Summary: It caused crash when clipper is NULL and it makes evas test-suite fail.
Test Plan: Run evas test-suite. (make check)
Reviewers: woohyun, tasn, herdsman, Hermet
Reviewed By: Hermet
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D2215
Summary:
Reinitialize FontConfig. If FontConfig has to be reinitialized
according to changes of system enviroments(ex. Changing font config files), it will be useful.
Reviewers: woohyun, seoz, tasn, cedric, raster
Reviewed By: raster
Subscribers: raster, herdsman, cedric
Differential Revision: https://phab.enlightenment.org/D1962
Summary:
If the clipper is image or has color, it affects to its clipees.
Even if we unset the clipper or change the clipper to another object,
it seems the clipper is not changed.
Test Plan:
Make two clipper objects and one clipee object.
And make clip the clipee according to following example
ex) Clipee object -> inner_clipper -> clipper
evas_object_clip_set(clipee, inner_clipper);
evas_object_clip_set(inner_clipper, clipper);
After checking the result and hide inner_clipper.
evas_object_clip_set(clipee, clipper);
evas_object_hide(inner_clipper);
See the result.
Reviewers: raster, cedric, Hermet, jpeg
Subscribers: woohyun, cedric
Differential Revision: https://phab.enlightenment.org/D2112
Signed-off-by: Jean-Philippe Andre <jp.andre@samsung.com>
Note: Technically we should not check the color of the fact that
the clipper is a mask and not a simple rect. But because of
real-life performance issues, damage_add was disabled so we're
trying to keep the perf in most cases while being correct in
cases where the clipper is visually important.
Summary:
This should enable applications to use GLES 3.0 through evas gl.
Todo: Fix indirect rendering issue occuring because texture objects
cannot be shared between different version of GLES contexts.
Todo: extension pointers need to be updated for GLES 3.0
Reviewers: wonsik, spacegrapher, jpeg
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D2017
@feature
In some rare cases, a mask would be rendered (from mask_subrender)
into a surface that is NOT an FBO. This would happen because the
previous surface was a "scaled GL image" and its size would
match the required geometry.
That took a while to figure out...
http://thecodinglove.com/post/111546429281/when-i-finally-solve-a-nasty-bug
The previous commit modifies the concept of direct rendering
vs. indirect rendering, so some runtime checks (in debug mode
only) will fail.
This commit introduces two new engine functions:
- gl_get_pixels_pre
- gl_get_pixels_post
The latter will be used in a later patch for optimization.
not changed.
Automatically fallback to indirect rendering on FBO or X11 Pixmap
if the Evas Object Image is not marked as dirty. This should
improve the performance and/or power consumption in those
rare cases where this area of the canvas needs to be redrawn
but the GL content has not changed.
@feature
The function image_scaled_update() frees() the old scaled image
passed as input if it doesn't match the old dimensions. This commit
will avoid double frees.
Edje may not set the filled flag on an image even if its fill
properties make it fill the whole object. For masking, it can
then be considered as a filled image.
If the image is not "filled", then we can't assume its image
source geometry is the same as its texture geometry.
Note: Implementing a fast path for non-filled images would
require a hell of a lot more work (need to cut the render
into a lot more triangles) for little real-life use.
Call object's function to get the private engine_data (here, the
image object). Thanks Dongyeon for your patch which inspired me to
do that instead of forcing pre_render.
This will currently optimize most of the masks when using the
GL engine[1].
This is a very special case that adds a highly optimized path
for masking in GL. It works by creating a virtual image, containing
a pointer to the original image and a new geometry[2].
Instead of creating a new FBO-based surface (image_map_surface),
we refer to the original image and adjust the mask geometry on
the fly.
KNOWN BUGS:
- masking a map with such a scaled image is now broken.
[1] Right now all masks are simple Evas Object Image, so that means
all cases of masking, except masks of masks, or masks of maps,
will be optimized with this new method.
[2] This virtual image mechanism is still quite hackish and may
be improved (for memory usage, refcounting, etc...)
