    15. 

  15. # in a vain attempt to add image size reporting for JFIF.  Note that these
    16. 

  16. # tests are not fool-proof since some perfectly valid JPEGs are currently
    17. 

  17. # impossible to specify in magic(4) format.
    18. 

  18. # First, a little JFIF version info:
    19. 

  19. >11	byte		x		\b %d.
    20. 

  20. >12	byte		x		\b%02d
    21. 

  21. # Next, the resolution or aspect ratio of the image:
    22. 

  22. >13	byte		0		\b, aspect ratio
    23. 

  23. >13	byte		1		\b, resolution (DPI)
    24. 

  24. >13	byte		2		\b, resolution (DPCM)
    25. 

  25. >14	beshort		x		\b X%d:
    26. 

  26. >16	beshort		x		\bY%d
    27. 

  27. #>4	beshort		x		\b, segment length %d
    28. 

  28. # Next, show thumbnail info, if it exists:
    29. 

  29. >18	byte		!0		\b, thumbnail %dx
    30. 

  30. >>19	byte		x		\b%d
    31. 

  31. # Here things get sticky.  We can do ONE MORE marker segment with
    32. 

  32. # indirect addressing, and that's all.  It would be great if we could
    33. 

  33. # do pointer arithemetic like in an assembler language.  Christos?
    34. 

  34. # And if there was some sort of looping construct to do searches, plus a few
    35. 

  35. # named accumulators, it would be even more effective...
    36. 

  36. # At least we can show a comment if no other segments got inserted before:
    37. 

  37. >(4.S+5)	byte		0xFE
    38. 

  38. >>(4.S+8)	string		>\0		\b, "%s"
    39. 

  39. #>(4.S+5)	byte		0xFE		\b, comment
    40. 

  40. #>>(4.S+6)	beshort		x		\b length=%d
    41. 

  41. #>>(4.S+8)	string		>\0		\b, "%s"
    42. 

  42. # Or, we can show the encoding type (I've included only the three most common)
    43. 

  43. # and image dimensions if we are lucky and the SOFn (image segment) is here:
    44. 

  44. >(4.S+5)	byte		0xC0		\b, baseline
    45. 

  45. >>(4.S+6)	byte		x		\b, precision %d
    46. 

  46. >>(4.S+7)	beshort		x		\b, %dx
    47. 

  47. >>(4.S+9)	beshort		x		\b%d
    48. 

  48. >(4.S+5)	byte		0xC1		\b, extended sequential
    49. 

  49. >>(4.S+6)	byte		x		\b, precision %d
    50. 

  50. >>(4.S+7)	beshort		x		\b, %dx
    51. 

  51. >>(4.S+9)	beshort		x		\b%d
    52. 

  52. >(4.S+5)	byte		0xC2		\b, progressive
    53. 

  53. >>(4.S+6)	byte		x		\b, precision %d
    54. 

  54. >>(4.S+7)	beshort		x		\b, %dx
    55. 

  55. >>(4.S+9)	beshort		x		\b%d
    56. 

  56. # I've commented-out quantisation table reporting.  I doubt anyone cares yet.
    57. 

  57. #>(4.S+5)	byte		0xDB		\b, quantisation table
    58. 

  58. #>>(4.S+6)	beshort		x		\b length=%d
    59. 

  59. 

  60. # HSI is Handmade Software's proprietary JPEG encoding scheme
    61. 

  61. 0	string		hsi1		JPEG image data, HSI proprietary
    62.