Shilpa Singh [Wed, 15 Nov 2017 12:48:33 +0000 (18:18 +0530)]
elc_ctxpopup: Add item_style_get API
Change-Id: I1911c4fb41a21e71ffeb9cfde193755872f6909b
Signed-off-by: Shilpa Singh <shilpa.singh@samsung.com>
(cherry picked from commit
62bfee336397cc980f003e27213f2ffefae4c545)
Jaehyun Cho [Tue, 9 Jan 2018 06:48:32 +0000 (15:48 +0900)]
naviframe: Move focus from last content to first content by focus next
Previously, if naviframe focus reaches to the last content, then the
next focus goes to naviframe itself and focus remains the last content.
Now, if naviframe focus reaches to the last content, then the next focus
goes to the first content.
Change-Id: I2bac890cd5953bfe77fee47044b2f86ce7d4c148
SangHyeon Lee [Fri, 5 Jan 2018 08:04:04 +0000 (17:04 +0900)]
genlist: fix white bg issue in item clip unset
Change-Id: If610c2ec963452e971aaa9ea08b443b13d4a695b
Signed-off-by: SangHyeon Lee <sh10233.lee@samsung.com>
Youngbok Shin [Tue, 2 Jan 2018 11:18:12 +0000 (20:18 +0900)]
layout: add elm_layout_valign_set/get APIs
@tizen_feature
Change-Id: Id49dff4811801aa4d6bbd691177003eeeda14612
Youngbok Shin [Wed, 27 Dec 2017 10:28:31 +0000 (19:28 +0900)]
entry: fix cursor position initialization issues when theme is changed
The issue was caused by the following patch.
"elm_entry: Save cursor position before entry text clear"
It needed to get cursor position at the text_set function to avoid
cursor position issue by delaying signal callback functions.
So, we can't remove the patch. In the theme_apply, we need to handle
cursor position more carefully to fix this side effect.
@tizen_fix
Change-Id: Id63aa401ba2572559bc241f54350b781c72f5092
Youngbok Shin [Wed, 20 Dec 2017 05:06:40 +0000 (05:06 +0000)]
Revert "Revert "entry: disable text selection by user interaction if elm_entry_select_allow_set() is called with EINA_FALSE""
This reverts commit
ef9dc3b049a2c7511b37fffb62146d982555a19b.
Change-Id: I4079041ae846223b16db40a980fb9ac591c7aa08
Youngbok Shin [Tue, 19 Dec 2017 07:05:48 +0000 (16:05 +0900)]
Revert "entry: disable text selection by user interaction if elm_entry_select_allow_set() is called with EINA_FALSE"
This reverts commit
a11b7e7b5eb67b701b72d1354afef9c32cc4bb0d.
Change-Id: I47ac1d84c1bbad7f21380aaad7adcfeb8f613fa8
SangHyeon Lee [Tue, 19 Dec 2017 06:34:55 +0000 (15:34 +0900)]
genlist: fix tooltip_cb_set wrong exception handling
Change-Id: I34fa14a91213917fb0842ba36e4bf0efb29ba953
Signed-off-by: SangHyeon Lee <sh10233.lee@samsung.com>
Youngbok Shin [Wed, 13 Dec 2017 10:29:47 +0000 (19:29 +0900)]
entry: disable text selection by user interaction if elm_entry_select_allow_set() is called with EINA_FALSE
@tizen_fix
Change-Id: I86a2bc892f8de66c3d59b270f6146f03ac225506
Shinwoo Kim [Fri, 8 Dec 2017 07:54:54 +0000 (16:54 +0900)]
atspi: not accept redundant leaf object
An object which role is ELM_ATSPI_ROLE_REDUNDANT_OBJECT, has been acceptable
so far regardless its children. So it was possible to find its children.
But it does not make sense to accept ELM_ATSPI_ROLE_REDUNDANT_OBJECT object if
it does not have children.
There was an issue on Browser application because we accepted REDUNDANT leaf
object. There is a button which role is ELM_ATSPI_ROLE_REDUNDANT_OBJECT, this
button would be used for conveying focus event to embedded web page.
The button was accepted, so it was not possible grab an object on web page.
Please refer to following part of accessibility tree. It is not possible to
get Elm_Atspi_Ewk_Wrapper, if Elm_Button is accepted for "GetNavigableAtPoint"
Elm_Layout
- Elm_Button [redundant object][0,148,1080,1604]
- Elm_Atspi_Ewk_Wrapper [embedded][0,148,1080,1604]
Change-Id: I355a913314d0132e0b0e0f7779e1d7c0b66b25db
Shinwoo Kim [Wed, 29 Nov 2017 12:14:19 +0000 (21:14 +0900)]
atspi: do not accept ELM_ATSPI_ROLE_IMAGE_MAP
There are two widgets using ELM_ATSPI_ROLE_IMAGE_MAP role.
One is elm_mapbuf which does not have to grab highlight.
It does not make sense because the elm_mapbuf only works for its content.
The other is elm_map. So far the elm_map does not have to grab highlight, and
it seems that the ELM_ATSPI_ROLE_IMAGE_MAP is improper role for elm_map.
We have to find proper role for elm_map when it is neccesary to grab highlight.
So this patch set would be reasonable.
