Incorrect OS/2 Table unicode and codepage range

For Tuffy.ttf, run subsetter, and use Microsoft ttfdump to observe dumped file. 
 (Microsoft Font Tools can be downloaded from 
http://www.microsoft.com/typography/tools/tools.aspx).

Expected result:

  Unicode Range 1( Bits 0 - 31 ): 80000003
  Unicode Range 2( Bits 32- 63 ): 00000002
  Unicode Range 3( Bits 64- 95 ): 00000000
  Unicode Range 4( Bits 96-127 ): 00000000

  CodePage Range 1( Bits 0 - 31 ): 00000001
  CodePage Range 2( Bits 32- 63 ): 80D40000

Actual result:

  Unicode Range 1( Bits 0 - 31 ): 1800CD1E
  Unicode Range 2( Bits 32- 63 ): 7977C949
  Unicode Range 3( Bits 64- 95 ): 8043FEFF
  Unicode Range 4( Bits 96-127 ): FFFF7A38

  CodePage Range 1( Bits 0 - 31 ): 3F000201
  CodePage Range 2( Bits 32- 63 ): 0000002F

Note: Hex editor reveals that the bytes are written correctly, therefore this 
could be due to invalid byte paddings.

Java version does not have the problem.

Original issue reported on code.google.com by arthur...@google.com on 12 Jul 2011 at 7:55

linxingang / sfntly

Incorrect OS/2 Table unicode and codepage range #2