Commit Graph

3390 Commits

Author SHA1 Message Date
Neil Roberts
6ea2fbc729 Add an internal _cogl_bitmap_new_from_buffer
This function creates a CoglBitmap which internally references a
CoglBuffer. The map and unmap functions will divert to mapping the
buffer. There are also now bind and unbind functions which should be
used instead of map and unmap whenever the data doesn't need to be
read from the CPU but will instead be passed to GL for packing or
unpacking. For bitmaps created from buffers this just binds the
bitmap.

cogl_texture_new_from_buffer now just uses this function to wrap the
buffer in a bitmap rather than trying to bind the buffer
immediately. This means that the buffer will be bound only at the
point right before the texture data is uploaded.

This approach means that using a pixel array will take the fastest
upload route if possible, but can still fallback to copying the data
by mapping the buffer if some conversion is needed. Previously it
would just crash in this case because the texture functions were all
passed a NULL pointer.

http://bugzilla.clutter-project.org/show_bug.cgi?id=2112
2010-07-15 17:27:15 +01:00
Neil Roberts
7eb0d01f9f Add a Cogl debug flag for BITMAP
CoglBitmap will soon want to report notes as it gets more complicated
so this adds a debug flag for it.
2010-07-15 17:27:15 +01:00
Neil Roberts
b780413e5a Try to avoid copying the GdkPixbuf when it is tightly packed
The docs for GdkPixbuf say that the last row of the image won't
necessarily be allocated to the size of the full rowstride. The rest
of Cogl and possibly GL assumes that we can copy the bitmap with
memcpy(height*rowstride) so we previously would copy the pixbuf data
to ensure this. However if the rowstride is the same as bpp*width then
there is no way for the last row to be under-allocated so in this case
we can just directly upload from the gdk pixbuf. Now that CoglBitmap
can be created with a destroy function we can make it keep a reference
to the pixbuf and unref it during its destroy callback. GdkPixbuf
seems to always pack the image with no padding between rows even if it
is RGB so this should end up always avoiding the memcpy.

The fallback code for when we do have to copy the pixbuf is now
simplified so that it copies all of the rows in a single loop. We only
copy the useful region of each row so this should be safe. The
rowstride of the CoglBitmap is now always allocated to bpp*width
regardless of the rowstride of the pixbuf.
2010-07-15 17:25:36 +01:00
Neil Roberts
fc12c42a83 cogl-bitmap: Encapsulate the CoglBitmap even internally
The CoglBitmap struct is now only defined within cogl-bitmap.c so that
all of its members can now only be accessed with accessor
functions. To get to the data pointer for the bitmap image you must
first call _cogl_bitmap_map and later call _cogl_bitmap_unmap. The map
function takes the same arguments as cogl_pixel_array_map so that
eventually we can make a bitmap optionally internally divert to a
pixel array.

There is a _cogl_bitmap_new_from_data function which constructs a new
bitmap object and takes ownership of the data pointer. The function
gets passed a destroy callback which gets called when the bitmap is
freed. This is similar to how gdk_pixbuf_new_from_data
works. Alternatively NULL can be passed for the destroy function which
means that the caller will manage the life of the pointer (but must
guarantee that it stays alive at least until the bitmap is
freed). This mechanism is used instead of the old approach of creating
a CoglBitmap struct on the stack and manually filling in the
members. It could also later be used to create a CoglBitmap that owns
a GdkPixbuf ref so that we don't necessarily have to copy the
GdkPixbuf data when converting to a bitmap.

There is also _cogl_bitmap_new_shared. This creates a bitmap using a
reference to another CoglBitmap for the data. This is a bit of a hack
but it is needed by the atlas texture backend which wants to divert
the set_region virtual to another texture but it needs to override the
format of the bitmap to ignore the premult flag.
2010-07-15 17:24:01 +01:00
Neil Roberts
fd886d9fc2 cogl-texture-2d-sliced: Store the internal format not image format
The 'format' member of CoglTexture2DSliced is returned by
cogl_texture_get_format. All of the other backends return the internal
format of the GL texture in this case. However the sliced backend was
returning the format of the image data used to create the texture. It
doesn't make any sense to retain this information because it doesn't
necessarily indicate the format of the actual texture. This patch
changes it to store the internal format instead.
2010-07-15 14:31:23 +01:00
Emmanuele Bassi
fd27ca7398 Mark property strings for translation
Both the nick and the blurb fields should be translatable, for UI
builders and other introspection-based tools.
2010-07-15 14:07:07 +01:00
Emmanuele Bassi
8be43b5285 Add P_() macro for translating property strings
The P_() macro adds a context for the property nick and blurb. In order
to make xgettext recognize it, we need to drop glib-gettexize inside the
autogen.sh script and ship a modified Makefile.in.in with Clutter.
2010-07-15 12:09:01 +01:00
Nate Stedman
fff5558ad2 Fixes build with GLX on Mac OS X.
Moves preprocessor #ifdef __linux_ above else statement, avoiding the
lack of an else block if __linux__ is not defined.