Change-Id: I8fc82c8f278b12f1e6367f4e40bf6b37c44b1481
Youngbok Shin [Mon, 4 Dec 2017 08:36:30 +0000 (17:36 +0900)]
elementary: apply slide ALWAYS mode for Edje marquee
Dependency warning: It should be applied with the following patch.
https://review.tizen.org/gerrit/162567
@tizen_fix
Change-Id: I0ff3a8dee81aef26f3cbce38477b816e7bd576f2
SangHyeon Jade Lee [Thu, 30 Nov 2017 04:24:15 +0000 (13:24 +0900)]
genlist: enabled select on focus
Change-Id: I9740b7ef0148c24bacc126a439fc1f8cfd64ceb0
Signed-off-by: SangHyeon Jade Lee <dltkdgus1764@gmail.com>
Jeonghyun Yun [Wed, 29 Nov 2017 06:29:39 +0000 (15:29 +0900)]
focus: Add second_level for focus direction weight calculate
@tizen_fix
Change-Id: Ifab1a45485741b12ac68a17a5fb24581dab3bce8
Signed-off-by: Jeonghyun Yun <jh0506.yun@samsung.com>
Youngbok Shin [Tue, 28 Nov 2017 11:53:34 +0000 (20:53 +0900)]
entry: improve performance for updating cursors
There is a patch on upstream EFL.
https://phab.enlightenment.org/D4223
@tizen_feature
Change-Id: I8abdfbf96ea757c5e1ab48a235315cc4aa00b858
SangHyeon Jade Lee [Wed, 22 Nov 2017 05:12:00 +0000 (14:12 +0900)]
genlist: fix item calculation performance issue
in calc(),
we only need to process first viewport items,
but currently it forcely calculate block sized items in
every single calc() funciton calls because of || operator.
it must be changed as && operator for orignal purpose.
Change-Id: If57081046d67c09d66c360dbbb75b31101ecc1e6
Signed-off-by: SangHyeon Jade Lee <dltkdgus1764@gmail.com>
Shinwoo Kim [Mon, 6 Nov 2017 12:04:13 +0000 (21:04 +0900)]
atspi: Use bottom up approach for PAGE_TAB
The naviframe item is elm_layout, and its atspi role is ELM_ATSPI_ROLE_PAGE_TAB.
The screen-reader could ask an object at point by using the naviframe item as a
root object. In this case, elementary did not us bottom up approach.
For the consistency, and to make it work better, This patch set is using the
bottom up approach for the naviframe item.
Change-Id: Icfe0791e340dac0c3998b56217fab8c34922cc5b
Shinwoo Kim [Mon, 6 Nov 2017 12:20:06 +0000 (21:20 +0900)]
Suppress build warning message
Change-Id: I320984158df1ad037005472c40451546d4e9ea3a
Youngbok Shin [Tue, 14 Nov 2017 04:49:19 +0000 (13:49 +0900)]
rename .gbs.conf to gbs.conf.bak temporarily
There was a build failure issue with the latest gbs (0.24.9).
I found removing .gbs.conf will fix the issue.
Also, modifying upstream_branch's value "upstream" to
"origin/upstream" or "origin/tizen" etc will fix it, too.
I don't know which way is correct way to fix this issue.
To keep history of this file, I just rename it.
So, you can use gbs 0.24.9 to build the packages.
@tizen_fix
Change-Id: I668f5c6aa0bff1e1fc5cbd5d8600f87335e30844
Myoungwoon Roy, Kim [Mon, 13 Nov 2017 00:36:54 +0000 (09:36 +0900)]
Remove invalid link for examples
Change-Id: Idff895c2ec25d1aa0a6b777fc6b6200c46ecd269
Jeonghyun Yun [Thu, 9 Nov 2017 01:19:17 +0000 (10:19 +0900)]
elm_panel: show and hide necessary object when scrollable is set
@tizen_fix
Change-Id: I229f9872551a5ba656c9fd91a1f62b34db282824
Signed-off-by: Jeonghyun Yun <jh0506.yun@samsung.com>
Shinwoo Kim [Tue, 31 Oct 2017 10:54:04 +0000 (19:54 +0900)]
atspi: use correct object for highlight_grabbing
elm_object_accessibility_highlight_set is using VIEW object as a "target"
object for the ELM_WIDGET_ITEM_CLASS object. And the "target" object is used
for all highlight related lines in the elm_win including highlight_grabbing.
So when the elm_widget sets highlight_grabbing for the ELM_WIDGET_ITEM_CLASS
object, the VIEW object should be used.
Change-Id: I6bd1e40e69883e81f968ea82a6a48be9204206d2
Shinwoo Kim [Mon, 23 Oct 2017 08:53:39 +0000 (17:53 +0900)]
Enhacne documentation
- elm_win.eo
Change-Id: Idd209c7f17ea5e5be362be5a92c18c7fc2487a62
Taehyub Kim [Thu, 26 Oct 2017 02:19:59 +0000 (11:19 +0900)]
Revert "atspi: fix translation domain for naviframe item"
This reverts commit
8878a133735f1a67752051e64298d1650a8aa861.
Change-Id: I5d28b247334752a01e9cd6f42c497cd008a267b9
Taehyub Kim [Thu, 26 Oct 2017 02:19:35 +0000 (11:19 +0900)]
Revert "atspi: fix name, description callback issue"
This reverts commit
1932e8f4831a8762611bee2c0a1ea0befd01f4d2.
Change-Id: I3f79dc65dbccdfded1b028326ebc542ca1bffadf
Taehyub Kim [Thu, 26 Oct 2017 02:19:05 +0000 (11:19 +0900)]
Revert "atspi : Do not send signal, if current object and highlight object are same"
This reverts commit
093c6b99032156e726b88c7ab09b4a54adec4ce1.
I9d64b853a7218269d3498c09ca4c1cef461d0b98
Change-Id: I9e9ac517df185ec45e992aedf88ecab847c52f63
Taehyub Kim [Thu, 26 Oct 2017 02:17:50 +0000 (11:17 +0900)]
Revert "[Accessibility]: Access frame going behind popup issue fix when popup is created in transition,finished callback."
This reverts commit
70f64ff792790cf06528bd1bd98fd37ed05dfda2.
