babl.git
8 years agosse2-float: use babl_linear_to_gamma_2_2f() (instead of double)
Ell [Thu, 11 Jan 2018 14:22:46 +0000 (09:22 -0500)]
sse2-float: use babl_linear_to_gamma_2_2f() (instead of double)

Use the single-precision babl_linear_to_gamma_2_2f() function,
instead of the double-precision version, when converting "leftover"
samples, so that the result is consistent with the vectorized
samples.

8 years agoutil, sse2-float: make sure 1 maps to 1 in linear -> gamma conversions
Ell [Thu, 11 Jan 2018 14:19:50 +0000 (09:19 -0500)]
util, sse2-float: make sure 1 maps to 1 in linear -> gamma conversions

Add a small offset to the result of the linear -> gamma
conversions, such that an input value of 1.0 maps to an output
value of 1.0, exactly.

8 years agobabl: shuffle some dispatch functions to be inlineable
Øyvind Kolås [Sun, 7 Jan 2018 04:37:39 +0000 (05:37 +0100)]
babl: shuffle some dispatch functions to be inlineable

8 years agobabl: strip out some internal length returning code
Øyvind Kolås [Sat, 6 Jan 2018 23:42:33 +0000 (00:42 +0100)]
babl: strip out some internal length returning code

8 years agobabl: prepare babl_fish_process for direct calling
Øyvind Kolås [Sat, 6 Jan 2018 23:35:22 +0000 (00:35 +0100)]
babl: prepare babl_fish_process for direct calling

8 years agopalette: expect palettes/formats to be gamma corrected in 8bit
Øyvind Kolås [Wed, 3 Jan 2018 16:12:41 +0000 (17:12 +0100)]
palette: expect palettes/formats to be gamma corrected in 8bit

A correctness correction related to bug #763581 the formats wanted
by GIMP and expected are gamma corrected not linear, I expect this
fix either does not much for GIMP or fixes an unknown 8bit linear
vs gamma quantization problem that has been lurking in INDEXED mode.

8 years agobabl: improve babl_trc_formula_srgb sRGB detection constants
Øyvind Kolås [Tue, 2 Jan 2018 20:05:02 +0000 (21:05 +0100)]
babl: improve babl_trc_formula_srgb sRGB detection constants

For bug #791816. Use constants as passed from GIMP when its built in sRGB ICC
profile is passed.

8 years agoCIE: Use a faster cbrtf implementation
Debarshi Ray [Thu, 21 Dec 2017 09:14:53 +0000 (10:14 +0100)]
CIE: Use a faster cbrtf implementation

This is the approximate cube root of an IEEE float implementation from
Hacker's Delight. The elimination of all conditional branches probably
makes it a better candidate for future SIMD accelerated code paths.

On an Intel i7 Haswell, it now takes 0.27s to convert a 15 megapixel
buffer from "RGBA float" to "CIE Lab alpha float" instead of the
earlier 0.35s. A "Y float" to "CIE L float" conversion takes 0.085s
instead of 0.102s.

Original code: http://www.hackersdelight.org/hdcodetxt/acbrt.c.txt
Permissions: http://www.hackersdelight.org/permissions.htm

https://bugzilla.gnome.org/show_bug.cgi?id=791837

8 years agoextensions-gggl: clamp when converting float to 16bit
Øyvind Kolås [Wed, 20 Dec 2017 02:22:22 +0000 (03:22 +0100)]
extensions-gggl: clamp when converting float to 16bit

Improve the quality of a conversion that was picked in error for bug #790658

8 years agobabl: re-enable earlier bailing for big path errors
Øyvind Kolås [Wed, 20 Dec 2017 02:22:00 +0000 (03:22 +0100)]
babl: re-enable earlier bailing for big path errors

This is one way of dealing with 790658 - that in general is a good idea as
well.

8 years agomeson.build: babl's license is LGPL3+
Øyvind Kolås [Tue, 19 Dec 2017 01:13:26 +0000 (02:13 +0100)]
meson.build: babl's license is LGPL3+

8 years agoSimpler library handling. PKGBUILD+release, lto.
Félix Piédallu [Fri, 15 Dec 2017 09:54:49 +0000 (10:54 +0100)]
Simpler library handling. PKGBUILD+release, lto.