http://bugzilla.clutter-project.org/show_bug.cgi?id=2212

Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2010-07-15 00:31:23 +01:00
Emmanuele Bassi
e7ae71e072 build: Fix include path for Cally's introspection
The introspection scanner does not include '.' by default, so it was
always using the installed copy of Clutter-1.0.gir. Which obviously
wouldn't work if we didn't have one.
2010-07-15 00:27:43 +01:00
Neil Roberts
54b2e1781e gles/cogl-texture-driver.c: Fix the include for cogl-material-private
In ddb9016be4 the GL texture driver backend was changed to include
cogl-material-opengl-private.h instead of cogl-material-private.h.
However the gles texture backend was missed from this so it was giving
a compiler warning about using an undeclared function.
2010-07-14 17:49:16 +01:00
Neil Roberts
ccdcbbb46b cogl-texture-3d: Use glTexSubImage3D through an indirect pointer
glTexSubImage3D was being called directly in cogl-texture-3d.c but the
function is only available since GL version 1.2 so on Windows it won't
be possible to directly link to it. Also under GLES it is only
available conditionally in an extension.
2010-07-14 17:45:15 +01:00
Neil Roberts
943266db3b cogl-texture-3d: Fix the cogl-material-private header include
In ddb9016be4 the texture backends were changed to include
cogl-material-opengl-private.h instead of cogl-material-private.h.
However the 3D texture backend was missed from this so it was giving a
compiler warning about using an undeclared function.
2010-07-14 16:35:33 +01:00
Neil Roberts
d0d72e4d97 cogl-texture-3d: Don't include cogl-texture-2d-private.h
I think this was included by a cut-and-paste error as it isn't needed
anywhere in the source.
2010-07-14 16:34:42 +01:00
Robert Bragg
5e66970385 material: splits out all the state flushing code
This moves the code supporting _cogl_material_flush_gl_state into
cogl-material-opengl.c as part of an effort to reduce the size of
cogl-material.c to keep it manageable.
2010-07-13 19:26:58 +01:00
Robert Bragg
ddb9016be4 material: split the texture unit management out
In general cogl-material.c has become far to large to manage in one
source file. As one of the ways to try and break it down this patch
starts to move some of lower level texture unit state management out
into cogl-material-opengl.c. The naming is such because the plan is to
follow up and migrate the very GL specific state flushing code into the
same file.
2010-07-13 19:26:58 +01:00
Robert Bragg
9e893e684e material: copy_differences: handle copying fog state
When the support for redirecting the legacy fog state through cogl
material was added in 9b9e764dc, the code to handle copying the fog
state in _cogl_material_copy_differences was missed.
2010-07-13 19:26:57 +01:00
Neil Roberts
ceb57087a7 Add a GL_GENERATE_MIPMAP fallback to the texture 2d and 3d backends
The CoglTexture2DSliced backend has a fallback for when the
framebuffer extension is missing so it's not possible to use
glGenerateMipmap. This involves keeping a copy of the upper-left pixel
of the tex image so that we can temporarily enable GL_GENERATE_MIPMAP
on the texture object and do a sub texture update by reuploading the
contents of the first pixel. This patch copies that mechanism to the
2D and 3D backends. The CoglTexturePixel structure which was
previously internal to the sliced backend has been moved to
cogl-texture-private.h so that it can be shared.
2010-07-13 18:41:01 +01:00
Emmanuele Bassi
ff56f4ac6f Merge branch 'wip/xkb-support'
* wip/xkb-support:
  x11: Use XKB to translate keycodes into key symbols
  x11: Use XKB to track the Locks state
  x11: Use XKB detectable auto-repeat
  x11: Add a Keymap ancillary object
  x11: Store the group inside the event platform data
  events: Add platform-data to allocated Events
  build: Check for the XKB extension
2010-07-13 16:12:14 +01:00
Emmanuele Bassi
0cbfabcda4 stage: Add the NO_CLEAR_ON_PAINT hint
Some apps or some use cases don't need to clear the stage on immediate
rendering GPUs. A media player playing a fullscreen video or a
tile-based game, for instance.

These apps are redrawing the whole screen, so we can avoid clearing the
color buffer when preparing to paint the stage, since there is no
blending with the stage color being performed.