Those 2 new values are here to avoid using environment variables
that have side effects on the whole application.
I'm actually wondering if we shouldn't just kill off the env
vars altogether. Also, direct override is a terrible option that
should never be used.
Memory optimization can make sense (needs more testing tho).
Summary:
Add class and type Evas_3D_Callback like wrapper under smart object
Incapsulate Evas_3D_Callback in Evas_3D_Object
Add virtual function register and unregister in Evas_3D_Object
Add function evas_3d_callback_call
Add callbacks clicked and collision for Evas_3D_Node
@feature
Reviewers: raster, Hermet, cedric
Reviewed By: cedric
Subscribers: artem.popov, cedric
Differential Revision: https://phab.enlightenment.org/D1914
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
this fixes a lot of noise for filled image objects that set fill to
0x0 if objetc is 0x0. this clamps minimum fill at 1x1 and thus keeps
things silent and sensible. also just silently ignore 0x0 filled image
objects. noise is not that useful.
this adds a lock for when walking all the objects to generate render
commands for an async render. this allows even the object tree walk
plus update area caluclation to be moved off into async if every oject
that can change canvas state actually does so correctly. this change
adds all those lock block calls to synchronise with an async object
tree walk.
Summary:
See first part https://phab.enlightenment.org/D1811 (backend, gl)
Add get/set for color pick mode at evas_3d_mesh and evas_3d_scene
Add evas_3d_node_color_node_mesh_collect function to collect data at force rendering
Add state flag for scene to avoid useless force rendering in case scene wasn't changed
Add functionality for color pick in evas_3d_scene_pick method
Reviewers: Hermet, raster, cedric
Reviewed By: cedric
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1956
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Summary:
When getting ellipsis value from evas text object fails,
the most reasonable return value is -1.0.
Currently, evas_object_text_ellipsis_get API with NULL returns 0.0.
It means ellipsis is not off. It must return -1.0 when API fails.
@fix
Comments by Tom: until now, this was inconsistent. With this change, it
now returns -1.0 consistently. Also, fixed commit summary.
Reviewers: woohyun, Hermet, seoz, tasn
Reviewed By: tasn
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1944
Summary:
The order of tranformation changed to scale, orientation, position as
in some cases it can lead to incorrect value for the bounding box.
@fix
Reviewers: cedric, Hermet, raster
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1942
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Summary:
During filling evas pick public data by API evas_3d_scene_pick
segfault can occur if mesh was created without texcoords.
See functions - _pick_data_mesh_add, _pick_data_texcoord_update
@fix
Reviewers: cedric, Hermet, raster
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1941
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Fixes rendering in the following case:
- Object with a map has a mask
- Object is child of smart object which also has a map (eg. transit)
--> Masking did not apply to the children before this patch.
NOTE: This works fine in SW but still didn't work in GL, see the
following commit...
I know. This title does not explain anything. Whatever.
This fixes the following issue:
- Mask a genlist (big mask)
- Each item has an icon masked (small mask)
- Apply a map to the genlist
- Scrolling the genlist
--> The big mask still works but totally screws up the
small icons with masks.
Note: Once again this patch only affects code paths where an
object is a mask.
Yeah, mixing maps and masks of masks in a genlist leads
to tons of amazing bugs :)
This commit removes x,y from the "mask" field in an object,
as they are duplicates of cur->geometry.{x,y} but were not
properly kept in sync.
This patch fixes a situation of:
- A genlist in a map
- Each item has an icon masked
- Scrolling the genlist
--> The masked items would not render properly before this
patch.
Remaining known problem:
- Mask a genlist (big mask)
- Each item has an icon masked (small mask)
- Apply a map to the genlist
- Scrolling the genlist
--> The big mask still works but totally screws up the
small icons with masks.
Note: These changes look scary just before the release
but I would have to backport them to 1.13.x as they
definitely are bug fixes. Also, they only concern
code paths used exclusively by masking.
All those masking bug fixes become harder to explain. But here goes:
- Take a genlist, apply a mask to it (for example put everything
in an elm_layout). Also play with various objects in the genlist.
