Commit Graph

51 Commits

Author SHA1 Message Date
Oliver Tonnhofer
147a0204eb fix FASTOCTREE quantizer for large images (>16MP)
the color value counter sums all color values and this overflows
a uint32 at 16M white pixels, 32M gray pixels, etc.
2013-09-26 10:15:42 +02:00
wiredfool
ac1d43bdef 8 character tabs -> spaces 2013-07-10 22:21:49 -07:00
wiredfool
993e075c0d Add conversions from I;16 to F to retain 16 bit precision 2013-07-08 23:21:41 -07:00
wiredfool
c7264b5d4b 8 char tabs -> whitespace 2013-07-08 22:39:16 -07:00
Alexey Buzanov
df0e214d46 hide tiff warnings 2013-07-01 19:48:21 +04:00
Alex Clark
bb1b3a532c Cleanup WS, courtesy of @Arfrever
find * -type f "-(" -name "*.bdf" -o -name "*.c" -o -name "*.h" -o -name "*.py" -o -name "*.rst" -o -name "*.txt" "-)" -exec sed -e "s/[[:space:]]*$//" -i {} \;
2013-06-30 18:42:19 -04:00
Alex Clark ☺
aea9570a2c Merge pull request #221 from wiredfool/jpeg_memoryleak
Jpeg Decode memory leak fix.
2013-05-16 12:42:45 -07:00
wiredfool
f6381ba00c Added cleanup method to prevent memory leaks from truncated jpeg decodes 2013-05-15 21:16:37 -07:00
wiredfool
454a914b56 Whitespace, 8 space tabs -> 4 spaces 2013-05-15 19:57:45 -07:00
wiredfool
37359369ce Code to ensure buffer space for exif write and jpeg_write_scanline calls, tests, corrected buffer length calculation 2013-05-14 22:29:31 -07:00
wiredfool
11ce826121 mixed 8 char tabs & spaces -> spaces 2013-05-14 20:45:07 -07:00
Evert Rol
8e74b50e0f Fix issue #193: remove double typedef declaration. 2013-04-17 14:27:07 +02:00
Sandro Mani
e2c62ff1d2 Fix warning: pointer targets differ in signedness 2013-04-12 17:54:10 +02:00
Alex Clark ☺
21562761ac Merge pull request #179 from manisandro/quantization
Make quantization code more portable
2013-04-09 06:14:03 -07:00
Sandro Mani
07f632194d Make quantization code more portable 2013-04-08 00:52:15 +02:00
homm
ebb9029fbe add parentheses to avoid warnings 2013-03-27 23:19:04 +04:00
homm
a511ef8b05 move declarations to make ANSI compilers happy 2013-03-27 23:18:10 +04:00
homm
a83ef0f0ed ≈20% faster, 2 times more precisely (0.17% difference against 0.37%) 2013-03-26 01:15:18 +04:00
homm
af96b5d4b2 Precision improvement. No performance affected. 2013-03-25 00:06:02 +04:00
homm
0663e14444 speedup. divisions replaced by shifts. 2013-03-23 16:54:02 +04:00
homm
e88490b4f2 Precision improvement. ≈60% performance lost 2013-03-23 14:26:14 +04:00
homm
c39a51bcf0 Significant performance improvement of alpha_composite function. 2013-03-23 05:46:39 +04:00
Sandro Mani
b57c2ae0e9 Squash a memory leak 2013-03-18 00:54:54 +01:00
Alex Clark ☺
a6792ff7e4 Merge pull request #134 from cgohlke/patch-4
Disable msvc10 compiler optimizations for topalette and tobilevel
2013-03-15 05:08:14 -07:00
Christoph Gohlke
bb61c7633a Disable msvc10 compiler optimizations for topalette and tobilevel
Fixes issue #90
2013-03-14 23:35:02 -07:00
wiredfool
b5335c0cf3 Switching to strncpy, fixing potential buffer overflow in palette as well 2013-03-14 20:46:23 -07:00
wiredfool
66965b6bb4 fix buffer overflow for YCbCr histograms 2013-03-14 20:46:23 -07:00
wiredfool
d98abd877b merge to master 2013-03-14 14:38:49 -07:00
wiredfool
7d99141405 force seek to beginning of file 2013-03-14 13:25:23 -07:00
Christoph Gohlke
26c037a08d Disable debug prints in g4 tiff decoder 2013-03-14 09:03:18 -07:00
Alex Clark
990f5ddd15 Add contributors ; rename Tiff.h, fixes #103 2013-03-13 21:37:42 -04:00
Christoph Gohlke
9f09a9253c Fix compile error with msvc 2013-03-13 17:33:16 -07:00
wiredfool
e83dbaccce scaling back on redefinitions 2013-03-13 15:11:01 -07:00
wiredfool
35ce34a5fa enabling debug build to get better info on travis errors 2013-03-12 22:54:13 -07:00
wiredfool
7226d0dc26 Merge branch 'master' into g4_support 2013-03-12 22:24:36 -07:00
Alex Clark ☺
d6f597a1ad Merge pull request #95 from olt/pil-png-transp-backport
Backport PNG/ZIP improvements from PIL repo
2013-03-12 03:43:06 -07:00
Oliver Tonnhofer
d4a93754ee release GIL during Image.fill 2013-03-11 20:45:15 +01:00
Oliver Tonnhofer
6537ba19c3 backport PIL's PNG/Zip improvements
- add new FASTOCTREE quantizer with alpha support
- make ZIP compress level and type configurable
- support reading/writing PNGs with paletted alpha

source 3637439d51
2013-03-11 20:33:04 +01:00
wiredfool
5ba6564e26 initial merge of wiredfool/Python-Imaging-Library-G4-Tiff-Support 2013-03-08 19:51:59 -08:00
wiredfool
e782fe721e initial merge of wiredfool/Python-Imaging-Library-G4-Tiff-Support 2013-03-08 19:51:10 -08:00
Christoph Gohlke
cee97b83cc 64 bit proof ImagingNew 2013-03-08 11:03:43 -08:00
etienne
55f2dc7b0b Added quantization tables and subsampling options to the JPEG encoder. 2013-03-06 20:23:02 -05:00
Sandro Mani
416d5f70af Cast hash table values to unsigned long 2013-02-18 16:48:46 +01:00
Sandro Mani
8c7c373676 Fix buffer overflow
This patch was carried by fedora for python-imaging
See https://bugzilla.redhat.com/show_bug.cgi?id=703212
2013-02-06 00:51:25 +01:00
d-schmidt
addf0f4d95 Added possibility to save exif information in jpeg-files 2013-01-10 21:36:21 +01:00
Brian Crowell
4459715b6e py3k: Fix strict aliasing warnings under Python 3 2013-01-10 08:46:37 -06:00
Tom Payne
9d35947363 Avoid rounding error in Image.alpha_composite 2012-12-19 14:25:22 +01:00
Tom Payne
e414c66070 Add Image.alpha_composite 2012-12-04 16:44:26 +01:00
Lars Yencken
155347fe87 Fix a bug in quantize() causing PIL crashes.
From German Bravo's Image-SIG post:
http://mail.python.org/pipermail/image-sig/2012-June/007047.html
2012-09-05 13:23:57 +10:00
Matti Picus
2ba3bf681d UNDEF more types before including windows headers 2012-03-09 00:00:04 +02:00