John Cupitt
43288833b9
Merge pull request #810 from lovell/smartcrop-clamp-sigma
...
Smartcrop: clamp sigma to avoid out of range warnings
2017-11-20 17:41:33 +00:00
Lovell Fuller
caa0988256
Clamp minimum sigma to avoid out of range warnings
2017-11-19 20:22:19 +00:00
John Cupitt
a4b89aa370
credit for affine fixup
2017-11-18 15:11:02 +00:00
John Cupitt
65cfe2e7c0
Merge branch 'master' into fix-affine-jaggies
2017-11-18 15:10:50 +00:00
John Cupitt
2815647ce3
fix background for affine on alpha images
...
need to convert to ink for premultiplied space
2017-11-18 13:42:00 +00:00
John Cupitt
e179e3b11f
premultiply alpha (if necessary) in affine
2017-11-18 10:59:20 +00:00
John Cupitt
cd898c52de
Merge pull request #808 from lovell/prefer-sqrt-to-pow-0.5
...
Prefer ~10x faster sqrt instead of pow when y=0.5
2017-11-17 22:11:23 +00:00
Lovell Fuller
bb58685d1c
Use ~10x faster sqrt instead of pow when y=0.5
2017-11-17 20:24:28 +00:00
John Cupitt
fcec6d639b
fix jaggies on the edge of affine output
...
and add a "background" param
2017-11-17 16:30:25 +00:00
John Cupitt
ef9b23b95a
update function list
2017-11-09 15:00:13 +00:00
John Cupitt
2bb7844f4c
update docs
...
some old stuff in examples
2017-11-09 14:53:20 +00:00
John Cupitt
c01afbc559
update docs
...
the examples page was using the old python binding
2017-11-04 14:50:49 +00:00
John Cupitt
4ea743f5e6
add combine mode to find_indexed
2017-11-03 16:36:09 +00:00
John Cupitt
487c112807
add VIPS_COMBINE_MIN combining mode
...
for vips_compass()
2017-11-02 12:34:01 +00:00
John Cupitt
55e577a2fb
rename nearest as fill_nearest
...
avoid a clash with the nearest-neighbour interpolator
2017-11-01 15:06:37 +00:00
John Cupitt
cdc286e9d6
switch to float distance calculations
...
removes dithering effects along edges
2017-11-01 13:16:36 +00:00
John Cupitt
cb37089c9b
seems to work
2017-11-01 13:16:36 +00:00
John Cupitt
ae54f35ed1
start tinkering
2017-11-01 13:16:36 +00:00
John Cupitt
d90f2e649d
no error for bad page-height in tiffsave
...
just don't write a toilet-roll image
the error was confusing and easy to trigger by mistake
see https://github.com/jcupitt/libvips/issues/782
2017-10-24 09:39:58 +01:00
John Cupitt
3b75b1e2ac
fix gcc7 build
...
removed exception specs from deprecated C++ binding
2017-10-24 09:31:23 +01:00
John Cupitt
5ec1d5b72f
fix vipsthumbnail --size Nx
...
oh dear, width but not height size specs to vipsthumbnail were not working
thanks jrochkind
see https://github.com/jcupitt/libvips/issues/781
2017-10-23 22:27:23 +01:00
John Cupitt
990b68b54b
typos
2017-10-23 18:05:41 +01:00
John Cupitt
d99a189846
missing cast
2017-10-19 17:00:56 +01:00
John Cupitt
dd503c03c6
add vips_value_set_blob_free()
...
simpler for language bindings to call
2017-10-19 16:37:43 +01:00
John Cupitt
b060883706
default tiff predictor to horizontal
...
reduces default tiff lzw / deflate file size by ~20 - 30% at no performance
cost ... thanks @jrochkind
See https://github.com/jcupitt/libvips/issues/777
2017-10-19 08:50:16 +01:00
John Cupitt
6912fe91ed
sniff webp from the magic number
...
removes the 100 byte minimum file size
see https://github.com/lovell/sharp/issues/996
2017-10-18 09:00:34 +01:00
John Cupitt
a8bbe05596
tiny optimisations
2017-10-16 20:55:58 +01:00
John Cupitt
99f02aeff1
allow approx gaussmat
...