- Also apply a map to it (for instance, elm_transit zoom).
--> Now some elements will be masked, some others will not,
and some may even not render at all.
This patch restores a mask in the current drawing context, instead
of just unsetting it.
In a situation where an object with mask of mask is in a map
(Yes! It can happen!) the masks would not get properly "multiplied".
Now the problem is that some objects still seem to bypass some
masks... Hmm...
This is a left-over from a previous fix a few weeks ago.
The point of this "if" is just to avoid writing the COW value
if not needed.
For reference:
commit f876cf31f8
Author: Jean-Philippe Andre <jp.andre@samsung.com>
Date: Tue Dec 23 18:57:45 2014 +0900
Evas masking: Fix invalid geometry after mask redraw
Summary:
This is an attempt at fixing:
- T1767: The ultimate evil map & clip bug
Force recalculation and re-propagation of clipper geometry
after or just before a map is applied (only when transiting
between map enabled and map disabled).
I realized that doing clip_unset+clip_set in the E widget
code would fix the issue, but this is not a solution that
makes a lot of sense.
Unfortunately I have no idea about the side effects of this
patch, especially in terms of performance.
Fixes T1767 and maybe T1630.
Test Plan:
Open PackageKit popup in E, check the animations
and that clipping works fine both during, before and after
the animations.
Reviewers: raster, cedric
Reviewed By: cedric
Subscribers: cedric, Hermet
Maniphest Tasks: T1767
Differential Revision: https://phab.enlightenment.org/D1897
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
While invesigating some clip & map issues, I found some very
strange piece of code:
{
tmp = a;
a = c;
a = tmp;
}
This actually comes from a very old code refactoring where a
line in-between was removed:
tobj = obj->cur.map_parent;
obj->cur.map_parent = obj->cur.clipper->cur.map_parent;
- evas_object_clip_recalc(obj);
obj->cur.map_parent = tobj;
Adding this line back there doesn't seem to do anything anyways.
So, let's just remove useless code.
For the record (legacy evas):
commit e1f6f3c5f239dfd95a307949acd5f98831c0c3c0
Date: Fri Aug 17 06:16:04 2012 +0000
evas/render - code refactoring.
SVN revision: 75351
Some complex examples of masking with mapped smart objects
would fail miserably, rendering the object without any mask,
and/or showing the mask itself somewhere in white color...
The memory usage graph was going up and to the right!
I was told this is always a good thing!
... maybe not this time :)
Hopefully I didn't forget a case. An intense session of
genlist scrolling with masks all over the place and masks
of masks didn't show any glitch, crash or memory leak.
The main difference between 1.12 and 1.13 memory foot print is actually
related to this two pointer to mask. I am wondering if there is not an
issue here also has we do have a duplicated pointer. We have prev_mask
and mask in both cur and prev state of an Evas_Object, but only mask
and prev_mask from the cur state seems to be accessed.
If we can remove two pointers from those state, we should have a decent
win in expedite benchmark. Hopefully 300KB to win there (Close to half
the additional cost in memory).
The flag should be set on the mask itself.
Checking for (x,y) being inside the mask can be an expensive operation,
so further optimization will be required.
Summary:
There was no conversion to the double quotation mark in the evas_textblock_text_utf8_to_markup function.
The price of the text coming out to API and text coming out to Textblock was different as a result.
As a result, Two text lengths came out differently.
So, I added the exceptional treatment part in the evas_textblock_text_utf8_to_markup function.
@fix
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1853
In case the clipper is a mask object, we should use precise
event masking. By default precise_is_inside is not enabled
because it is expensive, but it should probably be set by
the application when they use masks as clippers.
Signed-off-by: Jean-Philippe Andre <jp.andre@samsung.com>
This implements supports for masking inside evas_render, which
means:
- Render the mask itself into a surface (ALPHA if possible)
- Pass this mask surface to the draw context
- Apply mask recursively in case a masked object is contained
by another masked object.