8 years agofix check version
Félix Piédallu [Thu, 14 Dec 2017 14:52:57 +0000 (15:52 +0100)]
fix check version

8 years agoFix library name
Félix Piédallu [Thu, 14 Dec 2017 14:52:47 +0000 (15:52 +0100)]
Fix library name

8 years agoAdded PKGBUILD for archlinux, based on AUR.
Félix Piédallu [Thu, 14 Dec 2017 14:12:25 +0000 (15:12 +0100)]
Added PKGBUILD for archlinux, based on AUR.

8 years agoRemove useless join_paths arg
Félix Piédallu [Wed, 6 Dec 2017 14:22:41 +0000 (15:22 +0100)]
Remove useless join_paths arg

8 years agouse find_program. Use env for multiple commands as it allows to split into string...
Félix Piédallu [Sat, 2 Dec 2017 00:03:56 +0000 (01:03 +0100)]
use find_program. Use env for multiple commands as it allows to split into string list.

8 years agouse env instead of bash
Félix Piédallu [Fri, 1 Dec 2017 23:50:41 +0000 (00:50 +0100)]
use env instead of bash

8 years agoUse formatting instead of weird path reconstruction
Félix Piédallu [Fri, 1 Dec 2017 14:57:41 +0000 (15:57 +0100)]
Use formatting instead of weird path reconstruction

8 years agoDispatch the custom_target in two custom_targets for cleaner dependency declaration.
Félix Piédallu [Fri, 1 Dec 2017 14:31:43 +0000 (15:31 +0100)]
Dispatch the custom_target in two custom_targets for cleaner dependency declaration.

8 years agoFix push_web, add index_html generation. TODO: simplify this someday
Félix Piédallu [Fri, 1 Dec 2017 13:47:10 +0000 (14:47 +0100)]
Fix push_web, add index_html generation. TODO: simplify this someday

8 years agoAdd linker script for symbols export
Félix Piédallu [Tue, 28 Nov 2017 14:53:04 +0000 (15:53 +0100)]
Add linker script for symbols export

8 years agosome small cleanup, export script for scp
Félix Piédallu [Sun, 26 Nov 2017 12:51:31 +0000 (13:51 +0100)]
some small cleanup, export script for scp

8 years agoadd docs. fix install libs. Seems finished.~
Félix Piédallu [Thu, 23 Nov 2017 20:43:16 +0000 (21:43 +0100)]
add docs. fix install libs. Seems finished.~

8 years agoadd extensions, tools
Félix Piédallu [Thu, 23 Nov 2017 17:18:56 +0000 (18:18 +0100)]
add extensions, tools

8 years agoAdd tests
Félix Piédallu [Thu, 23 Nov 2017 17:12:09 +0000 (18:12 +0100)]
Add tests

8 years agoFix build, links.
Félix Piédallu [Thu, 23 Nov 2017 17:01:10 +0000 (18:01 +0100)]
Fix build, links.

8 years agoAdded basic meson files
Félix Piédallu [Thu, 23 Nov 2017 16:58:37 +0000 (17:58 +0100)]
Added basic meson files

8 years agobabl: make conversion to model double more generic
Øyvind Kolås [Sat, 18 Nov 2017 00:35:09 +0000 (01:35 +0100)]
babl: make conversion to model double more generic

Iterate over the components of the model, and fetch the relevant
component from the source format or use a filler value if not found.

Some changes by Debarshi Ray.

https://bugzilla.gnome.org/show_bug.cgi?id=790860

8 years agoCIE: refactor away some intermediate variables/assignments
Øyvind Kolås [Sat, 9 Dec 2017 22:38:38 +0000 (23:38 +0100)]
CIE: refactor away some intermediate variables/assignments

8 years agoCIE: add CIE XYZ and CIE XYZ alpha models
Øyvind Kolås [Sat, 9 Dec 2017 19:51:38 +0000 (20:51 +0100)]
CIE: add CIE XYZ and CIE XYZ alpha models

Even more useful now with custom RGB primaries - and being able to use babl for
conversions when needed is convenient. This is with reference paths - which
means fast paths for double precision float - single precision would have to be
added for it to be used and higher throughput in most common scenarios.

8 years agotests: not all UNIX-like OSes have a libpthread.
Jehan [Mon, 4 Dec 2017 02:33:47 +0000 (03:33 +0100)]
tests: not all UNIX-like OSes have a libpthread.

