python-pillow/Pillow
1
1
Fork 0
mirror of https://github.com/python-pillow/Pillow.git synced 2024-12-25 17:36:18 +03:00
Code Issues Packages Projects Releases Wiki Activity

Comments

Browse Source 
Fixes #414
This commit is contained in:
wiredfool 2013-11-20 17:03:46 -08:00
parent e4fe6ae7a3
commit cc5e8f0f6d
1 changed files with 10 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
libImaging/Unpack.c
View File

@ -797,21 +797,22 @@ unpackI16N_I16(UINT8* out, const UINT8* in, int pixels){
static void
unpackI12_I16(UINT8* out, const UINT8* in, int pixels){
/* Fillorder 1/MSB -> LittleEndian
/* Fillorder 1/MSB -> LittleEndian, for 12bit integer greyscale tiffs.
According to the spec:
According to the TIFF spec:
FillOrder = 2 should be used only when BitsPerSample = 1 and
the data is either uncompressed or compressed using CCITT 1D
or 2D compression, to avoid potentially ambigous situations.
Yeah. I thought so. See how well people read the spec.
Yeah. I thought so. We'll see how well people read the spec.
We've got several fillorder=2 modes in TiffImagePlugin.py
So, it appears that the layout is: 00 80 00 ... -> (128 , 0
...). The samples are stored in a single big bitian 12bit
block, but need to be pulled out to little endian format to be
stored in a 2 byte int.
There's no spec I can find. It appears that the in storage
layout is: 00 80 00 ... -> (128 , 0 ...). The samples are
stored in a single big bitian 12bit block, but need to be
pulled out to little endian format to be stored in a 2 byte
int.
*/
int i;
@ -1216,7 +1217,7 @@ static struct {
{"I;16L", "I;16N", 16, unpackI16N_I16}, // LibTiff native->image endian.
{"I;16B", "I;16N", 16, unpackI16N_I16B},
{"I;16", "I;12", 12, unpackI12_I16}, //
{"I;16", "I;12", 12, unpackI12_I16}, // 12 bit Tiffs stored in 16bits.
{NULL} /* sentinel */
};