Compare commits

..

17 Commits

Author SHA1 Message Date
f49704b2df Update dependency version to 6.10.0 2025-01-03 16:15:48 +01:00
43f3fd05f7 avif: color profiles improvements
With Qt 6.8.x, PQ and HLG transfer functions are supported.
Better support for GRAY profiles
2025-01-02 14:11:55 +01:00
a7cf1a87f9 Update heif.json 2024-12-31 17:33:58 +01:00
fe28130cb3 HEIF plug-in extended to read AVCI format
AVCI is H.264 encapsulated in HEIF container
2024-12-29 23:44:18 +01:00
c0d5b8854b Readme updated with some clarification
- `Contributing` section: `qimageformats module of Qt` changed to `image formats of Qt` because the Qt plugins are also in the core module (e.g. PNG)
- `The HEIF plugin` section: clarified which codec is needed for testing (HEVC)
- Added missing info about EPS, KRA and ORA
- Other minor changes
2024-12-29 07:37:06 +00:00
0b2c6d725d DDS: enable plugin by default
- Enabled the plugin as OSS Fuzz did not find anything yet
- Modified README by adding more info about contributing, DDS plugin and some clarifications

Solves CCBUG: 380956 CCBUG: 389900
2024-12-24 08:59:32 +00:00
894524f7e4 PSD: Added support to MCH1 and MCH2
- Multichannel images are treat as CMYK image when the number of channels are more than 1: when exists, channel 5 is used as alpha. Channels higher than 5 are discarded.
- Multichannel images are treat as Grayscaleimage when the number of channels are equals to 1.
- Device transactions removed (where possible)
- Fix clang-format issues
2024-12-23 23:58:40 +00:00
348ddce987 XCF: Fix OSS Fuzz issue 42527849
Fixes integer overflow when the value is -INT_MAX-1
2024-12-23 22:05:05 +00:00
1cb294545f JXR: Fix loss of HDR data on write for format RGBA16FPx16_Premultiplied
When writing RGBA16FPx16_Premultiplied format, the image was converted to RGBA64 by clamping the float values. With this patch float values ​​outside the range [0, 1] are preserved.
2024-12-23 06:36:39 +00:00
b649cca304 DDS: Fix for OSS Fuzz issue 384974505 2024-12-19 07:18:37 +00:00
adc5c7ae9a DDS: improved read/write support
The following changes have been made:

- Improved writing speed by using scanLine() instead of pixel()
- Optimized memory usage on writing by using ScanlineConverter class
- Added native write support for RGBA32FPx4, RGBA16FPx4, RGB8, Grayscale8 and Indexed8 uncompressed formats
- Grayscale DDS without alpha are loaded in a Grayscale8 image
- Fixed warnings about wrong PITCH reported by GIMP on DDSs saved by this plugin
- Initial support for loading DX10 formats (R16F, RG16F, RGBA16F, RGBPreMulA16F, R32F, RG32F, RGBA32F, RGBPreMulA32F)
- Fixed alignment issues and A8P8 format support of the attached images*

Tested using GIMP and [NVIDIA Texture Tools](https://developer.nvidia.com/texture-tools-exporter) plugin for Photoshop.

(*) The following images (taken from [here](https://github.com/walbourn/directxtexmedia)) cannot be added to read tests due to license issue:
[test8_DWORD.dds](/uploads/449b5a0d886aaf6764af554fe38e2b09/test8_DWORD.dds)
[dx5_logo.dds](/uploads/6f5f27df752890b227ef07e0195435d4/dx5_logo.dds)
[test888_DWORD.dds](/uploads/c8bc355c5749cf203d47e0b3073ad419/test888_DWORD.dds)
2024-12-17 23:08:43 +00:00
a6f7482957 Read / Write test: added NULL device test
DDS plugin crashes if I request supportedSubTypes()