Ic52cb8faa0c122883a2bff33fb853a51f8f54ed5
Change-Id: I5790d4d1f1575a8cd7ec0e160426a4b13003a17d
Jeonghyun Yun [Mon, 23 Oct 2017 11:44:18 +0000 (20:44 +0900)]
elm_url: matched between url and proxy environment
@tizen_fix
Change-Id: Ic46db15739a1c86d39b223aa47cf114d52419f3e
Signed-off-by: Jeonghyun Yun <jh0506.yun@samsung.com>
Shilpa Singh [Wed, 18 Oct 2017 13:31:04 +0000 (19:01 +0530)]
[Accessibility]: Access frame going behind popup issue fix when popup is created in transition,finished callback.
Change-Id: I4e4bdf9c7ce6bd9528311319119c5f8c9e287d17
JunsuChoi [Fri, 20 Oct 2017 05:28:18 +0000 (14:28 +0900)]
atspi : Do not send signal, if current object and highlight object are same
Change-Id: I25a207cbba4562ae364412d1753d9034c96243c3
Shinwoo Kim [Fri, 20 Oct 2017 05:51:28 +0000 (14:51 +0900)]
atspi: fix name, description callback issue
The accessible name, description callback did not work for the naviframe item.
Because the naviframe item has used VIEW object for its accessible object.
If user calls elm_atspi_accessible_name_cb_set(naviframe_item, cb, data),
then the callback cb should set to VIEW object, not naviframe item, because
atspi_accessible_name_get is using the VIEW object.
Change-Id: I8b0d25d36b8e66a45a3757efa900ff9082d6fbf3
Shinwoo Kim [Fri, 20 Oct 2017 01:12:19 +0000 (10:12 +0900)]
atspi: fix translation domain for naviframe item
The naviframe item have used its VIEW object for accessible name and
description. But the naviframe item did not care about the translation
domain. So there was a translation problem.
Change-Id: I3163f4b078a62c82686806f265487c9a3082abc9
Sungtaek Hong [Thu, 19 Oct 2017 08:30:49 +0000 (17:30 +0900)]
efl_ui_textpath: fix potential memory leak
Change-Id: I7a1b171354ff66bb87a75b38aaa4ea5decea7a6d
Signed-off-by: Sungtaek Hong <sth253.hong@samsung.com>
Jeonghyun Yun [Wed, 18 Oct 2017 22:48:46 +0000 (07:48 +0900)]
elm_scroller: fix scrolling with key move
Use content_region_show instead of content_pos_set in _key_action_move
Summary:
When user keep pressing key down or else on scroller content, scroller
animation is lagging because of elm_interface_scrollable_content_pos_set
by step_x or step_y value. When focus moved to next object by press key
down or else, content_pos_set by ecore_animator continuously. In this
time, content_pos_set in _key_action_move by step_x or step_y value
caused animation lagging problem. I fixed to use content_region_show
instead of content_pos_set in _key_action_move for remove exist
animator.
Test Plan:
1. elementary_test -> Scroller3
2. Press 3 times "Append 10 Items in 3s" button
3. focus to Item1 and keep pressing key_down
Reviewers: jpeg, woohyun
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D5278
Change-Id: Ia112e52594ad1189fe767bcc7bb0c37a82bcc3e6
Signed-off-by: Jeonghyun Yun <jh0506.yun@samsung.com>
JunsuChoi [Fri, 13 Oct 2017 03:13:31 +0000 (12:13 +0900)]
atspi : Change default label append of popup from show_finish_cb to on_show
show_finish_cb is called after animation of popup has finished.
default label time limit for screen-readers is 0.3 seconds.
Default label does not work if animation does not finish in 0.3 seconds
So change it so that it can be processed by smart show.
Change-Id: I1a626a6cf650a0fc05aed22c38fc192267aeb8ef
Thiep Ha [Fri, 1 Sep 2017 02:21:52 +0000 (11:21 +0900)]
migrate from upstream efl_ui_textpath
migrate patches from upstream with adaptation:
b0d185ac127297f5dd73cb8b58966fd0256611d7
172cf67d161626bd97e16a9508e4e803c711c156
b55cb4feb1c35a0a6b31ac1ccbadc76432f581bc
Change-Id: Ia04e81150822b40c2f4ec0b76f40348adeb09c25
Shinwoo Kim [Tue, 17 Oct 2017 02:23:21 +0000 (11:23 +0900)]
Enhance documentation
- elm_entry.eo
- elm_flip.h
- elm_flipselector.h
- elm_config.h
- elm_focus.h
- elm_gen.h
- elm_object.h
- elm_genlist.eo
- elm_genlist.h
- elm_grid.eo
- elm_icon.h
- elm_image.h
- elm_image_legacy.h
- elm_index.h
Change-Id: Ic82c977595f8819f53552b38174815e06268952b
Sungtaek Hong [Wed, 5 Apr 2017 06:43:33 +0000 (15:43 +0900)]
elm_map: fix copyright scalability issue
- Since copyright may have long text, the text cannot be shown properly
in low resolution.
@tizen_fix
Change-Id: I063b386bf4ec2c4b8a1f29d7d3cbff6abed9802e
Signed-off-by: Sungtaek Hong <sth253.hong@samsung.com>
Wonki Kim [Tue, 10 Oct 2017 07:56:44 +0000 (16:56 +0900)]
elm_box: fix a potential null pointer dereferencing in elm_box
Summary:
if 'evas_object_smart_data_get' return null somehow,
logic that dereference the smart data pointer will cause problems.
This patch prevent a potential bug in advance.