In particular Android systems don't need to link with -lpthread
(actually the link would fail with "ld: cannot find -lpthread").
Use the $(THREAD_LIB) variable which is set correctly during configure
since commit c02af82.

8 years agoCIE: Simplify code
Debarshi Ray [Sat, 18 Nov 2017 01:09:03 +0000 (02:09 +0100)]
CIE: Simplify code

Exploit the fact that fy is not needed to calculate the a and b
components. This is consistent with the other monochrome conversions.

8 years agoconfigure.ac: post release version bump
Øyvind Kolås [Wed, 15 Nov 2017 22:18:20 +0000 (23:18 +0100)]
configure.ac: post release version bump

8 years agoconfigure.ac: release 0.1.38
Øyvind Kolås [Wed, 15 Nov 2017 22:16:36 +0000 (23:16 +0100)]
configure.ac: release 0.1.38

8 years agoNEWS: update
Øyvind Kolås [Wed, 15 Nov 2017 22:15:25 +0000 (23:15 +0100)]
NEWS: update

8 years agobabl_pow_24f(): add missing float suffix
Edward E [Tue, 14 Nov 2017 04:43:29 +0000 (22:43 -0600)]
babl_pow_24f(): add missing float suffix

8 years agoCIE: Add "CIE L float"
Debarshi Ray [Thu, 9 Nov 2017 06:48:26 +0000 (07:48 +0100)]
CIE: Add "CIE L float"

Some of these conversions will be leveraged by gegl:shadows-highlights
which needs to go from "Y float" or "YaA float" to "CIE L float".

The conversion from "RGBA float" was added to aid "YaA float" to
"CIE L float" fishes. They go via:
  "YaA float"     to "RaGaBaA float"
  "RaGaBaA float" to "RGBA float"
  "RGBA float"    to "Y float"
  "Y float"       to "CIE L float"

A direct conversion from "YaA float" to "Y float" in simple C is
hindered by the need to check every pixel's alpha value to avoid
dividing by zero. The pipeline stalls make it lose out to the look-up
table and SIMD based RGB conversions to unassociated alpha.

However, we can trivially cut out the third step and still reduce some
memory traffic.

https://bugzilla.gnome.org/show_bug.cgi?id=790111

8 years agobabl: fix memory leaks on error paths
Tobias Stoeckmann [Fri, 13 Oct 2017 16:32:52 +0000 (18:32 +0200)]
babl: fix memory leaks on error paths

The database-focused functions of babl-cache are prone to
memory leaks when errors occur.

Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
8 years agoextensions: protect another conversion against /0.0
Øyvind Kolås [Fri, 10 Nov 2017 09:44:21 +0000 (10:44 +0100)]
extensions: protect another conversion against /0.0

8 years agoextensions: protect gray unpremuls from gggl against division by 0.0
Øyvind Kolås [Fri, 10 Nov 2017 09:34:57 +0000 (10:34 +0100)]
extensions: protect gray unpremuls from gggl against division by 0.0

8 years agoconfigure.ac: post release version bump
Øyvind Kolås [Fri, 10 Nov 2017 00:08:10 +0000 (01:08 +0100)]
configure.ac: post release version bump

8 years agorelease: 0.1.36
Øyvind Kolås [Fri, 10 Nov 2017 00:05:32 +0000 (01:05 +0100)]
release: 0.1.36

8 years agoNEWS: update for 0.1.36
Øyvind Kolås [Fri, 10 Nov 2017 00:03:37 +0000 (01:03 +0100)]
NEWS: update for 0.1.36

8 years agoCIE: Make the Lab to RGB fast paths work with non-sRGB primaries
Debarshi Ray [Tue, 7 Nov 2017 09:52:44 +0000 (10:52 +0100)]
CIE: Make the Lab to RGB fast paths work with non-sRGB primaries

This is similar to commit 72a074df6545fcb3.

https://bugzilla.gnome.org/show_bug.cgi?id=790011

8 years agoCIE: Make the Lab to RGB fast paths as accurate as the reference
Debarshi Ray [Tue, 7 Nov 2017 09:38:12 +0000 (10:38 +0100)]
CIE: Make the Lab to RGB fast paths as accurate as the reference

This is similar to commit e17807f235557d5d.

https://bugzilla.gnome.org/show_bug.cgi?id=790011

8 years agobabl, sse2-float: fall back to slow, accurate path for large pow-2.4 inputs
Ell [Tue, 7 Nov 2017 20:44:39 +0000 (15:44 -0500)]
babl, sse2-float: fall back to slow, accurate path for large pow-2.4 inputs

The approximations we use for pow_24() and pow_1_24() diverge from
the actual function for large-enough input values.  This can lead
not just to inaccurate results, but also to infinities and NaNs,
especially when multiple conversions are strung in a row.