@feature
There was a problem when checking whether the current surface
is compatible with direct rendering. In case of client-side
rotation (it's a flag set on the surface by the app), a surface
can be directly rendered even if the rotation is not 0.
But, before this patch, it was assumed that the surface was
current. Which doesn't make sense because make_current is
called by the pixel callback, from the application, and this
happens *after* we check for direct rendering.
As a consequence, it was not possible to mix directly rendered
surfaces with FBO-based ones, and use client-side rotation.
This patch should solve that issue.
Summary:
Move check visibility of node from evas_3d_node to evas_3d_camera
Move functionality (normalize, check distance, calculate frustum)
in evas_3d_utils.h (we are planing use evas_is_sphere_in_frustum in evas_gl_3d.c -
don't render mesh if it non visible)
Add possibility check frustum by box, aabb, central point
Refactor example frustum culling
@feature
Reviewers: Hermet, raster, cedric
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1420
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
It's been so long. even SLP is not a valid name anymore.
No idea whether the problem still exist or not.
If it is then it should be reported and fixed.
Summary:
The selection performance is slow if we select large chunk of text.
This is caused by many rectangles created and deleted.
This patch provides a way to improve it by combine selection rectangles
of line in middle into one rectangles (i.e, if we have N lines,
the selection rectangle for lines 2 to N-1 will be combined into one.)
@feature
Reviewers: raster, cedric, tasn
Subscribers: herdsman, woohyun, cedric
Differential Revision: https://phab.enlightenment.org/D1508
Summary:
If a RTL textblock has right margin, text is wrongly placed
(right margin is moved to left).
This patch fixes this issue.
Test cases are also added to test text position with margins.
@fix
Reviewers: tasn, herdsman
Subscribers: woohyun, cedric
Differential Revision: https://phab.enlightenment.org/D1691
Summary:
Since Evas_Textblock_Cursor has pos of type size_t so changed
pos argument in _find_layout_item_line_match from int to size_t
Also Evas_Object_Textblock_Item has text_pos of size_t so defined
variable p of type size_t
Signed-off-by: kabeer khan <kabeer.khan@samsung.com>
Reviewers: tasn
Subscribers: devilhorns, cedric
Differential Revision: https://phab.enlightenment.org/D1692
Summary:
This patch set adds the necessary code to expose device axis state to applications. This was primarily written with graphics tablets in mind, which -- in addition to acting like a mouse -- also provide information about pen pressure, tilt, etc. Other devices could potentially benefit from this API as well: touchscreens, joysticks, knob controllers, "spaceballs", etc.
Whenever an update to the device state is recieved, an "Axis update" event is synthesized. This event contains the updated information, typically scaled and normalized to a particular logical range (e.g. zero to one for pressure, -pi to pi radians for angles, etc.). Information about the tool which generated the event is also stored so that applications can disambiguate events from multiple devices (or in the case of multitouch screens, individual fingers).
This API is only wired up for use with X11 at the moment. Support for other backends (e.g. Wayland) should be easy to add for those familiar them.
**Note**: The following is a list of changes from the "v2" patches originally sent to the mailinglist
//Define and implement new Ecore_Event_Axis_Update events//
* Harcode axis labels instead of including xserver-properties.h
* Use C89-style comments
* Use doxygen comments
* Update comment text to note axes with unbounded/undefined ranges/units
* Create "Ecore_Axis" and "Ecore_Axis_Label" typedefs
* Reference typedef'd instead of raw types
* Adjust how we count through valuators to support tilt/az
* Add support for tilt and azimuth
* Tweak memory management in case number of valuators differ
* Expand TWIST axis normalization to declared range
* Only normalize TWIST axis if resolution == 1 (wacom bug)
* Cache label atoms on first use to minimize round-trips
//Implement EVAS_CALLBACK_AXIS_UPDATE event and friends//
* Update to doxygen comments
* Update comment text to note axes with unbounded/undefined ranges/units
* Typedef 'Evas_Axis_Label', 'Evas_Axis'
* Move typedef for 'Evas_Event_Axis_Update'
* Reference typedef'd instead of raw types
//Wire the Ecore and Evas implementations of axis update events together//
* Expose ecore_event_evas_axis_update in Ecore_Input_Evas.h
* Move ecore_event_evas_axis_update to more logical position
//DEBUG: Add axis update logging to evas-multi-touch.c//
* Removed from patch set
//Make evas-multi-touch demo use new axis functionality//
* Have pressure adjust rectangle brightness instead of size
* Use more available axis data when rendering rectangle (azimuth, tilt, twist)
Test Plan: The evas-multi-touch demo was updated to support axis update events. A graphics tablet was then used to verify that the pressure, azimuth, tilt, and twist data was coming through correctly.