We can add an private set of hints to ClutterStage, and expose accessors
for each potential hint; the first hint is the 'no-clear' one.

http://bugzilla.clutter-project.org/show_bug.cgi?id=2058
2010-07-13 14:57:48 +01:00
Neil Roberts
984e04cae8 Make the material functions for setting the p wrap mode public
Now that we have 3D texture support it makes sense to expose the wrap
mode for the p coordinate.
2010-07-13 14:29:07 +01:00
Neil Roberts
192cfa7a58 Rename the third texure coordinate from 'r' to 'p'
Using 'r' to name the third component is problematic because that is
commonly used to represent the red component of a vector representing
a color. Under GLSL this is awkward because the texture swizzling for
a vector uses a single letter for each component and the names for
colors, textures and positions are synonymous. GLSL works around this
by naming the components of the texture s, t, p and q. Cogl already
effectively already exposes this naming because it exposes GLSL so it
makes sense to use that naming consistently. Another alternative could
be u, v and w. This is what Blender and Direct3D use. However the w
component conflicts with the w component of a position vertex.
2010-07-13 14:29:07 +01:00
Neil Roberts
dfea57b45a Add a Cogl texture 3D backend
This adds a publicly exposed experimental API for a 3D texture
backend. There is a feature flag which can be checked for whether 3D
textures are supported. Although we require OpenGL 1.2 which has 3D
textures in core, GLES only provides them through an extension so the
feature can be used to detect that.

The textures can be created with one of two new API functions :-

cogl_texture_3d_new_with_size

 and

cogl_texture_3d_new_from_data

There is also internally a new_from_bitmap function. new_from_data is
implemented in terms of this function.

The two constructors are effectively the only way to upload data to a
3D texture. It does not work to call glTexImage2D with the
GL_TEXTURE_3D target so the virtual for cogl_texture_set_region does
nothing. It would be possible to make cogl_texture_get_data do
something sensible like returning all of the images as a single long
image but this is not currently implemented and instead the virtual
just always fails. We may want to add API specific to the 3D texture
backend to get and set a sub region of the texture.

All of those three functions can throw a GError. This will happen if
the GPU does not support 3D textures or it does not support NPOTs and
an NPOT size is requested. It will also fail if the FBO extension is
not supported and the COGL_TEXTURE_NO_AUTO_MIPMAP flag is not
given. This could be avoided by copying the code for the
GL_GENERATE_MIPMAP TexParameter fallback, but in the interests of
keeping the code simple this is not yet done.

This adds a couple of functions to cogl-texture-driver for uploading
3D data and querying the 3D proxy
texture. prep_gl_for_pixels_upload_full now also takes sets the
GL_UNPACK_IMAGE_HEIGHT parameter so that 3D textures can have padding
between the images. Whenever 3D texture is uploading, both the height
of the images and the height of all of the data is specified (either
explicitly or implicilty from the CoglBitmap) so that the image height
can be deduced by dividing by the depth.
2010-07-13 14:28:52 +01:00
Neil Roberts
8ccf8f6f08 configure: Include gl2ext.h or glext.h under GLES
Under big GL, glext.h is included automatically by gl.h. However under
GLES this doesn't appear to happen so it has to be included explicitly
to get the defines for extensions. This patch changes the
clutter_gl_header to be called cogl_gl_headers and it can now take a
space seperated list of multiple headers. This is then later converted
to a list of #include lines which ends up cogl-defines.h. The gles2
and gles1 backends now add their respective ext header to this list.
2010-07-13 14:28:51 +01:00
Neil Roberts
8940c30681 Make a public CoglBitmapError enum
There are many places in the texture backend that need to do
conversion using the CoglBitmap code. Currently none of these
functions can throw an error but they do return a value to indicate
failure. In future it would make sense if new texture functions could
throw an error and in that case they would want to use a CoglBitmap
error if the failure was due to the conversion. This moves the
internal CoglBitmap error from the quartz backend to be public in
cogl-bitmap.h so that it can be used in this way.
2010-07-13 14:28:45 +01:00
Neil Roberts
539b3f83e1 Rename COGL_ERROR_MISSING_FEATURE to COGL_ERROR_UNSUPPORTED
We can use this error in more unsupported situations than just when we
have a Cogl feature flag for the error. For example if a non-sliced
texture is created with dimensions that are too large then we could
throw this error. Therefore it seems good to rename to something more
general.
2010-07-13 14:28:36 +01:00
Neil Roberts
423f712871 Move _cogl_texture_2d_is_pot to cogl-util.h
This function could be used in many places in Cogl so it makes sense
to share it in cogl-util.h as _cogl_util_is_pot().
2010-07-13 14:28:35 +01:00
Neil Roberts
31c127bfa7 cogl-gles2-wrapper: Layers aren't equal if one is enabled and one is not
Previously when comparing whether the settings for a layer are equal
it would only check if one of them was enabled. If so then it would
assume the other one was enabled and continue to compare the texture
environment. Now it also checks whether the enabledness differs.
2010-07-13 14:28:35 +01:00
Emmanuele Bassi
6c913aa55e x11: Use XKB to translate keycodes into key symbols
And fall back to XKeycodeToKeysym() if XKB is not available.
2010-07-13 11:54:44 +01:00
Emmanuele Bassi
1ea4c50041 x11: Use XKB to track the Locks state
For Caps Lock and Num Lock. Store the state of both in the
platform-specific per-event data structure. No accessors, yet.
2010-07-13 09:15:21 +01:00
Emmanuele Bassi
bf2f8d670d x11: Use XKB detectable auto-repeat
If we have XKB support then we should be using it to turn on the
detectable auto-repeat; this allows avoiding the peeking trick
that emulates it inside the event handling code.
2010-07-13 08:27:48 +01:00
Emmanuele Bassi
bea657d3d5 x11: Add a Keymap ancillary object
We should try to abstract everything that is related with the key mapping
to its own object, to avoid complicating ClutterBackendX11 any further.
2010-07-13 08:27:48 +01:00
Emmanuele Bassi
d345a61e6c x11: Store the group inside the event platform data
Now that we have private, per-event platform data, we can start putting
it to good use. The first, most simple use is to store the key group
given the event's modifiers. Since we assume a modern X11, we use XKB
to retrieve it, or we simply fall back to 0 by default.