When the input value is large enough to produce notable divergence
(the difference between the approximate and actual values is ~1% at
the chosen limits,) fall back to a slower, but more accurate
version.

For the SSE2 float conversions, this results in an increase of ~5%
in conversion time, when all values are below the limit.  When most
values are above the limit, performance can be 10x slower or worse.

8 years agoexport-symbols: add babl_format_exists
Øyvind Kolås [Tue, 31 Oct 2017 19:48:23 +0000 (20:48 +0100)]
export-symbols: add babl_format_exists

8 years agobabl: add babl_format_exists
Øyvind Kolås [Tue, 31 Oct 2017 19:48:23 +0000 (20:48 +0100)]
babl: add babl_format_exists

A function to check if a specific format - by name - is already known by babl.
Should be used in conjunction with babl_format() for user supplied format
strings.

8 years agoCIE: Make the RGB to Lab fast paths work with non-sRGB primaries
Debarshi Ray [Tue, 31 Oct 2017 11:20:45 +0000 (12:20 +0100)]
CIE: Make the RGB to Lab fast paths work with non-sRGB primaries

The older constants were tied to sRGB primaries. Using the matrices
from the Babl space removes this restriction.

https://bugzilla.gnome.org/show_bug.cgi?id=789695

8 years agoCIE: Make the RGB to Lab fast paths as accurate as the reference
Debarshi Ray [Tue, 31 Oct 2017 08:43:53 +0000 (09:43 +0100)]
CIE: Make the RGB to Lab fast paths as accurate as the reference

The matrices used by the fast paths to convert from RGB to XYZ didn't
exactly match the reference matrices in the Babl space. This caused a
measurable error in their output.

https://bugzilla.gnome.org/show_bug.cgi?id=789695

8 years agoCIE: Use cbrt instead of pow for the reference XYZ to LAB conversion
Debarshi Ray [Tue, 31 Oct 2017 08:36:39 +0000 (09:36 +0100)]
CIE: Use cbrt instead of pow for the reference XYZ to LAB conversion

The fast-paths use an inlining-friendly version of cbrt(3). Using
something similar removes superficial differences between the two
conversion paths. It's not like the C library's cbrt(3) will perform
any worse than its own pow(3).

https://bugzilla.gnome.org/show_bug.cgi?id=789695

8 years agoCIE: Add conversion from "RGBA float" to "CIE Lab float"
Debarshi Ray [Sun, 29 Oct 2017 12:52:26 +0000 (13:52 +0100)]
CIE: Add conversion from "RGBA float" to "CIE Lab float"

Conversions from "RaGaBaA float" to "CIE Lab float", as seen when
using gegl:shadows-highlights" go via:
  "RaGaBaA float" to "RGBA float"
  "RGBA float"    to "RGB float"
  "RGB float"     to "CIE Lab float"

A direct conversion from "RaGaBaA float" to "CIE Lab float" in simple
C is hindered by the need to check every pixel's alpha value to avoid
dividing by zero. The pipeline stalls make it lose out to the look-up
table and SIMD based conversions to unassociated alpha.

However, we can trivially cut out the second step and still reduce
some memory traffic.

https://bugzilla.gnome.org/show_bug.cgi?id=789695

8 years agobabl: use snprintf instead of sprintf
Tobias Stoeckmann [Tue, 24 Oct 2017 18:02:15 +0000 (20:02 +0200)]
babl: use snprintf instead of sprintf

Using sprintf with environment variables is dangerous, because it can
easily lead to out of boundary writes on heap space.

While at it, replace sprintf calls with snprintf where proper
boundary checks are possible and required.

Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
8 years agobabl: properly handle large files on 32 bit systems
Tobias Stoeckmann [Fri, 13 Oct 2017 16:39:02 +0000 (18:39 +0200)]
babl: properly handle large files on 32 bit systems

If large file support is enabled on 32 bit systems, it is possible
to trigger an out of boundary write with files larger than 2 GB.

Always check if fseek and ftell are successful and if the file is
small enough to fit into memory.

Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
8 years agobabl: always end strncpy strings with NUL
Tobias Stoeckmann [Fri, 13 Oct 2017 17:25:01 +0000 (19:25 +0200)]
babl: always end strncpy strings with NUL

The function strncpy(3) does not guarantee to end the destination string
with NUL character if not enough space was available. This could happen
on systems which allow paths which are longer than 4096 characters.

Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
8 years agopost release version bump
Øyvind Kolås [Fri, 6 Oct 2017 13:16:25 +0000 (15:16 +0200)]
post release version bump

8 years agorelease 0.1.34, update NEWS
Øyvind Kolås [Fri, 6 Oct 2017 13:15:57 +0000 (15:15 +0200)]
release 0.1.34, update NEWS

8 years agoexport-symbols: add babl_backtrack for gimp-2.8
Øyvind Kolås [Thu, 5 Oct 2017 21:12:57 +0000 (23:12 +0200)]
export-symbols: add babl_backtrack for gimp-2.8

8 years agobabl: rearrange BablModel struct
Øyvind Kolås [Wed, 4 Oct 2017 14:47:39 +0000 (16:47 +0200)]
babl: rearrange BablModel struct

Babl is doing casting that relies on BablFormat / BablModel having the same
layout for shared members.

8 years agobabl: do not use model.data for storing original model backpointer
Øyvind Kolås [Wed, 4 Oct 2017 14:16:45 +0000 (16:16 +0200)]
babl: do not use model.data for storing original model backpointer

Fixing bug #788471

8 years agoupdate NEWS
Øyvind Kolås [Tue, 3 Oct 2017 15:31:11 +0000 (17:31 +0200)]
update NEWS

8 years agopost-release version bump
Øyvind Kolås [Tue, 3 Oct 2017 15:17:00 +0000 (17:17 +0200)]
post-release version bump

8 years agorelease 0.1.32
Øyvind Kolås [Tue, 3 Oct 2017 15:11:44 +0000 (17:11 +0200)]
release 0.1.32

8 years agodocs: update with api changes
Øyvind Kolås [Tue, 3 Oct 2017 15:14:02 +0000 (17:14 +0200)]
docs: update with api changes

8 years agobabl: acknowledge some used trademarks
Øyvind Kolås [Tue, 3 Oct 2017 15:06:06 +0000 (17:06 +0200)]
babl: acknowledge some used trademarks

8 years agobabl: move to/from xyz and rgbtoxyz matrix functions out of babl.h
Øyvind Kolås [Tue, 3 Oct 2017 14:41:26 +0000 (16:41 +0200)]
babl: move to/from xyz and rgbtoxyz matrix functions out of babl.h

Both of these are useful, but we might want a wider range, permitting
both linear and non-linear variants to be converted to both XYZ and LAB.

The matrix is for now used in babl for the RGB->Y conversion - this might
warrant a more direct API.

8 years agobabl: rename babl_space_from_icc to babl_icc_make_space
Øyvind Kolås [Tue, 3 Oct 2017 13:54:19 +0000 (15:54 +0200)]
babl: rename babl_space_from_icc to babl_icc_make_space

Thus being more consistent that functions that take icc_data + length have the
prefix babl_icc_, since each icc space in the future potentially can have more
spaces than the relative-colorimetric space defined, _make_ seems better than
_to_ or _new_.

8 years agobabl: fix clang warnings in babl-polynomial
Ell [Tue, 3 Oct 2017 14:01:58 +0000 (10:01 -0400)]
babl: fix clang warnings in babl-polynomial

Cast the input polynomial from BablPolynomial* to
BablBigPolynomial* in the evaluator functions, so that clang
doesn't warn us about out-of-bounds access to its coefficient
array.

