/*
This file is part of Magnum .
Copyright © 2010 , 2011 , 2012 , 2013 , 2014 , 2015 , 2016 , 2017 , 2018 , 2019 ,
2020 , 2021 , 2022 , 2023 Vladimír Vondruš < mosra @ centrum . cz >
Permission is hereby granted , free of charge , to any person obtaining a
copy of this software and associated documentation files ( the " Software " ) ,
to deal in the Software without restriction , including without limitation
the rights to use , copy , modify , merge , publish , distribute , sublicense ,
and / or sell copies of the Software , and to permit persons to whom the
Software is furnished to do so , subject to the following conditions :
The above copyright notice and this permission notice shall be included
in all copies or substantial portions of the Software .
THE SOFTWARE IS PROVIDED " AS IS " , WITHOUT WARRANTY OF ANY KIND , EXPRESS OR
IMPLIED , INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY ,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT . IN NO EVENT SHALL
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM , DAMAGES OR OTHER
LIABILITY , WHETHER IN AN ACTION OF CONTRACT , TORT OR OTHERWISE , ARISING
FROM , OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE .
*/
# include "DistanceFieldGlyphCache.h"
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
# include "Magnum/Image.h"
# include "Magnum/ImageView.h"
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
# include "Magnum/PixelFormat.h"
# include "Magnum/GL/Context.h"
# include "Magnum/GL/Extensions.h"
Pixel storage support, part 2: renamed ColorFormat to PixelFormat etc.
Yeah, sorry, I know, the enums are renamed for second or third time in a
row, first they were Image::Format, then ImageFormat, then ColorFormat
and now PixelFormat. But this time it's final and last time they are
renamed and now everything is finally consistent:
* ColorFormat::DepthComponent -- depth is not a color, thus
PixelFormat::DepthComponent makes a lot more sense.
* There will be PixelStorage classes, which will be stored in images
alonside PixelFormat/PixelType enums, making everything nicely
aligned.
* The GL documentation about glTexImage2D() etc. denotes the <format>
and <type> parameters as format and type of *pixel* data, so now we
are _finally_ consistent with the official naming.
I wonder why did I not choose PixelFormat originally. Anyway, the old
<Magnum/ColorFormat.h> header, ColorFormat, ColorType and
CompressedColorFormat types are now aliases to the new ones, are marked
as deprecated and will be removed in some future release (as always, I'm
waiting at least six months before removing the deprecated
functionality).
11 years ago
# ifndef CORRADE_NO_ASSERT
# include "Magnum/GL/PixelFormat.h"
Pixel storage support, part 2: renamed ColorFormat to PixelFormat etc.
Yeah, sorry, I know, the enums are renamed for second or third time in a
row, first they were Image::Format, then ImageFormat, then ColorFormat
and now PixelFormat. But this time it's final and last time they are
renamed and now everything is finally consistent:
* ColorFormat::DepthComponent -- depth is not a color, thus
PixelFormat::DepthComponent makes a lot more sense.
* There will be PixelStorage classes, which will be stored in images
alonside PixelFormat/PixelType enums, making everything nicely
aligned.
* The GL documentation about glTexImage2D() etc. denotes the <format>
and <type> parameters as format and type of *pixel* data, so now we
are _finally_ consistent with the official naming.
I wonder why did I not choose PixelFormat originally. Anyway, the old
<Magnum/ColorFormat.h> header, ColorFormat, ColorType and
CompressedColorFormat types are now aliases to the new ones, are marked
as deprecated and will be removed in some future release (as always, I'm
waiting at least six months before removing the deprecated
functionality).
11 years ago
# endif
# include "Magnum/GL/TextureFormat.h"
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
# include "Magnum/Math/Range.h"
# include "Magnum/TextureTools/DistanceField.h"
namespace Magnum { namespace Text {
DistanceFieldGlyphCache : : DistanceFieldGlyphCache ( const Vector2i & sourceSize , const Vector2i & size , const UnsignedInt radius ) :
# if !(defined(MAGNUM_TARGET_GLES) && defined(MAGNUM_TARGET_GLES2))
GlyphCache ( GL : : TextureFormat : : R8 , sourceSize , size , Vector2i ( radius ) ) ,
# elif !defined(MAGNUM_TARGET_WEBGL)
/* Luminance is not renderable in most cases */
GlyphCache ( GL : : Context : : current ( ) . isExtensionSupported < GL : : Extensions : : EXT : : texture_rg > ( ) ?