Reviewers: jpeg, woohyun, cedric
Differential Revision: https://phab.enlightenment.org/D5290
Change-Id: I3bb685468cc3bdf44bddff1c51988f8ab37e7627
Signed-off-by: Wonki Kim <wonki_.kim@samsung.com>
Shinwoo Kim [Mon, 16 Oct 2017 06:58:56 +0000 (15:58 +0900)]
atspi: not use destructed eo object
It is possible to access destructed eo object using ATSPI interface.
The following sequence is possilbe, because AT-client such as screen-reader
does not know whether a widget object deleted or not.
1. _elm_radio_evas_object_smart_del
2. _elm_radio_elm_interface_atspi_accessible_state_set_get
So before using eo interface, we have to check if the eo object is vlid or
not using eo_destructed_is.
Change-Id: I7ffaf7bd905dcc2ef5770d9c1cef9de52ec77322
Youngbok Shin [Tue, 17 Oct 2017 04:02:43 +0000 (13:02 +0900)]
examples: remove doxygen warning messages
It only remove warning messages which don't need example files.
Change-Id: Ia152012a0a197d7cc0472fefddeeccc090300f0f
Myoungwoon Roy, Kim [Wed, 11 Oct 2017 05:40:10 +0000 (14:40 +0900)]
Remove @ref for Start
Summary:
Since there is no Start page on upstream,
Remove @ref mark for Start.
Change-Id: I88ca7a1f5888c6090ecaf9d49ed7277595dcae74
Woochan Lee [Tue, 10 Oct 2017 10:57:59 +0000 (19:57 +0900)]
Spinner: Fix can't input number in case of min value is bigger than 1.
Min Max validate logic has been changed to support decimal point counting.
It makes this side effect.
Change-Id: Iede1a1ee48b60680387a770eaecd68a499d9eed4
JunsuChoi [Wed, 11 Oct 2017 03:12:45 +0000 (12:12 +0900)]
atspi : During the highlight grab, out signal is not sent.
There is an action that takes time to complete
the highlight grab, such as scroll bring in.
So the highlight box can be outside the screen.
After the highlight grab process is completed,
will check the highlight box status.
Change-Id: I5b9535b9f1ef7a515e3958af07dc0a61f1802782
JunsuChoi [Wed, 11 Oct 2017 13:09:14 +0000 (22:09 +0900)]
atspi : Do not center align when genlist item is highlighted in wearable profile
Change-Id: Ief1158011fd204cb10eed4b3576518dd7d55d64f
JunsuChoi [Wed, 13 Sep 2017 03:04:42 +0000 (12:04 +0900)]
atspi : Add check for chain end object with attribute
When searching for a neighbor object,
check whether there is an "relation_chain_end" attribute.
If it does, it returns the current object.
Change-Id: Ic5f9ded9c9f6a77dae8f940c2271544d6007813d
JunsuChoi [Fri, 29 Sep 2017 04:10:39 +0000 (13:10 +0900)]
atspi : Remove role check when getting an accessible object to a point.
Change-Id: Ie676a19164929d94c95d43044f1e585056d55874
JunsuChoi [Fri, 29 Sep 2017 11:55:32 +0000 (20:55 +0900)]
atspi: Improves how to find the can_highlight of the widget.
Returns EINA_FALSE if any of parents is EINA_FALSE
when determining can_highlight.
To determine can_highlight for the parent of an object,
we try to find the parent of the parent again.
As it repeats, it traverses as much as depth size factorial (d!)
However, if any of the parent of the object has can_highlight of the widget data is EINA_FALSE,
it is eventually EINA_FALSE
Change-Id: I8ad128bba766c97ddf3dd28ad788c0e2dd1d3554
Jeonghyun Yun [Wed, 11 Oct 2017 01:45:57 +0000 (10:45 +0900)]
elm_image: stop preloading before intenal_file_set as it may hit to-be-freed memory
Change-Id: I8df88ac5736058983a5c3db1addd573402b214bc
Signed-off-by: Jeonghyun Yun <jh0506.yun@samsung.com>
JunsuChoi [Fri, 29 Sep 2017 07:35:41 +0000 (16:35 +0900)]
add return check for eldbus
Change-Id: If736b96e08b1ac991a6676743f670ca68ae9672c
Shinwoo Kim [Mon, 25 Sep 2017 13:57:02 +0000 (22:57 +0900)]
atspi: use detail value of window activated signal
The detail value of window activated signal is using for following cases
(1) support "default_label" attribute of window
(2) notify existing keyboard window.
The keyboard window would exist always, once it shows.
So if screen-reader turns off, and turn on again, there was no ATSPI signal.
We are using 'activate' signal for this case to get the reference of
keyboard window. And we are using detail value to reduce a call of IPC.
Change-Id: I11db347b89f283ad2ab7fd5378ca123948dbbaf5
Shinwoo Kim [Sat, 23 Sep 2017 13:34:52 +0000 (22:34 +0900)]
atspi: win overrides atspi component get interface
(1) The efl_ui_win overrides _elm_interface_atspi_component_extents_get to give
correct value if screen_coord is EINA_FALSE which means relative position.
The efl_ui_win has given its object geometry value + ecore evas geometry value.
The object geometry value was equal to the ecore evas geometry value, so the
relative position was not correct(twice of ecore evas geometry value).
(2) Allow ELM_ROLE_INPUT_METHOD_WINDOW to send signal
Change-Id: I1b6832e960b3123d2a817d6253313447bc31af08
Shinwoo Kim [Wed, 27 Sep 2017 09:30:06 +0000 (18:30 +0900)]
atspi: make datetime itself unacceptable object
This depends on UI definition. The datetime itself does not have highlight.
Only each part objects such as inc/dec button, year, month, and day in it.
Change-Id: I6a14dec2a24ee280ad19fa885acec0992acc9400
Shinwoo Kim [Wed, 20 Sep 2017 04:14:27 +0000 (13:14 +0900)]
atspi: fix infinite loop issue
It is possilbe to get a target object as a return value of get_object_at_point.