8 years agoHSL: make hue be set in all switch cases for rgb_to_hsl
Øyvind Kolås [Tue, 3 Oct 2017 13:53:23 +0000 (15:53 +0200)]
HSL: make hue be set in all switch cases for rgb_to_hsl

8 years agobabl: revert and clarify libtool flag for win32 builds
Edward E [Sat, 30 Sep 2017 19:32:49 +0000 (14:32 -0500)]
babl: revert and clarify libtool flag for win32 builds

8 years agobabl: improve alpha calc. in pal+alpha -> rgba conversion
Ell [Sat, 30 Sep 2017 11:44:41 +0000 (07:44 -0400)]
babl: improve alpha calc. in pal+alpha -> rgba conversion

... in particular, so that fully opaque pixels remain fully opaque

8 years agobabl: consistently use unsigned char in u8 palette conversions
Ell [Sat, 30 Sep 2017 11:42:56 +0000 (07:42 -0400)]
babl: consistently use unsigned char in u8 palette conversions

... to avoid erroneous sign extensions

8 years agosse4-int8: avoid using uninitialized variable
Øyvind Kolås [Fri, 29 Sep 2017 23:21:44 +0000 (01:21 +0200)]
sse4-int8: avoid using uninitialized variable

8 years agobabl: use -no-install to avoid libtool .sh wrappers
Øyvind Kolås [Thu, 28 Sep 2017 21:44:10 +0000 (23:44 +0200)]
babl: use -no-install to avoid libtool .sh wrappers

8 years agobabl: fix crash when doing full introspection
Øyvind Kolås [Mon, 25 Sep 2017 22:21:53 +0000 (00:21 +0200)]
babl: fix crash when doing full introspection

also add skeleton functions for introspection of trcs/spaces

8 years agobabl: do not try to enumerate sampling conversions
Øyvind Kolås [Mon, 25 Sep 2017 22:14:11 +0000 (00:14 +0200)]
babl: do not try to enumerate sampling conversions

8 years agobabl: set babl_rel_avg_error to a high value on NaN
Øyvind Kolås [Mon, 25 Sep 2017 20:46:21 +0000 (22:46 +0200)]
babl: set babl_rel_avg_error to a high value on NaN

In relation to bug #787441, if the sum of errors contain a nan the sum
is nan, we return pi as a much higher than expected average error to
indicate that this is not 0.0.

8 years agotests/grayscale_to_rgb: increase printed debug precision
Øyvind Kolås [Mon, 25 Sep 2017 19:41:36 +0000 (21:41 +0200)]
tests/grayscale_to_rgb: increase printed debug precision

8 years agotests: move non-test binaries from here to ../tools
Øyvind Kolås [Mon, 25 Sep 2017 19:06:46 +0000 (21:06 +0200)]
tests: move non-test binaries from here to ../tools

8 years agobabl: check both source/dest bits/component in bad_idea()
Øyvind Kolås [Mon, 25 Sep 2017 19:00:08 +0000 (21:00 +0200)]
babl: check both source/dest bits/component in bad_idea()

8 years agobabl: add decrease of bits per components to bad_idea()
Øyvind Kolås [Mon, 25 Sep 2017 18:56:55 +0000 (20:56 +0200)]
babl: add decrease of bits per components to bad_idea()

related to bug #787441

8 years agotests: use a epsilon instead of == for float compare
Øyvind Kolås [Mon, 25 Sep 2017 18:33:17 +0000 (20:33 +0200)]
tests: use a epsilon instead of == for float compare

8 years agobuild: link extensions against internal .la again
Øyvind Kolås [Mon, 25 Sep 2017 18:04:54 +0000 (20:04 +0200)]
build: link extensions against internal .la again

8 years agobabl: limit exported symbols
Øyvind Kolås [Mon, 25 Sep 2017 17:53:52 +0000 (19:53 +0200)]
babl: limit exported symbols

For now, also including symbols needed for internal tools and tests, that are
not part of the public API; the API promised by babl remains the public
headers, but this reduces the public dynamic loaders namespace a little.

8 years agobabl: tidying of public headers
Øyvind Kolås [Mon, 25 Sep 2017 10:56:17 +0000 (12:56 +0200)]
babl: tidying of public headers

8 years agobabl: in babl_polynomial_approximate_gamma(), verify that x0 >= 0
Ell [Sun, 24 Sep 2017 21:46:50 +0000 (17:46 -0400)]
babl: in babl_polynomial_approximate_gamma(), verify that x0 >= 0

Now that the previous commit (hopefully) makes us iPhone-proof, we
can add an assertion for this condition.  If we want to replace the
assertions with some sort of bailing, we'll need to replace the rest
of them too anyway.