- Fixed DDS plugin crash
- Added NULL device test on both read and write tests

Closes #14
2024-12-16 13:52:06 +00:00
d91c7dd912 DDS: multiple FP improvements
- Use of native qfloat16 format
- FP16 and FP32 HDR images are preserved and sRGB Linear profile is set
- Images are always saved as 8-bit so, if valid profile is present the image is converted to sRGB 
- It also fixes (left image) a problem (right image) with HDR images:
![_DF316651-18AF-498F-8689-B3613F8C9D45_](/uploads/f5deb9bff6d22dae94ecf7942d8fd806/_DF316651-18AF-498F-8689-B3613F8C9D45_.png){width=281 height=157}![_AF65379D-E6A5-4007-863D-7B60EDFB7383_](/uploads/92460f2dabbb763061f6e6143bc36719/_AF65379D-E6A5-4007-863D-7B60EDFB7383_.png){width=271 height=169}
2024-12-12 22:07:07 +00:00
86865223d2 PFM: fix error when loading images with comments
I tested the plugin with the images stored [here](https://github.com/walbourn/directxtexmedia). It contains PFM and PHM images but I didn't find any license on the repo, so the images was not added to test cases.

I attach here the images that this patch allows to load:
[grad4d_mono.pfm](/uploads/6e41fcb64d0651a6abd78cffc5ff86b2/grad4d_mono.pfm)
[grad4d_mono.phm](/uploads/348a51476068aa344f67826006cb65c4/grad4d_mono.phm)
[grad4d.phm](/uploads/2ab8330466f4dd0fc2fda00711270ce9/grad4d.phm)
2024-12-12 09:48:34 +00:00
ecbcf3b7f4 DDS: fix buffer overflow in readCubeMap
The issue was identified by OSS Fuzz and the feature was not covered by our tests.

- Added earth-cubemap.dds under MIT licenses taken from [Open Toolkit library](https://github.com/mono/opentk/tree/main/Source/Examples/Data/Textures)
- Fix a wrong image size returned by a cubemap image
- Read test skips .license files
2024-12-12 09:42:24 +00:00
87eff569a4 Re-added DDS plugin support
Fork of [Qt 5.6 DDS plugin](https://code.qt.io/cgit/qt/qtimageformats.git/tree/src/plugins/imageformats/dds/qddshandler.cpp?h=5.6) under LGPL2.1.

- Merged all files in dds_p.h and dds.cpp
- Added support for Qt 6 image allocation limit
- Added checks for null image and datastream errors
- The plugin is disabled by default

CCBUG: 380956

Closes: #12
2024-12-11 06:45:00 +00:00
a531978e2a Update version to 6.10.0 2024-12-06 16:07:49 +01:00
119 changed files with 3198 additions and 195 deletions

View File

@ -7,4 +7,4 @@ Dependencies:
Options:
test-before-installing: True
require-passing-tests-on: [ 'Linux', 'FreeBSD', 'Windows' ]
cmake-options: "-DKIMAGEFORMATS_JXR=ON"
cmake-options: "-DKIMAGEFORMATS_DDS=ON -DKIMAGEFORMATS_JXR=ON"

View File

@ -1,11 +1,11 @@
cmake_minimum_required(VERSION 3.16)
set(KF_VERSION "6.9.0") # handled by release scripts
set(KF_DEP_VERSION "6.9.0") # handled by release scripts
set(KF_VERSION "6.10.0") # handled by release scripts
set(KF_DEP_VERSION "6.10.0") # handled by release scripts
project(KImageFormats VERSION ${KF_VERSION})
include(FeatureSummary)
find_package(ECM 6.9.0 NO_MODULE)
find_package(ECM 6.10.0 NO_MODULE)
set_package_properties(ECM PROPERTIES TYPE REQUIRED DESCRIPTION "Extra CMake Modules." URL "https://commits.kde.org/extra-cmake-modules")
feature_summary(WHAT REQUIRED_PACKAGES_NOT_FOUND FATAL_ON_MISSING_REQUIRED_PACKAGES)
@ -62,6 +62,8 @@ set_package_properties(libavif PROPERTIES
PURPOSE "Required for the QImage plugin for AVIF images"
)
option(KIMAGEFORMATS_DDS "Enable plugin for DDS format" ON)
option(KIMAGEFORMATS_HEIF "Enable plugin for HEIF format" OFF)
if(KIMAGEFORMATS_HEIF)
pkg_check_modules(LibHeif IMPORTED_TARGET libheif>=1.10.0)
@ -82,6 +84,7 @@ set_package_properties(LibRaw PROPERTIES
PURPOSE "Required for the QImage plugin for RAW images"
)
# JXR plugin disabled by default due to security issues
option(KIMAGEFORMATS_JXR "Enable plugin for JPEG XR format" OFF)
if(KIMAGEFORMATS_JXR)
find_package(LibJXR)