In this case _elm_widget_elm_interface_atspi_component_accessible_at_point_get
works again with the target which is same with root object.
[Enhanced]
- We do not have to go parent if the bottom up logic meets root object.
- Visiblity of panel is always true. So we are checking its SHOWING state.
If the panel is not SHOWING, remove it from valid children list.
Change-Id: I2edeefdc9a0c5955f1c28926321b74424c71bc40
Shinwoo Kim [Fri, 29 Sep 2017 05:28:21 +0000 (14:28 +0900)]
atspi: set elm_access layer to MAX
Set access object layer to max, because the highlight frame could be covered
by other object which has higher layer.
Change-Id: I16e54469f0c749fb2caab6da9ea68b8dc97d9b02
Shinwoo Kim [Thu, 28 Sep 2017 11:46:44 +0000 (20:46 +0900)]
elementary: atspi - fix dbus abort
This patch set fix following abort occured on dbus side.
(gdb) bt
0 __GI_raise (sig=sig@entry=6)
1 __GI_abort ()
2 _dbus_abort ()
3 _dbus_warn_check_failed
4 dbus_message_iter_append_basic
5 append_basic
6 eldbus_message_iter_basic_append
7 _bridge_signal_send
8 _text_text_removed_send
9 _bridge_accessible_event_dispatch
10 _elm_interface_atspi_accessible_event_emit
11 elm_interface_atspi_accessible_event_emit
12 _entry_changed_user_signal_cb
(gdb) f 12
(gdb) p *$1
$2 = {change = {insert = {content = 0x0, pos = 0, plain_length = 1},
del = {content = 0x0, start = 0, end = 1}},
insert = 0 '\000', merge = 0 '\000'}
The following patch set would fix the root cause of this problem.
https://phab.enlightenment.org/D5240
Change-Id: Ifea8c2fead631f06b7d54ce9a6517f795b3cbf06
Shinwoo Kim [Sat, 23 Sep 2017 08:42:55 +0000 (17:42 +0900)]
atspi: block event which is not interested in
The following ATSPI dbus signal is not interested in.
Moreover the following ATSPI dbus signal occurs a lot at init time.
This could cause a performance issue.
- 'ChildrenChanged' with added child
- 'ChildrenChanged' with deleted child
- 'AddAccessible'
- 'RemoveAccessible'
Change-Id: Iebfb4ef75b5f32bf025e275dd9b548bd3581b2de
Shinwoo Kim [Thu, 21 Sep 2017 07:40:49 +0000 (16:40 +0900)]
atspi: support "default_label" attribute of window
Wearable profile application does not want to use default label functionality
because it is hard to meet application UI definition.
So if the window has attribute "default_label" key, and its "disabled" value,
then screen-reader does not read out default label, and does not clear
currently highlight object.
Application should append attribute as below.
elm_atspi_accessible_attribute_append(window, "default_label", "disabled");
This patch set works with following patch set.
https://review.tizen.org/gerrit/151570 (screen-reader)
Change-Id: I354c6ecf30db3a0011658d69577ad384fcb900db
Shinwoo Kim [Tue, 19 Sep 2017 11:28:12 +0000 (20:28 +0900)]
atspi: Handle default label object on toolkit side
There are unexpected difficult cases to make default label work correctly.
The following would be the case.
(top of accessible tree) - (bottom side)
PageTab1 - Panel1 - PageTab2 - Panel2 - PageTab3 - Panel3(currently showing)
Application could make as below:
PageTab1 - Panel1 - PageTab3 - Panel3 - PageTab2 - Panel2(currently showing)
or following tree would normally be possilbe:
PageTab1 - Panel1 - PageTab2 - Panel2(currently showing)
There are much of complicated case over there.
So we are handling the default label object stack on toolkit(Elementary) side.
Please refer to following patch set:
https://review.tizen.org/gerrit/#/c/151025/ (at-spi2-core)
https://review.tizen.org/gerrit/#/c/151030/ (screen-reader)
Change-Id: I0834f9d0625f4e787d6315ce4858bdec4a67ab18
Shinwoo Kim [Sat, 16 Sep 2017 08:09:33 +0000 (17:09 +0900)]
atspi: enhance line related to highlight frame
1. Before elm_widget_theme_object_set for a11y highlight frame was called in
_elm_win_accessibility_highlight_show which is called in highlighted object
move, resize callback.
2. Before evas_object_smart_changed was not called even though a11y highlight
frame is added as a smart member of highlighted object.
※ Refer to commit: c3b5da9 (atspi: make highlight frame be a smart member)
Change-Id: I1f5dc367d2fc5565d61ae12d44fb1e8a49eeb979
Hosang Kim [Thu, 28 Sep 2017 07:05:21 +0000 (16:05 +0900)]
elm_need_efret() - improve error handling according to coverity
fix CID 1360466, CID 1353589
we don't actually chekc if efreet inits right with elm_need_efreet()
as we just assume it inits (And there pretty much is no case where it
will not), but make coverity happy and check.
@fix
Change-Id: Ieb37ab35a6ebc3a7e30b9fd5be6bb0e5b06aa64f
(cherry picked from commit
1b28da85f8e76a9052e99e0bc9394cb26a42cdac)
Hosang Kim [Thu, 28 Sep 2017 06:40:44 +0000 (15:40 +0900)]
elm dbus menu - handle return value of eldbus_message_iter_arguments_get
handle possible error return of eldbus_message_iter_arguments_get()
according to coverity CID 1353591. this should fix that.