GL : : TextureFormat : : R8 : GL : : TextureFormat : : RGB8 , sourceSize , size , Vector2i ( radius ) ) ,
# else
GlyphCache ( GL : : TextureFormat : : RGB , sourceSize , size , Vector2i ( radius ) ) ,
# endif
_size { size } , _distanceField { radius }
{
# ifndef MAGNUM_TARGET_GLES
MAGNUM_ASSERT_GL_EXTENSION_SUPPORTED ( GL : : Extensions : : ARB : : texture_rg ) ;
# endif
/* Replicating the assertion from TextureTools::DistanceField so it gets
checked during construction already instead of only later during the
setImage ( ) call */
CORRADE_ASSERT ( sourceSize % size = = Vector2i { 0 } & &
( sourceSize / size ) % 2 = = Vector2i { 0 } ,
" Text::DistanceFieldGlyphCache: expected source and destination size ratio to be a multiple of 2, got " < < Debug : : packed < < sourceSize < < " and " < < Debug : : packed < < size , ) ;
# if defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
/* Luminance is not renderable in most cases */
if ( ! GL : : Context : : current ( ) . isExtensionSupported < GL : : Extensions : : EXT : : texture_rg > ( ) )
Warning ( ) < < " Text::DistanceFieldGlyphCache: " < < GL : : Extensions : : EXT : : texture_rg : : string ( ) < < " not supported, using inefficient RGB format for glyph cache texture " ;
# endif
}
DistanceFieldGlyphCache : : DistanceFieldGlyphCache ( NoCreateT ) noexcept : GlyphCache { NoCreate } , _distanceField { NoCreate } { }
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
GlyphCacheFeatures DistanceFieldGlyphCache : : doFeatures ( ) const {
return GlyphCacheFeature : : ImageProcessing
# ifndef MAGNUM_TARGET_GLES
| GlyphCacheFeature : : ProcessedImageDownload
# endif
;
}
void DistanceFieldGlyphCache : : doSetImage ( const Vector2i & offset , const ImageView2D & image ) {
GL : : Texture2D input ;
input . setWrapping ( GL : : SamplerWrapping : : ClampToEdge )
. setMinificationFilter ( GL : : SamplerFilter : : Linear )
. setMagnificationFilter ( GL : : SamplerFilter : : Linear ) ;
/* The constructor already checked that the ratio is an integer multiple,
so this division should lead to no information loss */
CORRADE_INTERNAL_ASSERT ( size ( ) . xy ( ) % _size = = Vector2i { 0 } ) ;
const Vector2i ratio = size ( ) . xy ( ) / _size ;
/* Upload the input texture and create a distance field from it. On ES2
without EXT_unpack_subimage and on WebGL 1 there ' s no possibility to
upload just a slice of the input , upload the whole image instead by
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
ignoring the PixelStorage properties of the input and also process it as
a whole . */
# ifdef MAGNUM_TARGET_GLES2
# ifndef MAGNUM_TARGET_WEBGL
if ( ! GL : : Context : : current ( ) . isExtensionSupported < GL : : Extensions : : EXT : : unpack_subimage > ( ) )
# endif
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
{
input . setImage ( 0 , GL : : textureFormat ( image . format ( ) ) , ImageView2D { image . format ( ) , size ( ) . xy ( ) , image . data ( ) } ) ;
_distanceField ( input , texture ( ) , { { } , size ( ) . xy ( ) / ratio } , size ( ) . xy ( ) ) ;
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
# ifdef MAGNUM_TARGET_WEBGL
static_cast < void > ( offset ) ;
# endif
}
# ifndef MAGNUM_TARGET_WEBGL
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
else
# endif
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
# endif
# if !(defined(MAGNUM_TARGET_GLES2) && defined(MAGNUM_TARGET_WEBGL))
{
/* Create an image view that includes the distance field radius as
well , to be sure the edges are processed appropriately as well */
/** @todo this feels more like a hack than a real solution, any better
idea ? Tried to make TextureTools : : DistanceField pad the processing
area on its own , but eventually gave up as it would ( a ) made the
processing do more work than necessary in many cases , with no easy
way to opt out of that , ( b ) wasn ' t really working if the input
didn ' t have a black strip around the edge on platforms that don ' t
support border clamp , ( c ) required the input texture to have
certain wrapping mode set */
CORRADE_INTERNAL_ASSERT ( image . storage ( ) . skip ( ) . xy ( ) = = offset ) ;
# ifdef CORRADE_NO_ASSERT
static_cast < void > ( offset ) ;
# endif
const Vector2i paddedMin = Math : : max ( Vector2i { 0 } ,
image . storage ( ) . skip ( ) . xy ( ) - padding ( ) ) ;
const Vector2i paddedMax = Math : : min ( size ( ) . xy ( ) ,
image . size ( ) + image . storage ( ) . skip ( ) . xy ( ) + padding ( ) ) ;
/* TextureTools::DistanceField expects the input size and output
rectangle size ratio to be a multiple of 2 in order for the shader
to perform pixel addressing correctly . That might not always be the
case with the rectangle passed to flushImage ( ) , so round the
paddedMin * down * to a multiple of the ratio and paddedMax * up * to a
multiple of the ratio . */
const Vector2i paddedMinRounded = ratio * ( paddedMin / ratio ) ;