The data is exposed as a ClutterX11-specific function, within the
sanctioned clutter_x11_* namespace.
2010-07-13 08:27:48 +01:00
Emmanuele Bassi
f44ccba42e events: Add platform-data to allocated Events
Events allocated by Clutter should have a pointer to platform-specific
data; this would allow backends to add separate structures for holding
ancillary data, whilst retaining the ClutterEvent structure for use on
the stack.

In theory, for Clutter 2.x we might just want to drop Event and use an
opaque structure, or a typed data structure inheriting from
GTypeInstance instead.
2010-07-13 08:27:48 +01:00
Øyvind Kolås
25dd0ae356 state: swap clutter_state_set_state and clutter_state_warp_to_state
Got the boolean arguments for the old clutter_state_change wrong,
making the methods do the opposite of what they should.
2010-07-12 17:34:17 +01:00
Neil Roberts
90acc6a2a9 Add the deprecated ref-counting for cogl_program
cogl_program has always had cogl_program_ref and cogl_program_unref
but this was missed from 89cb325fd4 so they got removed.
2010-07-09 19:09:49 +01:00
Neil Roberts
8a85088196 Don't define public cogl_is_* functions for internal types
This adds a COGL_OBJECT_INTERNAL_DEFINE macro and friends that are the
same as COGL_OBJECT_DEFINE except that they prefix the cogl_is_*
function with an underscore so that it doesn't get exported in the
shared library.
2010-07-09 18:57:54 +01:00
Neil Roberts
0e839c3769 Add cogl_vertex_buffer_is_indices to the public headers
This function has always been defined in the shared library but it was
missed from the public headers.
2010-07-09 18:57:54 +01:00
Neil Roberts
89cb325fd4 cogl: Don't define the deprecated ref/unref accessors for new types
Previously COGL_OBJECT_DEFINE would always define deprecated
cogl_$type_{ref,unref} functions even if the type is new or if the
type is entirely internal. An application would still find it
difficult to use these because they wouldn't be in the headers, but it
still looks bad that they are exported from the shared library. This
patch changes it so that the deprecated ref counting functions are
defined using a separate macro and only the types that have these
functions in the headers call this macro.
2010-07-09 18:57:54 +01:00
Emmanuele Bassi
903560c6c8 text: Layouts for single-line-mode entries are boundless
A PangoLayout for an editable, single-line text entry is by definition
without a given width, since the actor will be clipped when being drawn.
2010-07-09 18:38:45 +01:00
Robert Bragg
1635a2db12 material: use common node type for materials and layers
Since 365605cf42, materials and layers are represented in a tree
structure that allows traversing up through parents and iterating down
through children.  This re-works the related typedefs and reparenting
code so that they can be shared.
2010-07-09 18:04:27 +01:00
Damien Lespiau
9e82fe3944 cogl-texture-pixmap-x11: This API is new in 1.4, not 1.2
CoglTexturePixmapX11 has been introduced in the 1.3 development cycle. A
Stability: Unstable tag was missing too, so add it.
2010-07-09 17:55:31 +01:00
Damien Lespiau
892aa45513 cogl: Add Stability tag to new experimental API
Functions guarded with COGL_ENABLE_EXPERIMENTAL API should be also maked
as being Unstable with the Stability gtk-doc tag.

