David Bremner wrote: > You can find the tarball here: > > http://www.math.tu-berlin.de/polymake/download-alpha/polymake-2.9.6.tar.bz2
Reproduced & added to test suite. bzip2 -9 gets closest to this, but only about 256k of the 1.1 mb are identical. That includes the header of the file, and quite a lot of the beginning of the compressed data. Then it diverges: @@ -11796,56895 +11796,56897 @@ 0002e130 28 2b 24 ca 6b 28 31 d4 08 a0 69 e0 2f ff ff ff |(+$.k(1...i./...| 0002e140 ff bc bf ff ff ff ff ff ff ff ff ff ff ff e0 21 |...............!| 0002e150 0e 26 10 08 41 00 90 40 20 08 14 80 61 0f 0c 66 |.&..A..@ ...a..f| -0002e160 23 c6 51 ed 1f 20 03 ea 9e d4 d6 53 0d 41 3b 2b |#.Q.. .....S.A;+| +0002e160 23 79 ca 3d a3 e4 00 7a a7 b5 35 ee d3 86 a0 9e |#y.=...z..5.....| If you know anything about the version of bzip2 that generated the file, or can find out, that'd be very useful. Perhaps it was an earlier version. I don't know anything yet about how stable bzip2's output has been historically. -- see shy jo
signature.asc
Description: Digital signature