View File

@ -1,6 +1,7 @@
# KImageFormats
Plugins to allow `QImage` to support extra file formats.
Plugins to allow [`QImage`](https://doc.qt.io/qt-6/qimage.html) to support
extra file formats.
## Introduction
@ -28,6 +29,7 @@ The following image formats have read-only support:
The following image formats have read and write support:
- AV1 Image File Format (avif)
- DirectDraw Surface (dds)
- Encapsulated PostScript (eps)
- High Efficiency Image File Format (heif)
- JPEG XL (jxl)
@ -41,16 +43,22 @@ The following image formats have read and write support:
## Contributing
See the [`QImageIOPlugin`](https://doc.qt.io/qt-6/qimageioplugin.html) documentation for information on how to write a
new plugin.
See the [`QImageIOPlugin`](https://doc.qt.io/qt-6/qimageioplugin.html)
documentation for information on how to write a new plugin.
The main difference between this framework and the qimageformats module
of Qt is the license. As such, if you write an imageformat plugin and
you are willing to sign the Qt Project contributor agreement, it may be
better to submit the plugin directly to the Qt Project.
The main difference between this framework and the image formats of Qt is
the license. As such, if you write an image format plugin and you are
willing to sign the Qt Project contributor agreement, it may be better to
submit the plugin directly to the Qt Project.
To be accepted, contributions must:
- Contain the test images needed to verify that the changes work correctly
- Pass the tests successfully
## Duplicated Plugins
### The TGA plugin
The TGA plugin supports more formats than Qt's own TGA plugin;
specifically, the one provided here supports indexed, greyscale and RLE
images (types 1-3 and 9-11), while Qt's plugin only supports type 2
@ -61,6 +69,12 @@ licensing. If anyone were willing to write fresh code to improve Qt's
TGA plugin, it would allow the TGA plugin in this framework to be
removed.
### The DDS plugin
The DDS plugin is a fork from Qt 5.6 with bug fixes and improvements.
The plugin was forked because Qt Project no longer supports its DDS plugin.
## License
This framework is licensed under the
@ -71,22 +85,23 @@ The CMake code in this framework is licensed under the
## Plugin status
The current implementation of a plugin may not be complete or may have limitations
of various kinds. Typically the limitations are on maximum size and color depth.
The current implementation of a plugin may not be complete or may have
limitations of various kinds. Typically the limitations are on maximum size
and color depth.
The various plugins are also limited by the formats natively supported by Qt.
The various plugins are also limited by the formats natively supported by Qt.
For example, native support for CMYK images is only available since Qt 6.8.
### HDR images
HDR images are supported via floating point image formats from EXR, HDR, JXL,
JXR, PFM and PSD plugins.
It is important to note that in the past these plugins stripped away HDR
HDR images are supported via floating point image formats from DDS, EXR, HDR,
JXL, JXR, PFM and PSD plugins.
It is important to note that in the past these plugins stripped away HDR
information, returning SDR images.
HDR images return R, G and B values outside the range 0.0 - 1.0.
While Qt painters handles HDR data correctly, some older programs may display
strange artifacts if they do not use a tone mapping operator (or at least a
While Qt painters handles HDR data correctly, some older programs may display
strange artifacts if they do not use a tone mapping operator (or at least a
clamp). This is not a plugin issue.
### Metadata
@ -114,11 +129,15 @@ plugin ('n/a' means no limit, i.e. the limit depends on the format encoding).
- ANI: n/a
- AVIF: 32,768 x 32,768 pixels, in any case no larger than 256 megapixels
- DDS: n/a
- EXR: 300,000 x 300,000 pixels
- EPS: n/a
- HDR: n/a (large image)
- HEIF: n/a
- JXL: 262,144 x 262,144 pixels, in any case no larger than 256 megapixels
- JXR: n/a, in any case no larger than 4 GB
- KRA: same size as Qt's PNG plugin
