Commit Graph

1115 Commits

Author SHA1 Message Date
John Cupitt
09b81b7580 im_maxpos_avg() avoids NaN 2015-03-26 17:16:52 +00:00
John Cupitt
9552315e80 version bump 2015-03-26 08:46:52 +00:00
John Cupitt
4d0887f5de fix .dz as a target
and add a test for this as well
2015-02-13 11:17:55 +00:00
John Cupitt
5b90e36559 notes 2015-02-10 14:44:35 +00:00
John Cupitt
346a9e70c0 remove --disable-cxx flag
it was getting too hard to support :( automake now generates libtool
lines before configure runs, so it can't know we don't want a C++ link

see https://github.com/jcupitt/libvips/issues/231
2015-02-07 10:37:13 +00:00
John Cupitt
3d902d2feb improvements to python configure 2015-02-06 14:54:52 +00:00
John Cupitt
9d11c02e35 version bump 2015-02-06 13:39:39 +00:00
John Cupitt
d6dbfea064 much faster RGB16 -> sRGB path 2015-02-04 16:48:25 +00:00
John Cupitt
0c466b9495 don't use isnormal() to test for crazy FP numbers
it fails for 0.0 and for subnormal numbers, thanks Murat Korkmazov
2015-02-04 15:13:11 +00:00
John Cupitt
e429772ae8 turn on leak testing in python test suite 2015-02-02 11:36:29 +00:00
John Cupitt
eb5fee5e65 add lower-level cache access 2015-02-01 17:56:48 +00:00
John Cupitt
ce557ba6de add VInterpolate wrapper class
to cplusplus, see https://github.com/jcupitt/libvips/issues/230
2015-01-30 21:19:58 +00:00
John Cupitt
e8f456a7e2 more tests 2015-01-27 21:45:14 +00:00
John Cupitt
70b60e1dd5 tiny cleanup 2015-01-22 13:19:21 +00:00
John Cupitt
229636fc66 rename vipsthumbnail -o as -f
-o stays as a hidden flag

see https://github.com/jcupitt/libvips/issues/216
2015-01-20 09:10:36 +00:00
John Cupitt
23071d73ac Merge branch 'master' of github.com:jcupitt/libvips
Conflicts:
	ChangeLog
2015-01-19 22:03:05 +00:00
John Cupitt
dbd98ba86c disable chroma subsample in jpeg-tiff if Q>=90
see https://github.com/jcupitt/libvips/issues/218
2015-01-19 22:01:13 +00:00
John Cupitt
f1d901a229 try to handle tiffs with 8-bit colormaps
see

https://github.com/jcupitt/libvips/issues/220
2015-01-19 14:52:57 +00:00
John Cupitt
c805b382fe vipsthumbnail sets an error on failure
conversion errors are ignored as before, but on final exit vipsthumbnail will
return an error code if one or more of the conversions failed

see https://github.com/jcupitt/libvips/issues/219
2015-01-13 17:31:09 +00:00
John Cupitt
eda4ed64f1 rename _!PIXEL etc. to help bindings
ruby gobject-introspection hates numbers at the start of _ separated
names
2015-01-07 14:22:13 +00:00
John Cupitt
6f101ae86b remove VImage::scale()
it clashed with the VipsScale operator, see

https://github.com/jcupitt/libvips/pull/213
2015-01-06 12:00:11 +00:00
John Cupitt
6440830315 required C++ output params were broken
see https://github.com/jcupitt/libvips/pull/213

this now works:

        VImage rows;
	VImage cols = in.profile( &rows );
2015-01-06 11:51:53 +00:00
John Cupitt
90725b9f5e more tests 2015-01-05 13:48:35 +00:00
John Cupitt
6c35ae5540 display default and range in usage
eg. we now see:

$ vips gamma
gamma an image
usage:
   gamma in out
where:
   in           - Input image, input VipsImage
   out          - Output image, output VipsImage
optional arguments:
   exponent     - Gamma factor, input gdouble
  			default: 2.4
			min: 1e-06, max: 1000
operation flags: sequential-unbuffered
2014-12-30 13:16:22 +00:00
John Cupitt
32625146db cpp can set enums from strings
eg.