Reviewers: cedric, raster
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1514
Conflicts:
src/lib/ecore_input/Ecore_Input.h
Carsten Haitzler -
** fixed forward enum typedefs (make things unhappy)
** fixed conflict above
** fixed wrong param type for _evas_canvas_event_feed_axis_update()
** fixed @sinces to be 1.13
** fixed formatting/indeting
** fixed order of operation reliance in if's with ()'s to be clear
** fixed functions to be static that should have been
Summary:
.ply format is important for relation blender and EFl, because in blender exist only two mesh export API: bpy.ops.import_mesh.ply and bpy.ops.import_mesh.stl. One of them is necessary for .edc 3D generator. Which I writing now.
Sorry, it isn't like image loader. Refactoring of import/export will be soon.
Reviewers: Oleksander, artem.popov, Hermet, raster, cedric
Reviewed By: cedric
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1544
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
These macros replace some very repetitive code:
- define ENFN obj->layer->evas->engine.func
- define ENDT obj->layer->evas->engine.data.output
This commit includes lots of space changes as well.
This is a purely cosmetic commit.
Here's a macro that's used for debugging in some of the ugliest
ways possible: avoid passing an extra argument to a function when the
cost of always passing it is negligible (it's an int).
Fixes T1749.
Summary:
Default orientation of node should be corrected, because its zero rotation matrix multiplies on rotation matrix of its children node and we also get zero orientation for children,
so the vector of orientation should be non-zero
@fix
Reviewers: Hermet, raster, cedric
Reviewed By: cedric
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1647
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Thanks cedric for the report.
We'll need a new function (after 1.12) to detect if the current
cow pointer is the default data or not. Maybe eina_cow_default_get()
or maybe something just returning a bool.
Eina_Cow does memory comparison during standby time. So in expedite as we don't
have any standby time, we end up allocating some vast amount of memory that
will never be compared. It is way simpler to compare that the data are not
going to change before hand. It should also reduce the CPU consumed during
idle time.
This patch save about 1MB of data at peak time in expedite.
As pointed out by Cedric, the memory usage of basic evas objects
has increased a lot in recent versions of EFL, in part due
to this excessive use of filters data.
This is a partial fix for ticket 1725.
Summary:
Now minimum width hint (for horizontal layout) and minimum height hint (for vertical)
are set to largest item size rather than largest row/column.
@fix
Test Plan:
1. Run "elementary_test"
2. Maximize window
3. Unmaximize it
ExR: button boxes should return to start state.
Reviewers: cedric, seoz, Hermet
Subscribers: cedric, reutskiy.v.v
Differential Revision: https://phab.enlightenment.org/D1511
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Summary:
All points in map are double, when try to get point coordinates, there
are issues with rounding.
@fix
Reviewers: Hermet, raster, seoz, cedric
Reviewed By: cedric
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1554
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
This commit adds lots of required definitions (GL_ stuff),
new API entries in the Evas_GL_API struct, and basic new APIs
to create OpenGL-ES 1.1 contexts.
@feature
When using EGL, this function should return the properties of
a surface. Limited to a subset of known attributes.
Right now, this function will only work fine with EGL.
GLX support or any other engine is not implemented.
@feature
This will be used to increase the chances of having direct
rendering (no fallback to FBO) even if the window is rotated.
The client is then responsible for handling the view rotation.
@feature
Summary:
This is a fix of those warnings. Sorry for existence of them.
