| Age | Commit message (Collapse) | Author |
|---|
|
Accessible's `IAccessibleComponent` interface [1] has no
`grabFocus` method and `CAccComponentBase::grabFocus`.
Therefore, drop it.
The `IAccessible` interface has an `IAccessible::accSelect`
method that can be used to request focus by passing the
`SELFLAG_TAKEFOCUS` flag [3], but that is implemented in
`CMAccessible::accSelect`.
[1] https://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/interface_i_accessible_component.html#a8709fdbf3f51ae079af274854a8bffd4
[2] https://learn.microsoft.com/en-us/windows/win32/api/oleacc/nf-oleacc-iaccessible-accselect
[3] https://learn.microsoft.com/en-us/windows/win32/winauto/selflag
Change-Id: I1ae99fdbf60aa18a041c086a73f9bd7c6a80cace
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173439
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
While IAccessible's `IAccessibleComponent` interface [1]
has a `get_locationInParent` method and
the implementation in `CAccComponent::get_locationInParent`
calls the base class one,
`CAccComponentBase::get_locationInParent`, there is no
`IAccessibleComponent::get_locationOnScreen` equivalent,
and `CAccComponentBase::get_locationOnScreen` is simply
unused.
Therefore, drop it.
The `IAccessible` interface has a `IAccessible::accLocation`
method to return the location on screen, but that is
implemented in `CMAccessible::accLocation`.
Therefore, drop
`CAccComponentBase::get_locationOnScreen`.
[1] https://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/interface_i_accessible_component.html#a8709fdbf3f51ae079af274854a8bffd4
[2] https://learn.microsoft.com/en-us/windows/win32/api/oleacc/nf-oleacc-iaccessible-acclocation
Change-Id: I1a38844101b4eca10afdc6356c46156ad2091f82
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173438
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
In `AccObjectWinManager::InsertAccObj`, when no HWND
was explicitly passed, it is attempted to retrieve the
HWND from the parent's `AccObject`.
If no parent AccObject exists yet, create/insert that
one first.
This fixes the following assert seen when using NVDA's SayAll
feature to read the whole document with sample document mentioned in NVDA
issue [1], also attached as attachment 196385 in tdf#91739
for reference:
Assertion failed: pIAccessible && "Couldn't retrieve IAccessible object for relation target.", file C:/tools/cygwin/home/user/development/git/libreoffice/winaccessibility/source/UAccCOM/MAccessible.cxx, line 2755
It's also reproducible when manually trying to get the next
flows-to relation target via NVDA's Python console when the
first cell in the table has focus:
>>> focus.flowsTo
The issue is not reproducible when focus had been in the
second cell before, as an accessible object has already
been created otherwise in that case.
Backtrace:
1 abort ucrtbase 0x7fff3b72286e
2 get_wpgmptr ucrtbase 0x7fff3b7241b5
3 wassert ucrtbase 0x7fff3b7244f1
4 CMAccessible::get_relationTargetsOfType MAccessible.cxx 2755 0x7fff1e028729
5 NdrSendReceive RPCRT4 0x7fff3d4ca2d3
6 NdrStubCall2 RPCRT4 0x7fff3d4625a7
7 CStdStubBuffer_Invoke combase 0x7fff3cfbc4ac
8 RoGetAgileReference combase 0x7fff3cf669c3
9 RoGetAgileReference combase 0x7fff3cf6674e
10 HSTRING_UserSize combase 0x7fff3cfbefb6
11 DllGetClassObject combase 0x7fff3cf470b3
12 CoWaitForMultipleHandles combase 0x7fff3cf6774d
13 RoGetActivatableClassRegistration combase 0x7fff3cf2eb26
14 CoGetMarshalSizeMax combase 0x7fff3cf5cfba
15 CallWindowProcW USER32 0x7fff3ccbef5c
16 DispatchMessageW USER32 0x7fff3ccbe684
17 ImplSalDispatchMessage salinst.cxx 475 0x7ffed452d378
18 ImplSalYield salinst.cxx 552 0x7ffed452da9d
19 WinSalInstance::DoYield salinst.cxx 581 0x7ffed452cfa1
20 ImplYield svapp.cxx 385 0x7ffed78befc4
21 Application::Yield svapp.cxx 474 0x7ffed78c2cd2
22 Application::Execute svapp.cxx 361 0x7ffed78bc656
23 desktop::Desktop::Main app.cxx 1691 0x7ffeecf689b7
24 ImplSVMain svmain.cxx 228 0x7ffed78d3d4c
25 SVMain svmain.cxx 261 0x7ffed78d45a2
26 soffice_main sofficemain.cxx 121 0x7ffeecfb9064
27 sal_main main.c 51 0x7ff782681013
28 main main.c 49 0x7ff78268105a
29 __scrt_common_main_seh exe_common.inl 288 0x7ff782681344
30 BaseThreadInitThunk KERNEL32 0x7fff3bb07374
31 RtlUserThreadStart ntdll 0x7fff3d85cc91
[1] https://github.com/nvaccess/nvda/issues/8152#issuecomment-2342167620
Change-Id: I246251f06d1885e0da96600ffc7dd0549854382f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173224
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
The `AccObject` ctor already asserts it's non-null,
so there's no need to check again in various places.
Change-Id: Ibba363d326198a89e391a4da54cff25c95eef776
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173223
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
... and rename to use the "x" prefix for this uno::Reference.