Change-Id: I213f644c6450e550f5a26fd9a0c11a707d71f94a
(cherry picked from commit
67fda5c6f55e323f72eba9ece0f35ff8822e7215)
Youngbok Shin [Thu, 28 Sep 2017 10:41:56 +0000 (19:41 +0900)]
elementary: fix memory leaks when using elm_prefs
"regex" is heap-allocated and is not handled by regfree().
We must explicitely call free() after a regfree() to remove
the regex_t from memory.
Change-Id: Ib278f4bd49b6246b7d6eaf7f646dd1c36aca4419
Author: Jean Guyomarc'h <jean@guyomarch.bzh>
Date: Fri Aug 12 16:36:12 2016 +0200
Youngbok Shin [Thu, 28 Sep 2017 10:13:05 +0000 (19:13 +0900)]
entry: fix memory leak issue for atspi
@tizne_fix
Change-Id: Ib7820537a74b504939c78848b03cf5f644e49970
Woochan Lee [Thu, 28 Sep 2017 07:47:06 +0000 (16:47 +0900)]
elm_win: prevent Null pointer ref.
_elm_config could be null. in case launch the app using app_launch command in shell.
@tizen_fix
Change-Id: I27006548a32c56fffd6ff9aea98941b5e6036593
Taehyub Kim [Thu, 28 Sep 2017 05:56:20 +0000 (14:56 +0900)]
elm_interface_atspi_accessible: modified since_tizen tags for relationships clear
Change-Id: I772ca0b888cba32ed9d1b99ccddad6c9cc12ca1a
Taehyub Kim [Thu, 28 Sep 2017 06:02:45 +0000 (15:02 +0900)]
elm_config: add since_tizen tags for language auto mirrored functions
Change-Id: I855d620001346327358fffaa25f8073611b2fe5e
JunsuChoi [Mon, 25 Sep 2017 09:58:23 +0000 (18:58 +0900)]
add color for build message(error:red, warning:ylw)
Change-Id: Idb94b1e1ab19a917deb2624c0975288352ee25d1
jiin.moon [Wed, 20 Sep 2017 10:18:07 +0000 (19:18 +0900)]
elm image: Fix async open to avoid multiple mmap
From
cb3b4cc8d7794bea575a85325c8a58f25f0507b2 Mon Sep 17 00:00:00 2001
From: Jean-Philippe Andre <jp.andre@samsung.com>
Date: Fri, 15 Sep 2017 14:39:14 +0900
Subject: [PATCH] elm image: Fix async open to avoid multiple mmap
Reported by @jiin.moon:
In case of async_open for an elm_image, we try and open a file in a
thread, then map it and populate a bit, as this may take some time
(blocking I/O). This creates a mmap with eina_file_map_new. But later
evas image loaders will (usually) try and map the entire file with
eina_file_map_all() which creates another mmap. Since the size is
different (32Kb first then all) the returned map might be different
(it's up to the kernel to decide at this point).
So, in order to avoid having multiple maps on the same file, and try to
reduce the peak memory usage, we should prefer using the same map all
the time, i.e. the global one returned by eina_file_map_all().
This patch relies on the previous patch in eina_file which fixes
eina_file_map_populate() for the global map.
@fix
Change-Id: If27d42b89269493ac2425737198338689687459d
Jaehyun Cho [Wed, 6 Sep 2017 11:04:41 +0000 (20:04 +0900)]
naviframe: Fix to finish transition for newly pushed item
Suppose naviframe top item is "A" and a new item "B" is newly pushed.
In this case, if "A" is deleted by elm_object_item_del() before
transition for "B" is started, then signal for "B" is also not sent and
"B" does not become visible.
The above issue happened if the transition effect was implemented by
using deferred signals (i.e. "pushed,deferred" and "popped,deferred").
To resolve the above issue, the signal only for the deleted item is not
sent.
Suppose naviframe top item is "A" and a new item "B" is newly pushed.
In this case, if "B" is deleted by elm_object_item_del() before
transition for "B" is started, then signal for "A" is still sent and "A"
becomes invisible.
To resolve the above issue, if the deleted item is the top item and it
is in the middle of item push, then all the signals related to the
deleted item are not sent.
Change-Id: I33c9f5039a8d3a3e5c60259a14edae395503ddd8
Wonki Kim [Mon, 11 Sep 2017 04:12:57 +0000 (13:12 +0900)]
interface_scrollable: Cast result of divide explicitly
cast from double to int explicitly to remove smell of code
Change-Id: Ic02d3c24f212be1d71ecb5a8b0e923795e8296d3
Signed-off-by: Wonki Kim <wonki_.kim@samsung.com>
SangHyeon Jade Lee [Thu, 21 Sep 2017 10:52:33 +0000 (19:52 +0900)]
genlist & gengrid : fix build warnings
Change-Id: Ia356f41294f8e0bb84446699a7de87757d1b23b3
Signed-off-by: SangHyeon Jade Lee <dltkdgus1764@gmail.com>
Jeonghyun Yun [Wed, 20 Sep 2017 01:02:56 +0000 (10:02 +0900)]
els_tooltip: Fix to use some functions before tooltip_test_set or tooltip_content_cb_set
Summary:
If user call tooltip_orient_set or tooltip_style_set or tooltip_window_mode_set
before tooltip_test_set or tooltip_content_cb_set, those functions doesn't work.
Because elm_tooltip will be created when tooltip_content_cb_set is called.
I fixed logic to use some functions before tooltip_test_set or tooltip_content_cb_set.
Test Plan: elementary_test -> Popups -> Tooltip
Reviewers: jpeg, Jaehyun
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D5183
Change-Id: Ia058679cd1645a12e048d8c5b0db0566f2742fca
Signed-off-by: Jeonghyun Yun <jh0506.yun@samsung.com>
Shinwoo Kim [Fri, 15 Sep 2017 10:14:19 +0000 (19:14 +0900)]
atspi: clear current highlighted object
The _elm_win_screen_reader does not clear current highlighted object when the
screen reader is disabled. It tried to clear window highlight even though the
window does not have highlight.