- ORA: same size as Qt's PNG plugin
- PCX: 65,535 x 65,535 pixels
- PFM: n/a (large image)
- PIC: 65,535 x 65,535 pixels
@ -162,10 +181,23 @@ without using the ICC profile.
JXR, PSD and SCT plugins natively support 4-channel CMYK images when compiled
with Qt 6.8+.
### The DDS plugin
**This plugin can be disabled by setting `KIMAGEFORMATS_DDS` to `OFF`
in your cmake options.**
The following defines can be defined in cmake to modify the behavior of the plugin:
- `DDS_DISABLE_STRIDE_ALIGNMENT`: disable the stride aligment based on DDS pitch: it is known that some writers do not set it correctly.
### The HEIF plugin
**This plugin is disabled by default. It can be enabled with the
`KIMAGEFORMATS_HEIF` build option in the cmake file.**
**This plugin is disabled by default. It can be enabled by settings
`KIMAGEFORMATS_HEIF` to `ON` in your cmake options.**
The plugin is disabled due to issues with the heif library on certain
distributions. In particular, it is necessary that the HEIF library has
support for HEVC codec. If HEVC codec is not available the plugin
will compile but will fail the tests.
### The EXR plugin
@ -173,6 +205,14 @@ The following defines can be defined in cmake to modify the behavior of the plug
- `EXR_CONVERT_TO_SRGB`: the linear data is converted to sRGB on read to accommodate programs that do not support color profiles.
- `EXR_DISABLE_XMP_ATTRIBUTE`: disables the stores XMP values in a non-standard attribute named "xmp". Note that Gimp reads the "xmp" attribute and Darktable writes it as well.
### The EPS plugin
The plugin uses `Ghostscript` to convert the raster image. When reading it
converts the EPS to PPM and uses the Qt PPM plugin to read the image.
When writing it uses [`QPrinter`](https://doc.qt.io/qt-6/qprinter.html) to
create a temporary PDF file which is then converted to EPS. Therefore, if
`Ghostscript` is not installed, the plugin will not work.
### The HDR plugin
The following defines can be defined in cmake to modify the behavior of the plugin:
@ -189,8 +229,8 @@ The following defines can be defined in cmake to modify the behavior of the plug
### The JXR plugin
**This plugin is disabled by default. It can be enabled with the
`KIMAGEFORMATS_JXR` build option in the cmake file.**
**This plugin is disabled by default. It can be enabled by settings
`KIMAGEFORMATS_JXR` to `ON` in your cmake options.**
The following defines can be defined in cmake to modify the behavior of the plugin:
- `JXR_DENY_FLOAT_IMAGE`: disables the use of float images and consequently any HDR data will be lost.
@ -198,11 +238,24 @@ The following defines can be defined in cmake to modify the behavior of the plug
- `JXR_DISABLE_BGRA_HACK`: Windows displays and opens JXR files correctly out of the box. Unfortunately it doesn't seem to open (P)RGBA @32bpp files as it only wants (P)BGRA32bpp files (a format not supported by Qt). Only for this format an hack is activated to guarantee total compatibility of the plugin with Windows.
- `JXR_ENABLE_ADVANCED_METADATA`: enable metadata support (e.g. XMP). Some distributions use an incomplete JXR library that does not allow reading metadata, causing compilation errors.
### The KRA plugin
The KRA format is a ZIP archive containing image data. In particular, the
rendered image in PNG format is saved in the root: the plugin reads this
image.
### The ORA plugin
The ORA format is a ZIP archive containing image data. In particular, the
rendered image in PNG format is saved in the root: the plugin reads this
image.
### The PSD plugin
PSD support has the following limitations:
- Only images saved by Photoshop using compatibility mode enabled (Photoshop default) can be decoded.
- Multichannel images are treated as CMY/CMYK and are only loaded if they have 3 or more channels.
- Multichannel images are treated as CMYK if they have 2 or more channels.
- Multichannel images are treated as Grayscale if they have 1 channel.
- Duotone images are treated as grayscale images.
- Extra channels other than alpha are discarded.