const Vector2i paddedMaxRounded = ratio * ( ( paddedMax + ratio - Vector2i { 1 } ) / ratio ) ;
/* As the size is also a multiple of ratio, the resulting size should
not get larger . */
CORRADE_INTERNAL_ASSERT ( paddedMaxRounded < = size ( ) . xy ( ) ) ;
const ImageView2D paddedImage {
PixelStorage { image . storage ( ) }
. setSkip ( { paddedMinRounded , image . storage ( ) . skip ( ) . z ( ) } ) ,
image . format ( ) ,
paddedMaxRounded - paddedMinRounded ,
image . data ( ) } ;
input . setImage ( 0 , GL : : textureFormat ( paddedImage . format ( ) ) , paddedImage ) ;
_distanceField ( input , texture ( ) , Range2Di : : fromSize ( paddedMinRounded / ratio , paddedImage . size ( ) / ratio ) , paddedImage . size ( ) ) ;
}
# endif
}
void DistanceFieldGlyphCache : : setDistanceFieldImage ( const Vector2i & offset , const ImageView2D & image ) {
CORRADE_ASSERT ( ( offset > = Vector2i { } & & offset + image . size ( ) < = _size ) . all ( ) ,
" Text::DistanceFieldGlyphCache::setDistanceFieldImage(): " < < Range2Di : : fromSize ( offset , image . size ( ) ) < < " out of range for texture size " < < _size , ) ;
# ifndef CORRADE_NO_ASSERT
const GL : : PixelFormat format = GL : : pixelFormat ( image . format ( ) ) ;
# endif
# if !(defined(MAGNUM_TARGET_GLES) && defined(MAGNUM_TARGET_GLES2))
CORRADE_ASSERT ( format = = GL : : PixelFormat : : Red ,
" Text::DistanceFieldGlyphCache::setDistanceFieldImage(): expected " < < GL : : PixelFormat : : Red < < " but got " < < format , ) ;
# else
# ifndef MAGNUM_TARGET_WEBGL
if ( GL : : Context : : current ( ) . isExtensionSupported < GL : : Extensions : : EXT : : texture_rg > ( ) )
CORRADE_ASSERT ( format = = GL : : PixelFormat : : Red ,
" Text::DistanceFieldGlyphCache::setDistanceFieldImage(): expected " < < GL : : PixelFormat : : Red < < " but got " < < format , ) ;
else
# endif
{
/* Luminance is not renderable in most cases */
CORRADE_ASSERT ( format = = GL : : PixelFormat : : RGB ,
" Text::DistanceFieldGlyphCache::setDistanceFieldImage(): expected " < < GL : : PixelFormat : : RGB < < " but got " < < format , ) ;
}
# endif
texture ( ) . setSubImage ( 0 , offset , image ) ;
}
Text: rework AbstractGlyphCache for better flexibility and efficiency.
The class now supports incremental filling, multiple fonts, texture
arrays, removes all reliance on STL containers and is finally properly
documented.
To avoid complete breakage of every use, as much as possible was kept as
deprecated APIs -- in particular the reserve() with the nasty
std::vectors, the insert() that assumes a 2D cache and a single font
and textureSize() that returns a 2D vector. Those behave the same as
before, but will assert if the cache is an array or contains more than
one font.
On the other hand, begin() / end() access with std::unordered_map iterators
(ew!) was removed as the internals simply aren't a hashmap anymore. The
image() that returned an Image2D is now used to fill the glyph cache
instead of querying its potentially processed contents, and returns a
MutableImageView3D. I considered keeping it and adding sourceImage()
instead, but such naming turned out to be too inconsistent. For querying
processed image data (such as with the distance field cache) there's a
new processedImage() query, guarded by new GlyphCacheFeature bits -- if
both ImageProcessing and ProcessedImageDownload is set, it can be used
to retrieve the processed image (so, similar as ImageDownload was
before), and if neither is set, the cache contents are queryable
directly through image(), without needing any special support from
the GPU API.
Existing code is updated only in the minimal way possible to ensure that
no serious breakage was introduced by reimplementing the deprecated APIs
on top of the new backend. Porting away from deprecated APIs will be
done in next commits. The GlyphCache and DistanceFieldGlyphCache have
their public API kept intact for now, as a similar rework will be needed
for them as well.
Additionally, the MagnumFont and MagnumFontConverter plugins aren't
compiling yet as they require substantial changes to deal with the new
glyph cache features. That is not the case with other plugins in the
magnum-plugins repository tho, for those the backwards compatibility
"just works". On the other hand, since layout of the AbstractGlyphChange
changed, I'm bumping the AbstractFont plugin interface version to
force-trigger a rebuild of dependent projects. Because I ran a stale
magnum-player binary, it worked without crashing or GL errors but just
didn't show ANY text whatsoever due to ABI differences, and I wasted
some precious minutes before realizing that a simple rebuild would fix
it.
3 years ago
# ifndef MAGNUM_TARGET_GLES
Image3D DistanceFieldGlyphCache : : doProcessedImage ( ) {
Image2D out = _texture . image ( 0 , PixelFormat : : R8Unorm ) ;
return Image3D { out . format ( ) , { out . size ( ) , 1 } , out . release ( ) } ;
}
# endif
} }