works with gaussblur now
2017-10-16 17:10:16 +01:00
John Cupitt
5290901f86
ban gcc 5.x from vector stuff too
...
it seems to be easy to get internal compiler errors, thanks homm
see https://github.com/jcupitt/pyvips/issues/21
2017-10-16 17:08:07 +01:00
John Cupitt
0767e4f4c9
remove a 0.5 pixel shift from triangle
...
lines up with cubic and lanczos now
2017-10-16 09:38:04 +01:00
John Cupitt
53119250cc
fix a crash in gifload with a missing colormap
...
it seems gifs can have a missing colormap, who knew
thanks Kleis
see https://github.com/jcupitt/libvips/issues/773
2017-10-15 22:35:52 +01:00
John Cupitt
68ed42e2fa
make linear and cubic adaptive
2017-10-15 22:09:44 +01:00
John Cupitt
cfa96f735a
only read ifd0 orientation
...
we used to set orientation from the first ifd, now just read ifd0, since
that's where the main image orientation should always be
see https://github.com/jcupitt/libvips/issues/772
2017-10-14 21:54:31 +01:00
John Cupitt
008fd21728
all done
2017-10-14 17:03:48 +01:00
John Cupitt
36f21138d7
builds half float mask
2017-10-14 13:46:51 +01:00
John Cupitt
ec99ff5000
compiles
2017-10-14 12:38:48 +01:00
John Cupitt
0b3d7fc185
start half-float experiment
2017-10-13 18:06:58 +01:00
John Cupitt
cf1e6e3ef8
oh dear, missing breaks in switch
...
too much ruby sigh
2017-10-13 12:16:03 +01:00
John Cupitt
c76b4893c0
convi was leaking vectors
...
we forgot to free the compiled code on dispose, oops
thanks MHeimbuc
see https://github.com/jcupitt/libvips/issues/771
2017-10-12 14:41:21 +01:00
John Cupitt
135d7a4dfb
Merge branch '8.5'
2017-10-11 12:57:18 +01:00
John Cupitt
b4a6e6e309
spelling
2017-10-11 11:24:43 +01:00
John Cupitt
fd1d332892
update change notes
2017-10-11 10:12:01 +01:00
John Cupitt
bfc339facb
make vips_image_write() sever connections
...
when writing to a non-partial image, vips_image_write() now carefully severs
all connections between the two images on completion
this fixes a couple of cases where we had bad behaviour: writing to a
temp file could leave dangling pointers, see
https://github.com/jcupitt/libvips/issues/708
and writing to a memory buffer during copy_memory coud leave dangling
pointers too, see:
https://github.com/jcupitt/ruby-vips/issues/140
2017-10-11 10:02:42 +01:00
John Cupitt
e96fd25a9f
more icc input profile sanity tests
...
check colourspace too, so we won't try to import an RGB image with a LAB
profile
see https://github.com/jcupitt/libvips/issues/730
2017-10-10 14:23:05 +01:00
John Cupitt
9e0ff87dd5
add docs for autofit_dpi
2017-10-10 13:10:17 +01:00
John Cupitt
40f20e5e0a
Merge branch 'master' into gargsms-text-autofit
2017-10-10 13:04:31 +01:00
John Cupitt
12db3e936c
Merge branch '8.5'
2017-10-09 22:35:29 +01:00
John Cupitt
6f16a9df71
fix a crash with heavy use of draw
...
we were leaving a few bits of the original image around after
vips_image_copy_memory(), including the progress signal ... if earlier images
were freed, perhaps by a GC in a language binding, we could get a dangling
pointer
thanks Nakilon, see https://github.com/jcupitt/ruby-vips/issues/140
2017-10-09 22:27:36 +01:00
John Cupitt
37a435ec48
missing - in travis.yml
2017-10-09 12:22:32 +01:00
John Cupitt
d9017b0cf4
let os x fail again
...
since homebrew is now requiring ruby2.3
2017-10-09 12:19:32 +01:00