View File

@ -76,6 +76,11 @@ kimageformats_read_tests(
tga
)
if(KIMAGEFORMATS_DDS)
kimageformats_read_tests(dds)
kimageformats_write_tests(dds-nodatacheck-lossless)
endif()
if (KF6Archive_FOUND)
kimageformats_read_tests(
kra
@ -106,6 +111,12 @@ if (LibHeif_FOUND)
hej2
)
endif()
if (LibHeif_VERSION VERSION_GREATER_EQUAL "1.19.0")
kimageformats_read_tests(FUZZ 4
avci
)
endif()
endif()
if (LibJXL_FOUND AND LibJXLThreads_FOUND)

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 250 KiB

Binary file not shown.

View File

@ -0,0 +1,2 @@
SPDX-FileCopyrightText: Copyright (c) 2006 - 2010 The Open Toolkit library.
SPDX-License-Identifier: MIT

Binary file not shown.

After

Width:  |  Height:  |  Size: 956 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.1 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.1 KiB

BIN
autotests/read/dds/r16.dds Normal file

Binary file not shown.

BIN
autotests/read/dds/r16.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.2 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.2 KiB

BIN
autotests/read/dds/r32.dds Normal file

Binary file not shown.

BIN
autotests/read/dds/r32.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.2 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.2 KiB

BIN
autotests/read/dds/rg16.dds Normal file

Binary file not shown.

BIN
autotests/read/dds/rg16.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.6 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.6 KiB

BIN
autotests/read/dds/rg32.dds Normal file

Binary file not shown.

BIN
autotests/read/dds/rg32.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.6 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.6 KiB

BIN
autotests/read/dds/rgba.dds Normal file

Binary file not shown.

BIN
autotests/read/dds/rgba.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.7 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.1 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.7 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.6 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.2 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.3 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.3 KiB

Binary file not shown.

View File

@ -0,0 +1,5 @@
[
{
"fileName" : "testcard_gray_half.png"
}
]

Binary file not shown.

After

Width:  |  Height:  |  Size: 26 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 16 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 53 KiB

Binary file not shown.

Binary file not shown.

After

Width:  |  Height:  |  Size: 31 KiB

Binary file not shown.

View File

@ -263,7 +263,7 @@ int main(int argc, char **argv)
}
for (const QFileInfo &fi : lstImgDir) {
TemplateImage timg(fi);
if (timg.isTemplate()) {
if (timg.isTemplate() || timg.isLicense()) {
continue;
}
@ -375,6 +375,66 @@ int main(int argc, char **argv)
}
}
// NULL device test
for (const QFileInfo &fi : lstImgDir) {
TemplateImage timg(fi);
if (timg.isTemplate() || timg.isLicense()) {
continue;
}
QTextStream(stdout) << "* Run on NULL device\n";
QImageReader reader;
reader.setFormat(fi.suffix().toLatin1());
if (reader.canRead() == true) {
QTextStream(stdout) << "FAIL : " << fi.suffix() << ": canRead() returns true\n";
++failed;
break;
}
if (!reader.read().isNull()) {
QTextStream(stdout) << "FAIL : " << fi.suffix() << ": read() returns a non-NULL image\n";
++failed;
break;
}
if (reader.size() != QSize()) {
QTextStream(stdout) << "FAIL : " << fi.suffix() << ": size() returns a valid size\n";
++failed;
break;
}
if (reader.imageFormat() != QImage::Format_Invalid) {
QTextStream(stdout) << "FAIL : " << fi.suffix() << ": size() returns a valid format\n";
++failed;
break;
}
// test for crash only
reader.textKeys();
reader.quality();
reader.clipRect();
reader.scaledSize();
reader.scaledClipRect();
reader.backgroundColor();
reader.supportsAnimation();
reader.transformation();
reader.autoTransform();
reader.subType();
reader.supportedSubTypes();
reader.jumpToNextImage();
reader.loopCount();
reader.imageCount();
reader.currentImageNumber();
reader.currentImageRect();
// success
QTextStream(stdout) << "PASS : " << fi.suffix() << "\n";
++passed;
// runs once for each format
break;
}
QTextStream(stdout) << "Totals: " << passed << " passed, " << skipped << " skipped, " << failed << " failed\n";
QTextStream(stdout) << "********* "
<< "Finished basic read tests for " << suffix << " images *********\n";

View File

@ -28,6 +28,11 @@ bool TemplateImage::isTemplate() const
return false;
}
bool TemplateImage::isLicense() const
{
return !m_fi.suffix().compare(QStringLiteral("license"), Qt::CaseInsensitive);
}
QFileInfo TemplateImage::compareImage(TestFlags &flags, QString& comment) const
{
auto fi = jsonImage(flags, comment);

View File

@ -47,6 +47,12 @@ public:
*/
bool isTemplate() const;
/*!
* \brief isLicense
* \return True if the file suffix is .license
*/
bool isLicense() const;
/*!
* \brief compareImage
* \param flags Flags for modifying test behavior (e.g. image format not supported by current Qt version).

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

Some files were not shown because too many files have changed in this diff Show More