Change-Id: Icc5de54361852c0210c0640f2801483f6a2c6468
Doyoun Kang [Fri, 15 Sep 2017 00:30:21 +0000 (09:30 +0900)]
elm_conform: consider window's geometry while reflecting keyboard conformant geometry
There was a bug that the partial window was resized when the keyboard was shown on landscape mode.
So, we add code to consider window's geometry while calculating conformant's geometry.
Change-Id: Iaa895335ddbc78697044f6712e7477317f4d3758
Shinwoo Kim [Wed, 13 Sep 2017 12:34:44 +0000 (21:34 +0900)]
atspi: enhance _accessible_at_point_get
The _elm_widget_elm_interface_atspi_component_accessible_at_point_get is
seaching accessible object from bottom of tree.
If elm_popup, or elm_panel comes as a root object, then
the _elm_widget_elm_interface_atspi_component_accessible_at_point_get does not
search from bottom of tree.
To make seach logic consistent, we are using bottom-up approach first for both
elm_popup and elm_panel from this patch.
Change-Id: I2a31725e8ff69455b836d6f888e0b42695ad2af5
Doyoun Kang [Tue, 12 Sep 2017 12:20:17 +0000 (21:20 +0900)]
elm_win: do deferred resize job first before calling rotation change callback
there was a bug that the elm_win didn't resize when an user called evas_object_resize
with previous width and height in "wm,rotation,changed" smart callback function.
Change-Id: I8ac58abbb51e4e4135ff742c1c8a77e0c246ac7d
Shinwoo Kim [Tue, 12 Sep 2017 05:44:39 +0000 (14:44 +0900)]
atspi: enhance _accessible_at_point_top_down_get
If there is only one valid child at point, then return it.
The evas_tree_objects_at_xy_get could not find proper object,
if application does not have well aligned objects.
This could be a role of application, but this patch set could solve somehow.
Change-Id: I6326a23b88904f6003a99e8b17885452890c7222
SangHyeon Jade Lee [Mon, 28 Aug 2017 10:55:05 +0000 (19:55 +0900)]
genlist: re-focus on genlist in item focus set when sub-object was focused
Change-Id: I6de90b593dc0e251bc401bdd245dcfb507a90a98
Signed-off-by: SangHyeon Jade Lee <dltkdgus1764@gmail.com>
Shinwoo Kim [Mon, 11 Sep 2017 05:33:41 +0000 (14:33 +0900)]
atspi: enhance po file to translate "Alert"
There was no information to translate "Alert".
Change-Id: I339548b707d2d34df77dcc36774d0a4665854e63
Shinwoo Kim [Mon, 11 Sep 2017 05:29:46 +0000 (14:29 +0900)]
atspi: follow edje_entry logic to send anchor name
The edje_entry has logic remove unnecessary character.
The elm_entry is following the logic to give correct information.
Change-Id: I55114193d9b23117ddc8b9863c2b14741f42a215
JunsuChoi [Fri, 8 Sep 2017 09:13:05 +0000 (18:13 +0900)]
atspi : Fix "bring_in" of the genlist item in wearable profile
Change-Id: I3409ad8d55998d01af28eb248d57a586a94d3cc4
Shinwoo Kim [Sun, 10 Sep 2017 00:50:21 +0000 (09:50 +0900)]
atspi: emit signal after atspi bridge is connected
This patch set emitting popup SHOWING signal, and window ACTIVE signal after
the atspi bridge is connectted.
The SHOWING signal of popup does not work properly, if the popup shows before
the atspi bridge is not connected. In this case, the screen-reader cannot read
popup information. So the SHOWING signal should be emitted after the connection
is established.
The ACTIVE signal does not work properly in the following case.
There are two windows show in one process at the same time. In this case,
the ACTIVE signal was not emitted because window_activated_broadcast_needed is
set to EINA_FALSE. This would be a bug.
Change-Id: Ieb1883c94c2570fdb92b9770a1c20e7d95c27a60
SangHyeon Jade Lee [Mon, 21 Aug 2017 09:48:19 +0000 (18:48 +0900)]
genlist: initialize process size when item is loaded
Change-Id: I6d983a56fae403444ffe55a68e276a5e4e42355e
Signed-off-by: SangHyeon Jade Lee <dltkdgus1764@gmail.com>
Shinwoo Kim [Fri, 8 Sep 2017 07:46:32 +0000 (16:46 +0900)]
atspi: remove unacceptable leaf node
An object could have children which have same position.
In this case, if first child of this object is not acceptable leaf node, then
there is no chance to find second child object.
Change-Id: If6887dc04bce4590ab4123cb79914172b5a4ca48
Jaeun Choi [Fri, 8 Sep 2017 07:15:41 +0000 (16:15 +0900)]
elm_color_class: add missing return value
Change-Id: Ic420433a43cfcd4aecdbe3f405dea27bd18665ef
Shinwoo Kim [Fri, 8 Sep 2017 02:30:30 +0000 (11:30 +0900)]
atspi: expose API
elm_atspi_accessible_relationships_clear
Change-Id: I297d379744f8df17d833d3b4172a0f70aba4e655
Shinwoo Kim [Wed, 6 Sep 2017 11:09:26 +0000 (20:09 +0900)]
atspi: remove memory leak
The _elm_util_mkup_to_text gives alloacated plain text.
This plain text should be freed.
Change-Id: I4fe6443a2be78a80b00e79925b9ac968e489df76
Shinwoo Kim [Tue, 5 Sep 2017 13:42:50 +0000 (22:42 +0900)]
atspi: use entry text and parent accessible name
(1) The entry is using entry text for its accessible name.