8 years agoBug 788093 - Crash loading (jpg with) ICC v4 profile with parametric TRCs
Ell [Sun, 24 Sep 2017 21:34:00 +0000 (17:34 -0400)]
Bug 788093 - Crash loading (jpg with) ICC v4 profile with parametric TRCs

Fix parameter offset calculation of parametric TRCs, so that we
parse them correctry.

8 years agobabl: avoid negative degree when shrinking a zero polynomial
Ell [Sun, 24 Sep 2017 17:43:02 +0000 (13:43 -0400)]
babl: avoid negative degree when shrinking a zero polynomial

When shrinking a zero polynomial, avoid trying to set its degree
to a negative value, and just keep it at 0, which is the minimal
degree we support.

Note that this also happens for polynomials whose coefficients are
all NaN, as is the case in bug 788093, so they also become zero
polynomials after shrinking, incidentally.

8 years agobabl: make babl_space_to_icc internal, for now
Øyvind Kolås [Fri, 22 Sep 2017 12:50:48 +0000 (14:50 +0200)]
babl: make babl_space_to_icc internal, for now

Also adds an optimized 26 entry LUT, as used in the creation of
http://pippin.gimp.org/sRGBz/
 Please enter the commit message for your changes. Lines starting

8 years agobabl: remove commented out debug printfs
Øyvind Kolås [Tue, 19 Sep 2017 14:02:22 +0000 (16:02 +0200)]
babl: remove commented out debug printfs

8 years agobabl: improve icc number serialization
Øyvind Kolås [Mon, 18 Sep 2017 17:24:35 +0000 (19:24 +0200)]
babl: improve icc number serialization

8 years agobabl: add matrix re-equalization
Øyvind Kolås [Mon, 18 Sep 2017 17:17:56 +0000 (19:17 +0200)]
babl: add matrix re-equalization

This optimizes the coefficients of the matrix ensuring that a RGB 1.0 1.0 1.0
results in exactly CIE Lab 100.0 0.0 0.0 and that equal R,G,B triples yield 0.0
for CIE a, b. This is achieved by rounding to 16.16 fixed point precision,
which can be exactly represented by IEEE double, and then brute-force jittering
the coefficients +/- 1 for the best solution. This is also the rounding needed
for making the matrix well behaved when used in an ICC profile.

8 years agobabl: tidy up icc writing
Øyvind Kolås [Mon, 18 Sep 2017 00:47:52 +0000 (02:47 +0200)]
babl: tidy up icc writing

8 years agobabl: make powf approximation yield 0 for negatives
Øyvind Kolås [Sun, 17 Sep 2017 18:24:16 +0000 (20:24 +0200)]
babl: make powf approximation yield 0 for negatives

8 years agobabl: improve log output on Android.
Jehan [Wed, 5 Jul 2017 00:15:27 +0000 (02:15 +0200)]
babl: improve log output on Android.

No need to log end of lines. This is not like a printf where we are
trying to pretty-print.

8 years agobabl: replace a fprintf (stderr, ...) with babl_log().
Jehan [Tue, 4 Jul 2017 14:47:52 +0000 (16:47 +0200)]
babl: replace a fprintf (stderr, ...) with babl_log().

It will make it visible in the logcat, for instance in Android Studio.
As a general rule, we should use babl_log() for anything which we
consider as "logs", unless we want to direct them to a given output for
sure (for instance in the case of CLI tools where we want the output to
be on standard I/O in terminal, fprintf() still is the best choice).

8 years agotests: make babl build on haiku
Øyvind Kolås [Thu, 14 Sep 2017 19:24:22 +0000 (21:24 +0200)]
tests: make babl build on haiku

Using -lphtread instead of -pthread in Makefile.am fixes the build when using
gcc-x86 under haiku, figured out by schumaml.

8 years agobabl/base: model-gray, avoid potential NaN
Øyvind Kolås [Thu, 14 Sep 2017 16:32:09 +0000 (18:32 +0200)]
babl/base: model-gray, avoid potential NaN

8 years agobabl: add rendering intent argument to icc loading API
Øyvind Kolås [Fri, 8 Sep 2017 19:26:04 +0000 (21:26 +0200)]
babl: add rendering intent argument to icc loading API