CC lib/evas/common/lib_evas_libevas_la-evas_alpha_main.lo
../../src/lib/evas/canvas/evas_3d_mesh_loader_eet.c:113:13: warning:
assignment from incompatible pointer type
eet_file = _evas_3d_eet_file_new();
^
CC lib/evas/common/lib_evas_libevas_la-evas_blend_main.lo
CC lib/evas/common/lib_evas_libevas_la-evas_blit_main.lo
CC lib/evas/common/lib_evas_libevas_la-evas_convert_color.lo
../../src/lib/evas/canvas/evas_3d_mesh_saver_eet.c: In function
'evas_3d_mesh_save_eet':
../../src/lib/evas/canvas/evas_3d_mesh_saver_eet.c:159:13: warning:
assignment from incompatible pointer type
eet_file = _evas_3d_eet_file_new();
^
CC lib/evas/common/lib_evas_libevas_la-evas_convert_colorspace.lo
CC lib/evas/common/lib_evas_libevas_la-evas_convert_gry_1.lo
../../src/lib/evas/canvas/evas_3d_eet.c: In function
'_evas_3d_eet_file_new':
../../src/lib/evas/canvas/evas_3d_eet.c:28:4: warning: return discards
'const' qualifier from pointer target type
return creating_file;
Reviewers: cedric, artem.popov
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1527
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
this is an unused code path UNLESS you:
export EVAS_RENDER2=1
so this will affect nothing unless you enable this and test it. it's
designe dto be done this way so we can slowly bring it in and bring it
up to speed and peolpe can easily test it and so on.
so now - content is rendered. most content is actually. only thing
that breaks that i can see so far is mapped smart objects.
Summary: The first version of .eet format is added. All changes due to discussion in D1307 are done.
Reviewers: artem.popov, se.osadchy, reutskiy.v.v, Hermet, raster, cedric, Oleksander
@feature
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D1477
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Summary:
Currently, in cursor geometry get function, the text direction is not
returned if cursor is under cursor.
This patch fixs it by returning text direction for under cursor.
@fix
Reviewers: tasn
Subscribers: herdsman, cedric
Differential Revision: https://phab.enlightenment.org/D1505
Summary:
This fixes an issue that causes BiDi text to get wrapped even when
resizing the textblock to its native size.
The way we find the last visual item in the native line is wrong. The
'visual_pos' describes the visual position in the formatted layout, in
which wrapping may occur. So, this is really bad to use it for native
width calculations as well.
Also, there's no need to actually reorder the line - we just need the
last visual item.
This adds and uses a function very similar to _layout_line_reorder, in
which we retrieve the last visual item in the native line. This
function does not do any actual reordering, as the native layout is
disposed of after calculation is done.
Also, added GET_ITEM_LEN macro for convenience.
Fixes T1532
@fix
Test Plan: Added to evas textblock test suite in this patch
Reviewers: tasn
Subscribers: id213sin, JackDanielZ, cedric
Maniphest Tasks: T1532
Differential Revision: https://phab.enlightenment.org/D1353
Summary:
In some cases of char or word wrapping, an empty line might be
accidentally added at the end of the paragraph. That line contains
no items. Of course, this line should not exist.
One outcome of this is that it causes wrong height values of the
paragraph, when the finalizing code uses the
_layout_last_line_max_descent_adjust_calc, which in turn
looks at that empty line to calculate the descent values.
@fix
Test Plan: Char-wrap and word-wrap tests to test suite included in this revision.
Reviewers: tasn
Subscribers: JackDanielZ, cedric
Projects: #efl
Differential Revision: https://phab.enlightenment.org/D1444
Before this change eo_add() used to create an object with 1 ref, and if
the object had a parent, a second ref.
Now, eo_add() always returns an object with 1 ref, and eo_add_ref()
preserves the old behaviour (for bindings).
eo_unref now un-parents if refcount is 0, and eo_del() is an alias for
eo_unref (will change to be a way to ensure an object is dead and goes
to zombie-land even if still refed).