(2) It is possible to overwrite default accessible name of entry by using
parent accessible name.
Change-Id: I29cdfe0fd6c517d9611ad46996eebe40c7be27e9
Shinwoo Kim [Tue, 5 Sep 2017 11:53:29 +0000 (20:53 +0900)]
atspi: find valid child of list item
This patch set is solving following case.
If genlist item uses evas_object_image_source_set to show item itself, then
evas_tree_objects_at_xy_get does not return below objects of image source.
So content objects of genlist item could not become cadicdates for
elm_interface_atspi_component_accessible_at_point_get.
@tizen_fix
Change-Id: I44e52b48054cf80fd102131c7f199a6b62a0464d
Shinwoo Kim [Tue, 5 Sep 2017 02:13:28 +0000 (11:13 +0900)]
atspi: use elm_entry to call smart callback
This patch set is using elm_entry to call "anchor,clicked" smart callback.
The elm_access for anchor has been used so far, so the "anchor,clicked" smart
callback did not work.
Change-Id: If09400e78a9fc488e307fd46e67b39bf6beeca19
JunsuChoi [Wed, 9 Aug 2017 07:41:47 +0000 (16:41 +0900)]
[atspi][panel] Adds accessibility support for the panel widget.
Panel handles similar popups.
When open or close it sends signal to screen-reader.
Change-Id: I9803c5d49507bd6deeadef195a609b6a22ffe511
Jaeun Choi [Tue, 5 Sep 2017 07:14:37 +0000 (16:14 +0900)]
elm_panel: set content as a sub object of panel object
since sd->bx is not an elm object, it is excluded from parent-child tree.
in case of a non-scrollable panel, the content needs to be attached to the tree
as a child of panel object directly.
Change-Id: I32b65fc2e2b213e510c3060ac877a4361653191a
origin: upstream (
61381f6e3584118a70565eea79c355dd049b9e28)
JunsuChoi [Thu, 31 Aug 2017 05:56:37 +0000 (14:56 +0900)]
[ATSPI] Add to find genlist item using proxy image
In a wearable profile, genlist displays the item using a proxy image.
When a screen-reader finds an object in the top_down, it can not find the item.
because, found item is proxy image.
So, using evas_object_image_source_get and gets item view object from proxy image.
Change-Id: Ia4ad0e7fa214c5c7c01ca33d4fe484c6e76a9673
Shinwoo Kim [Fri, 1 Sep 2017 07:35:55 +0000 (16:35 +0900)]
atspi: make highlight frame be a smart member
Make highlight frame object be a smart member of target object.
There are two cases covered by this commit.
1. ctxpopup of copy & paste module.
The ctxpopup item has higher object layer than highlight frame object.
In this case, the highlight frame is covered by the ctxpopup item.
2. genlist item proxy object.
The genlist item shows its proxy object(handled by evas_object_image_source*).
In this case, the highlight frame has improper position, and size.
Change-Id: I2b6470a8a695ea3739ddf80832e837dbd2c0d1f8
Myoungwoon Roy, Kim [Thu, 24 Aug 2017 04:57:05 +0000 (13:57 +0900)]
docs: Add information about texture size limitations
Summary: I had added some information about texture size limitations to Elm image API reference.
Test Plan: Doxygen Revision
Reviewers: raster, cedric, stefan, jpeg, Jaehyun_Cho
Reviewed By: raster
Differential Revision: https://phab.enlightenment.org/D5106
Change-Id: I5372b83b4e9e736813a679da5c885a95b273a0f8
Woochan Lee [Thu, 24 Aug 2017 04:27:47 +0000 (13:27 +0900)]
elm_spinner: Maintainance API internal logic.
Change fmt value in the API can break compatibility.
Change-Id: I94e2c9832f907c714653a43aa63cc9ac6c2584eb
Woochan Lee [Wed, 23 Aug 2017 11:40:25 +0000 (20:40 +0900)]
elm_spinner: Support min,max filter for float format case.
Summary: The previous min, max filter doesn't support float format case.
Test Plan: elementary_test -> spinner.
Reviewers: jpeg, cedric, woohyun, myoungwoon
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D5016
Change-Id: I60f97c1e8e27f9c3d09f85a5a4e0ccbb347d28e2
Woochan Lee [Wed, 23 Aug 2017 10:52:14 +0000 (19:52 +0900)]
elm_datetime: Add 'O' character to ignore POSIX alternative format in format.
Summary:
_parse_format() function filtering "_/-/0/^/*" to filter POSIX alternatvie/extension formats,
But the Datetime do not appear in some locale(fa_IR, lzh_TW, my_MM, or_IN) with 'O'.
+ improve code. (not compare with all of the characters. instead of that using strchr)
https://lh.2xlibre.net/values/d_t_fmt/
@fix
Test Plan: Set locale as fa_IR and create DateTime. and see fields, not appears.
Reviewers: jpeg, cedric
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D5020
Change-Id: I5c64f301324834b771c452c9cf39da6faa782274
Woochan Lee [Wed, 23 Aug 2017 11:02:38 +0000 (20:02 +0900)]
elm_datetime: Parses the format recursively.
Summary:
Some of locale formats convert as extension format.
For example,
uk_UA d_t_fmt is "%a, %d-%b-%Y %X %z".
The %X will convert as %T.
The %T format has to convert one more time to convert as %H:%M:%S.
ref : https://lh.2xlibre.net/locale/uk_UA/
We need to parse the format recursively to support that kind of case.
@fix
Test Plan:
Change the locale as uk_UA.
Run elementary_test -> datetime
See the time field is not appear.
Reviewers: jpeg, cedric
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D5021
Change-Id: Ie088a4558d3688542e9bcf43dacd1a9a339b433e