8e6ed3ef30
compiles
2015-11-16 14:58:51 +00:00
d32bc97ccb
start hacking it in
2015-11-16 11:08:10 +00:00
3b32200cc1
sizealike() and friends elide completely
...
sizealike() / formatalike() and bandsalike() used to just vips_copy() if
the image didn't need any changes ... this was fast, but left a
vips_copy_gen() in the pipeline, wasting a lot of space on the C stack
during recursion.
They now vanish completely if the image is already in the right format.
Since we call them before most image processing operations, and
often all three of them, this saves a lot of C stack, more than x2 even
in simple cases.
There might also be a measureable CPU saving if the operations are very
simple (eg. insert).
See:
http://stackoverflow.com/questions/33658795/difficulty-with-handling-very-large-image-using-vips
2015-11-13 14:48:02 +00:00
d88304a297
sync
2015-11-12 09:15:31 +00:00
03a4e54649
slightly less stupid float->int mask
...
still not quite right
2015-11-11 16:51:08 +00:00
f32ab06a1c
fix test suite
2015-11-11 15:00:15 +00:00
4c51f5b36d
tune cache sizes a bit more
2015-11-11 13:00:25 +00:00
d94a2807c9
smarter cache sizing for vips_resize
...
see comment in code, should now account for max delay possible
2015-11-11 10:23:50 +00:00
66fe338d36
better alpha handling for tiff save
...
the tiff saver was writing all five-band images as CMYKA, even if they
were tagged as srgb ... it now follows the interpretation tag and will
write many alpha channels instead
see https://github.com/jcupitt/libvips/issues/348
thanks sadaqatullahn
2015-11-11 08:55:42 +00:00
aedb174537
note
2015-11-11 08:33:41 +00:00
ca51b46e9a
docs, c++, all done
2015-11-07 20:54:00 +00:00
295b18585a
python uses bandjoin_const when possible
...
added a test case, speedup seems worthwhile
2015-11-07 20:34:56 +00:00
9cb2674a77
now works woo
...
python interface plus test and benchmark to go
update c++ too I guess
oh docs as well sigh
2015-11-07 18:47:02 +00:00
3ed0904fab
fix travis
2015-11-06 15:56:13 +00:00
2f98f38d8a
started bandjoinconst
2015-11-06 14:58:32 +00:00
de65da85b9
smarter conversion of float masks to int
...
we were just rint()ing masks going from float -> int, but this could
give bad results in some cases, with output images becoming darker
or lighter, as individual elemets moved up or down
now we adjust the output scale to try to keep the overall brightness
the same, if we can
see: https://github.com/jcupitt/libvips/issues/344
2015-11-06 14:29:23 +00:00
d38a78d0ea
notes
2015-11-05 12:40:03 +00:00
82d96fb73e
notes
2015-11-04 21:19:31 +00:00
30a3d2f7fa
fix shrink with non-int args
...
fix a silly bug, add a test
2015-11-04 12:00:49 +00:00
d126a733a0
better rgb16/grey16 conversion
...
things like float tagged as rgb16 did not save well as png ... new
conversion rules should help
see https://github.com/jcupitt/libvips/issues/344
2015-11-04 11:34:52 +00:00
7f400da698
better Py docs
...
talks about new_from_memory etc.
2015-11-04 11:03:36 +00:00
b853012787
add memory.h to Python API
...
this makes the memory high water visible to Python, handy for testing
2015-11-04 10:09:06 +00:00
11ce13d43d
enable new shrink code
...
2x faster, much lower memuse
2015-10-31 12:15:48 +00:00
582a7f1810
shrink2 now done
...
big mem saving, still a bit slower though
2015-10-31 10:43:08 +00:00
da8f236f95
try a split shrink
...
not quite done yet
2015-10-30 18:15:14 +00:00
07ca974273
note on IM link problems
...
vips won't build against current imagemagick
seems to be a change in IM, see:
http://www.imagemagick.org/discourse-server/viewtopic.php?f=3&t=28603
see also:
https://github.com/jcupitt/libvips/issues/343
2015-10-30 14:53:35 +00:00
c27fce059d
Merge branch 'master' of github.com:jcupitt/libvips
2015-10-27 12:08:53 +00:00
d73d7e6e18
type
2015-10-27 12:08:43 +00:00
3b34308e6d
only allow [] to enclose options
...
we could pick up ( ) too easily before
see https://github.com/jcupitt/libvips/issues/342
2015-10-26 13:59:55 +00:00
9cad049051
notes
2015-10-26 11:06:57 +00:00
5160010eda
add vips_array_image_empty()/_append()
...
to help bindings without init from array
2015-10-25 17:15:45 +00:00
36617bd2e4
added vips_blob_copy()
...
helps ruby-vips8
2015-10-24 11:12:08 +01:00
2b46b7e889
many more const decls
...
ruby gobject-introspection is quite fussy about needing a lot of const
declarations ... these changes help vips_image_matrix_from_array()
appear in Ruby
2015-10-23 11:45:11 +01:00
649c30018e
add more links to README
2015-10-23 09:45:20 +01:00
64d081e726
return default value for unset props
...
we used to issue a warning and return early, but this can leave garbage
in the *value pointer, I think
ruby gobject-introspection will walk object props during GC and can see
state inbetween init and build when not all objects have been given a
value ... we don't want these warnings
2015-10-16 14:50:44 +01:00
faf7b70e2e
Merge branch '8.1'
...
Conflicts:
ChangeLog
2015-10-14 13:04:36 +01:00
4b25ce2cf2
note pathname parse fix
2015-10-14 13:00:41 +01:00
4d95f4512c
Merge branch '8.1'
...
Conflicts:
ChangeLog
configure.ac
doc/libvips-docs.xml
2015-10-13 14:47:37 +01:00
9a4d82653c
ignore ':' in col 1 of vips7 paths
...
oh argh, we were mangling windows paths
see https://github.com/jcupitt/build-win32/issues/11
2015-10-13 14:21:17 +01:00
368a74abcd
better vips7 filename splitting
...
helps windows, see https://github.com/jcupitt/build-win32/issues/11
also add some tests
2015-10-13 10:29:53 +01:00
2b2ac1b077
rename vips-8.0 as vips-8.1
...
also bump configure version
thanks Danilo!
2015-10-13 08:55:54 +01:00
b1cd4d1124
add doc note for operation.new
...
since language bindings use it
2015-10-13 07:28:58 +01:00
ceb1b96d06
note that quadratic is unfinished
2015-10-08 17:40:37 +01:00
3e16302701
bump version to 8.2 for the next cycle
2015-10-07 14:55:28 +01:00
b8a28f1a8d
add vips7 wrapper for magickload_buffer
...
see https://github.com/jcupitt/ruby-vips/issues/69
2015-10-07 14:48:32 +01:00
a7766b28b3
improve py docs
2015-10-07 12:20:22 +01:00
2b2bf30fbd
oops, update cpp binding
2015-10-06 22:45:37 +01:00
db291daf81
two small fixes for dzsave
...
fix a tiny leak, fix coloured @background
2015-10-06 13:30:25 +01:00
8f4dfa5cb6
Merge pull request #335 from PolarNick239/patch-1
...
acinlude.m4: python3 compatible print
2015-10-05 08:38:44 +01:00
b28c530979
acinlude.m4: python3 compatible print
...
Python3 and python2 compatible print call in configure script.
2015-10-05 01:55:58 +03:00
John Cupitt
Nickolay Polyarniy