Fixes: http://bugzilla.clutter-project.org/show_bug.cgi?id=2175
2010-07-09 17:17:11 +01:00
Emmanuele Bassi
b7f99ddd3c script: Let ClutterActor parse behaviours
Up until now, the "behaviours" member of an actor definition was parsed
by the ClutterScript parser itself - even though it's not strictly
necessary.

In an effort to minimize the ad hoc code in the Script parser, we should
let ClutterActor handle all the special cases that involve
actor-specific members.
2010-07-09 14:59:32 +01:00
Neil Roberts
a65429363e Use GL_MAX_TEXTURE_SIZE on GLES when checking supported tex size
Under big GL, _cogl_texture_driver_size_supported uses the proxy
texture to check whether the given texture size is supported. Proxy
textures aren't available under GLES so previously this would just
return TRUE to assume all texture sizes are supported. This patch
makes it use glGetIntegerv with GL_MAX_TEXTURE_SIZE to give a second
best guess.

This fixes the sliced texture backend so that it will use slices when
the texture is too big.
2010-07-09 11:14:15 +01:00
Neil Roberts
eb24d2a252 Fix cogl_texture_get_data when an intermediate buffer is used
When an intermediate buffer is used for downloading texture data it
was using the wrong byte length for a row so the copy back to the
user's buffer would fail.
2010-07-09 11:14:15 +01:00
Neil Roberts
9e1bb31922 Use GL_NEAREST filter in the draw-and-read get_data texture fallback
The fallback for when glGetTexImage is not available renders the
texture to the framebuffer to read the data using glReadPixels. This
patch just sets the COGL_MATERIAL_FILTER_NEAREST filter mode on the
material before rendering to avoid linear filtering which would alter
the texture data.
2010-07-09 11:14:15 +01:00
Neil Roberts
25cf5979e6 Use cogl_read_pixels in the cogl texture draw-and-read fallback
The fallback for when glGetTexImage is not available draws parts of
the texture to the framebuffer and uses glReadPixels to extract the
data. However it was using cogl_rectangle to draw and then immediately
using raw glReadPixels to fetch the data. This won't cause a journal
flush so the rectangle won't necessarily have hit the framebuffer
yet. Instead it now uses cogl_read_pixels which does flush the
journal.
2010-07-09 11:14:15 +01:00
Neil Roberts
e454b9cadf cogl-material: Fix some problems with flushing texture overrides
There were a few problems flushing texture overrides so that sliced
textures would not work:

* In _cogl_material_set_layer_texture it ignored the 'overriden'
  parameter and always set texture_overridden to FALSE.

* cogl_texture_get_gl_texture wasn't being called correctly in
  override_layer_texture_cb. It returns a gboolean to indicate the
  error status but this boolean was being assigned to gl_target.

* _cogl_material_layer_texture_equal did not take into account the
  override.

* _cogl_material_layer_get_texture_info did not return the overridden
  texture so it would always use the first texture slice.
2010-07-09 11:14:15 +01:00
Neil Roberts
8beb49164f cogl-texture: Share the common code in the set_region virtual
There was a lot of common code that was copied to all of the backends
to convert the data to a suitable format and wrap it into a CoglBitmap
so that it can be passed to _cogl_texture_driver_upload_subregion_to_gl.
This patch moves the common code to cogl-texture.c so that the virtual
just takes a CoglBitmap that is already in the right format.
2010-07-09 11:14:14 +01:00
Neil Roberts
0577c81c3b cogl-texture: Share the common code in the get_data virtual
Previously cogl_texture_get_data would pretty much directly pass on to
the get_data texture virtual function. This ended up with a lot of
common code that was copied to all of the backends. For example, the
method is expected to return the required data size if the data
pointer is NULL and to calculate its own rowstride if the rowstride is
0. Also it needs to convert the downloaded data if GL can't support
that format directly.

This patch moves the common code to cogl-texture.c so the virtual is
always called with a format that can be downloaded directly by GL and
with a valid rowstride. If the download fails then the virtual can
return FALSE in which case cogl-texture will use the draw and read
fallback.
2010-07-09 11:00:48 +01:00