c90b8be0b8
add tiffsave_buffer
...
add tests
also some cleanups to foreign include files
see https://github.com/jcupitt/libvips/issues/417
2016-10-15 12:29:14 +01:00
6d1d6fd16c
oops
2016-10-14 09:31:12 +01:00
ded6f17fa8
fix up new buffer.c
...
and make getpoint() use the threading system
2016-10-14 09:00:21 +01:00
b4d6d6c590
compiles, but untested
2016-10-13 14:57:18 +01:00
c5c3d48da7
version bump for new cycle
2016-10-13 11:43:46 +01:00
6a6c8cc5dd
notes
2016-10-12 17:43:25 +01:00
105c396de5
argh revert that
...
seem to all work now, but buffer.c is getting ugly ... revise for 8.5
2016-10-12 16:40:06 +01:00
cf98b92dbf
more tiny buffer fixups
2016-10-12 12:46:27 +01:00
463d2e885f
a bit more debugging code in buffer.c
...
watch global buffercache
2016-10-12 12:07:09 +01:00
c87991c9f0
notes
2016-10-12 10:32:28 +01:00
d042b0e88a
Merge branch 'master' into fix-threadshutdown-race
2016-10-11 14:13:23 +01:00
bc0bde85cb
remove a couple of statics
...
g_once() is fast enough
2016-10-11 14:12:45 +01:00
7883e30d68
group all non-worker buffers
...
have a single place to keep all non-worker buffers, and guard it with a
lock
2016-10-11 10:47:01 +01:00
ccfe4cb215
seems to work?
...
needs stress testing still
see https://github.com/jcupitt/libvips/issues/535
2016-10-10 15:12:12 +01:00
524c395f01
tiny doc fixes
2016-10-07 13:31:06 +01:00
ee0499af1f
tiny typos
2016-10-04 15:58:03 +01:00
dff0960e5f
fix error message for metadata get
...
if the type of the metadata item didn't match theh gtype you were
fetching to, the error message was wrong
2016-09-30 08:54:46 +01:00
db90f9f271
tiny conv improvements
...
- we were still using the im_aconv() wrapper, oops
- add gate and pixel count markup
2016-09-30 07:37:57 +01:00
b559a4dc00
more doc fixes
2016-09-27 12:29:18 +01:00
246089dc82
bump version
2016-09-27 11:23:57 +01:00
35892fa93c
docs polishing
...
see https://github.com/jcupitt/ruby-vips/issues/90
2016-09-27 07:53:33 +01:00
1cc3b2bd5d
final tweaks for 8.4 release
2016-09-23 12:26:12 +01:00
5b31408fd3
note on use of new_from_memory_copy
...
for py docs
see https://github.com/jcupitt/libvips/issues/346#issuecomment-248536958
2016-09-21 10:50:53 +01:00
eb91dda83c
simplify eval handling
...
put the "hide-progress" test into the signal dispatch
2016-09-19 17:40:35 +01:00
74089aa2e4
call _setmaxstdio() on win32
...
Windows has a limit of 512 files open at once for the fopen()
family of functions, and 2048 for the _open() family.
Call _setmaxstdio() in vips_init() to try to raise the stdio level to
match _open().
See https://github.com/jcupitt/libvips/issues/527
2016-09-18 12:46:19 +01:00
c88fa5ac73
openexr read outputs scRGB
2016-09-17 10:08:13 +01:00
3b3d4f401f
oh heh revert
2016-09-16 11:32:00 +01:00
71bcc526b0
try Vips.py with tabs
2016-09-16 11:24:04 +01:00
5b64db0aa0
revise example
2016-09-15 13:18:36 +01:00
a99b4cc23e
argh stupid unicode
2016-09-14 14:31:38 +01:00
4958adfb66
strange
2016-09-14 14:06:13 +01:00
48edc5fec1
iupdate libtool version number
2016-09-14 14:02:56 +01:00
fd9c598584
make marble tesselate
2016-09-14 11:19:48 +01:00
3723df0607
add marble example to whatsnew
2016-09-14 09:48:36 +01:00
e241d13339
better matrix header read
2016-09-13 22:56:16 +01:00
1872529b76
add whatsnew
2016-09-13 16:38:07 +01:00
b96335365e
switch to 3.5 bits of precision for vec convi
...
lets us execute
3 3 8 0
-1 -1 -1
-1 16 -1
-1 -1 -1
on the vector unit
2016-09-12 14:48:23 +01:00
707235ae61
doc improvements
2016-09-12 12:23:53 +01:00
85066e9d35
doc improvements
...
see https://github.com/jcupitt/libvips/issues/514
2016-09-12 11:20:11 +01:00
d4d41eef6a
doc improvement
...
see https://github.com/jcupitt/libvips/issues/521
2016-09-12 11:15:56 +01:00
eb320f41e1
add VIPS_TRACE env var
...
to enable --vips-cache-trace
2016-09-12 11:01:40 +01:00
d3cafb3a9f
seems to work
...
add @centre option for downsampling, see
https://github.com/jcupitt/libvips/issues/504
2016-09-10 10:31:33 +01:00
1e9cdef615
try centre sampling option
2016-09-09 14:51:45 +01:00
fc9322f71b
move concurrency_set into the public API
...
oops, it was off in threadpool and wasn't being scanned
2016-09-09 08:52:14 +01:00
b566bc6b3d
radsave was polluting the error log
2016-09-08 13:55:37 +01:00
591d4472a9
notes on vips-properties changes
2016-09-08 12:45:50 +01:00
bb29f80b44
Merge pull request #464 from felixbuenemann/dzsave-fix-vips-properties-duplicate-path
...
Fix dzsave vips-properties path for non-dz layout
2016-09-08 12:41:02 +01:00
d54df515b8
add some tests
2016-09-08 10:41:01 +01:00
4ba083437a
Merge branch 'master' into new-dzsave-overlaps
2016-09-08 09:42:34 +01:00
e601fad38b
small doc fixes
2016-09-07 21:26:59 +01:00
John Cupitt