	set("extend", "copy")->
2014-12-30 11:27:04 +00:00
John Cupitt
d077d7c1fc small stuff 2014-12-27 10:00:01 +00:00
John Cupitt
b01dd59309 fix VipsBlob read
oops, forgot the .get()
2014-12-27 09:41:01 +00:00
John Cupitt
366e7e3a5e add gobject to Requires: line
vips C and C++ programs can link directly against things in gobject
without appearing to, for example:

	a = VIPS_OBJECT( b );

the macro will expand to include a g_type_check_instance_is_a(). So we
have to have gobject-2.0 in the link line of things like:

	g++ -g -Wall hello8.cc `pkg-config vips-cpp --cflags --libs`

see also

	https://github.com/jcupitt/libvips/issues/196
2014-12-24 10:10:14 +00:00
John Cupitt
1b86de44f4 "make check" works uninstalled
and skips tests for missing features
2014-12-22 10:35:26 +00:00
John Cupitt
abdd98409d add configure option for pyvips8
it tests for pygobject as well
2014-12-18 09:15:58 +00:00
John Cupitt
6e2135d8b5 mask builders take prec, not integer
vips_gaussmat() and vips_logmat() used to default to float, but
vips_conv() defaulted to int ... so this failed:

	vips gaussmat m.v 2 0.1 --separable
	vips convsep k2.jpg x.jpg m.v

since we were using a small float mask with an integer convolution

This change makes everything int by default, with gaussmat having a new
param called --precision whose arg can be int or float.

The old @integer param is deprecated.
2014-12-16 20:48:11 +00:00
John Cupitt
c3102b15bb add python3 support 2014-12-10 17:02:51 +00:00
John Cupitt
76653e9e78 add notes to logs for @density 2014-12-04 21:47:46 +00:00
John Cupitt
f8cca8e59d load and save XMP metadata in tiff
see https://github.com/jcupitt/libvips/issues/198
2014-12-03 09:03:09 +00:00
John Cupitt
ffe90ef38a add .szi as a synonym for .zip
in dzsave
2014-11-26 14:26:57 +00:00
John Cupitt
4c3461967b duuh fix up misplaced patch 2014-11-25 12:05:04 +00:00
John Cupitt
6bfe9b45b7 Merge branch '7.40' 2014-11-25 12:04:24 +00:00
John Cupitt
f66afb78da fix profile handling in XYZ2scRGB
see

https://github.com/lovell/sharp/issues/125
2014-11-25 12:01:13 +00:00
John Cupitt
4b51a9d61f don't strip profiles in XYZ2scRGB
see

https://github.com/lovell/sharp/issues/125
2014-11-25 11:43:48 +00:00
John Cupitt
384a5e05dc change parameters of vips_gaussblur()
sorry :-(

The old param (passing sigma*2 as an int) were just too confusing
coming from other packages. This new behaviour matches ImageMagick, and
almost everyone else in the world.

A new optional param lets you set the minimum amplitude as well.
2014-11-19 12:27:26 +00:00
John Cupitt
f16e252b3e Merge branch '7.40' 2014-11-11 20:43:19 +00:00
John Cupitt
8a44d6f7b3 fix memleak in sharpen
see https://github.com/jcupitt/libvips/issues/194
2014-11-11 20:42:26 +00:00
John Cupitt
cce90b4fba add pbm save
previously vips could load but not save pbm (one bit) images

fix a few bugs in float load and save as well, and improve one-bit load
2014-11-08 17:40:43 +00:00
John Cupitt
cfcd819246 better default resolution for png load
see

https://github.com/lovell/sharp/issues/96
2014-11-04 09:38:21 +00:00
John Cupitt
63469b1c9d merge new C++ API 2014-11-03 09:07:28 +00:00
John Cupitt
c56e23e3c2 Merge remote-tracking branch 'origin/7.40'
Conflicts:
	ChangeLog
	configure.ac
2014-10-27 13:47:17 +00:00
John Cupitt
965936307b better bigtiff detection
update Changelog and bump version
2014-10-27 13:45:45 +00:00
John Cupitt
620bff2d78 allow symbolic names for flags
you can now use numbers or names for GFlags values, eg.:

$ vips copy 50020484-00001.png x.png[filter=avg]
$ vips copy 50020484-00001.png x.png[filter=64]

was numbers only before
2014-10-27 11:40:43 +00:00
John Cupitt
ad86f45cd9 add autorot as an operator as well 2014-10-19 10:35:10 +01:00
John Cupitt
0a51536162 seems to work 2014-10-17 13:17:44 +01:00