This fixes a lot of cases where a genlist node (of any type)
with children may have appeared in the invalid expanded or
contracted state.
Before this patch, the test case "Genlist tree, Relative insert"
looked like below (all items are programmatically added):
> A
> 1
> 2
> B
> 3
> 4
The problem above is that A and B have visible children but
still believe they are in contracted state. This patch ensures
that A and B will be marked as expanded, but will do so without
firing an "expanded" signal and definitely without the
"expand,request" signal.
After this patch, the test case will look like this:
v A
> 1
> 2
v B
> 3
> 4
Which is more correct. Note that this test case does not handle
any expand/contract signal.
NOTE: This is a behaviour break!
This fixes a case where inserting item "C" after item "B" in this
tree would go wrong:
A
B
- 1
- 2
Before this patch, 1 and 2 lose their parent:
A
B
C
- 1
- 2
After this patch, 1 and 2 retain their parent:
A
B
- 1
- 2
C
Insert before worked by luck, no need to fix it.
Note that this patch may require the next one to actually
work (ensuring expanded state flag).
NOTE: This is a behaviour break!
This fixes the test case "Genlist Tree, Insert Sorted".
This is a pretty ugly patch... but the genlist code is already
pretty ugly, as it keeps a flat inlist of items (sd->items)
as well as a tree structure in parallel.
Before this patch, the following configuration led to issues:
1
3
- A
- B
Adding item "2" led to a crash. Adding item 4 led to this:
1
3
4
- A
- B
Items A and B lost their parent "3". Subsequent sorted inserts
would lead to insane bahaviour, where for instance "8" would
appear before "3".
This patch fixes all sorted inserts, at the cost of performance
(an optimized code path is avoided). Subsequent patches will
increase the robustness of the tree structure.
NOTE: This is a behaviour break!
Fixes T4850
Summary:
Change the item indexing to start 1 after reordering animation.
Change the logic of edge checking to get the row or col correctly.
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Test Plan:
elementary_test -to gengrid2
append 6 items
enable the reorder mode
check reordering is working properly (4 to 1)
@fix
Reviewers: singh.amitesh, cedric
Subscribers: jehun.lim, jpeg
Differential Revision: https://phab.enlightenment.org/D4676
With the new Efl.Ui.Clock widget, which implements Datetime,
most parts are relative to each other, and trigger a near-inifinite
recalc of all the parts. Indeed edje was recalculating a part even
if it had already calculated it (flag is set).
This seemingly simple patch fixes Datetime and probably a lot
of other edje widgets.
Fixes T4909
@fix
Some property set calls to Evas Text are NOPs if the property was
already set to the same value. Avoid doing any work then.
This adds a missing intercept call (size_set as it's overriden
by Evas.Text).
I was in a situation where my elm_config had the proper version
but did not contain a proper module name for the clock widget.
This resulted in not being able to create any field object in
the clock widget... in other words it made for a very useless
clock :)
Ideally the config should be updated at this point. I'm not
sure how to proceed exactly, though :(
This fixes a minor memory leak during shutdown.
Note: This does NOT perform the dlclose (and still leaks the
Eina_Module descriptor). Calling dlclose leads to a whole lot
of other issues, so we avoid it.
This fixes issues where Evas_Object_Intercept_Cb_Type was not
defined, as this is defined in Evas_Legacy.h (unfortunately...
as it's used by elementary), but the private headers defining
EFL_CANVAS_OBJECT_PROTECTED were included after Evas_Legacy.h.
clip_dirty is called A LOT. Unfortunately this patch does not
result in very measurable improvements.
**********************
Note about this merge:
After this series of patches, the CPU usage for a certain test
case has significantly gone down:
Based on c0e6a8d698c17fc16f9b67fc9 (upstream before git push):
NS since frame 2 = 28910806786 , 2937 frames = 9843652 / frame
After this patch:
NS since frame 2 = 19218592951 , 2928 frames = 6563727 / frame
1.18:
NS since 2 = 13105584220 , 2961 frames = 4426066 / frame
As we can see, 1.18 remains *much* better than 1.19. I'm still
struggling trying to figure out why (clip_recalc is called more,
but the call tree is hard to decypher).
The test case is:
EINA_FREEQ_BYPASS=1 ELM_TEST_AUTOBOUNCE=100 \
elementary_test -to "Scroller 2"
EFL was compiled with GCC 6.3.1 with -O3 -g
@optimization
This function was moved out of inline (see d7c6fca6c0) but
unfortunately the early checks at its beginning are likely
to result in an early return. Inline this part so we get back a
better performance. Inlining the whole function does not improve
the performance, as GCC simply gives up with inlining.
Note: Between 1.18 and master the number of calls to clip_recalc
has simply blown out. It is thus crucial to find out where those
calls come from but also micro-optimize the function itself. This
patch does the latter only.
@optimize
No need to verify that the clip is an evas object unless we're
actually going to change said clipper. This has a minor effect,
removing ~10% of all calls to efl_isa (0.1% of everything).
After all those efl_isa patches the number of calls has dropped
from 1.9M (~3.05%) to ~570k (0.76%). More can still be done,
surely. Now efl_isa accounts for ~0.05% of my CPU usage according
to perf top (down from ~0.7%).
If edje uses EO APIs instead of legacy image APIs, then many of
the remaining calls will vanish.
@optimization
In relative smart move, we only need to check once that the object
is indeed a valid smart object. This drops the call count from ~1.02M
to ~785k.
@optimization
This avoids calling efl_isa and locking the async mutex.
In callgrind analysis, this reduces the count of calls to
efl_isa from 1.96M to 1.02M.
@optimization
Summary: I had fixed some typos and some wrong expressions, such as capital letters, singular, and orders of groups in Eina API reference doxygen.
Test Plan: Doxygen Revision
Reviewers: stefan, cedric, raster, Jaehyun_Cho, jpeg
Reviewed By: jpeg
Subscribers: conr2d
Differential Revision: https://phab.enlightenment.org/D4674
Summary:
Also fix warning about lack of a format string:
warning: format not a string literal and no format arguments [-Wformat-security]
fprintf(stdout, commands);
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewers: cedric, jpeg
Reviewed By: jpeg
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D4673
If HAVE_GETPWENT isn't defined - the closing brace was missed.
Also prevent situation when strdup() tried to duplicate NULL
pointer, because that could cause segfault.
@fix
Summary:
When you set gravity 1 on scroller, scroller sticks to the bottom
even content is changed.
however, scroller don't work like above, if size of pan is changed.
this commit uses pan_pos_max rather than w/h of content_info
because pan_pos_max is related with both content_size and pan size.
gravity_set will work properly even if both size of content and pan are
changed simultaneously.
Test Plan:
1. Select 'scroll3' in the elementary_test
2. Append enough items so that scroll bar appears (about 30 items)
3. Go to the bottom and Set gravity 1.0
4. Check that scroller sticks to the bottom once you append another item
(it works)
5. Check that scroller sticks to to bottom once you resize window(pan)
(it doesn't work without this patch)
Reviewers: eagleeye, jpeg, cedric, woohyun, z-wony, herdsman
Differential Revision: https://phab.enlightenment.org/D4665
Summary: this should fix some spamming in e
Reviewers: jpeg
Reviewed By: jpeg
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D4675
Signed-off-by: Jean-Philippe Andre <jp.andre@samsung.com>
Lauro Moura
Daniel Kolesa
Jean-Philippe Andre
Andrii Kroitor
Amitesh Singh
Minkyu Kang
Myoungwoon Roy, Kim
Bryce Harrington
Stefan Schmidt
Mykyta Biliavskyi
Wonki Kim
Marcel Hollerbach
Jiyoun Park
Daniel Hirt
Andy Williams
Carsten Haitzler