Change-Id: Ibf422aa48d6fbada49975fa8c5f43ad6c4c3ddee
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173222
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
... and rename the variable to use the "x" prefix
in line with naming conventions.
Change-Id: Ife49161162654efdf1d543909218e31e464d2fc3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173221
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Don't fallthrough from the `AccessibleEventId::TEXT_CHANGED`
case to the `AccessibleEventId::VISIBLE_DATA_CHANGED` one.
commit 3ec901fab2cb3e1a1c8222498761e72c28175bd3
Date: Wed May 25 08:38:19 2016 +0200
Mark dubious fallthrough cases as "SAL_FALLTHROUGH; //TODO ???" (in Win-only)
had already added a TODO comment.
The IAccessible2 IDL describes `IA2_EVENT_VISIBLE_DATA_CHANGED` like
this:
/** A visible data event indicates the change of the visual appearance
of an accessible object. This includes for example most of the
attributes available via the IAccessibleComponent interface.
*/
IA2_EVENT_VISIBLE_DATA_CHANGED,
I see no real reason to always send that event as well
when text is updated.
In particular, since the description of the UNO equivalent
`AccessibleEventId::VISIBLE_DATA_CHANGED` matches the IA2
one very closely, it should explicitly be sent by the corresponding
XAccessible(Context) implementation in addition to the
`AccessibleEventId::TEXT_CHANGED` event if there are cases where
both events should actually be emitted on the platform layer
rather than being implicitly added in the platform-specific a11y bridge.
Change-Id: I00402e08278b0b0c70f8d784939344a952dfa668
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173120
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
When the text changes, don't only send an
MSAA/IAccessible `EVENT_OBJECT_VALUECHANGE` [1]
event, but also an IAccessible2 `IA2_EVENT_TEXT_UPDATED` [2]
event in the Windows a11y bridge.
As described at [2]:
> This event indicates general text changes, i.e. changes to
> text that are exposed through the IAccessibleText interface.
This e.g. makes that event type be sent when the value/text in
one of the editable combobox in Writer's "Formatting" toolbar
changes, e.g. the font size, which exposes its current text
via the IAccessibleText interface.
[1] https://learn.microsoft.com/en-us/windows/win32/winauto/event-constants
[2] https://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/_accessible_event_i_d_8idl.html#ae26846b6d521727ab696d20c3f43c0b5af9d3d9355e2cfd10b126d21513241609
Change-Id: I83d3f2299c59bd9da4246723f99d1c55a351e6d8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173080
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Send `IA2_EVENT_TEXT_UPDATED instead of the deprecated
`IA2_EVENT_TEXT_CHANGED` event. Quoting from the
IAccessible 2 doc [1]:
> IA2_EVENT_TEXT_CHANGED: Deprecated. This event is
> equivalent to IA2_EVENT_TEXT_UPDATED.
[1] https://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/_accessible_event_i_d_8idl.html#ae26846b6d521727ab696d20c3f43c0b5af9d3d9355e2cfd10b126d21513241609
Change-Id: Iec1ba33ba4591c7608438dfdf7ac597e1746fc52
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/173079
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Change-Id: I6a92f442e6540714f8f8f652d67a213e28a1f35a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172650
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The param is never explicitly passed, so is always
nullptr.
Change-Id: I7838a9e5e990ebef18f9f622e587aa2a64070e92
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172649
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: I95c591e2b79e59a50671967e6ab35b384d8d6c23
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172648
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
IAccessible2 PR [1] was merged now as upstream commit [2]
commit 2b8c2c79417bad4b464761a142fab45ffde8bfa8
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Thu Aug 29 07:55:56 2024 +0200
Add an "id" object attribute. (#31)
, so is part of the official IAccessible2 object
attribute specification now.
[1] https://github.com/LinuxA11y/IAccessible2/pull/31
[2] https://github.com/LinuxA11y/IAccessible2/commit/2b8c2c79417bad4b464761a142fab45ffde8bfa8
Change-Id: Ia613a673b832d35bcc6e3801325dc9f55b393a2c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172585
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Report the accessible ID as an "id" object attribute.
Firefox also already uses that object attribute.
Pending PR suggesting to add the "id" attribute to the
official IAccessible2 object attribute specification: [1]
Sample use in NVDA's Python console, with the caret/focus
in the Spelling dialog's "Not in Dictionary" edit:
>>> focus.parent.IAccessibleObject.attributes
'id:sentence;'
[1] https://github.com/LinuxA11y/IAccessible2/pull/31
Change-Id: I42250c3f46afbb2ab64c884d8bf5f2bf7d2a5ced
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172538
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Implement the `IAccessible2_2::get_relationTargetsOfType`
method [1] in the Windows accessibility bridge.
As described in NVDA issue comment [2], this method is
the preferred way that NVDA uses to get e.g. the target
object of the "flowsTo" relation.
With this commit in place, retrieving the next Writer
paragraph in the NVDA Python console using that relation
now also works with the fallback code path in NVDA [3]
dropped in a local build of NVDA:
>>> focus.flowsTo
<NVDAObjects.Dynamic_SymphonyParagraphSymphonyTextEditableTextWithAutoSelectDetectionIAccessible object at 0x0B8F5890>
[1] https://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/interface_i_accessible2__2.html#a63f214b322c663caf01d4bb67277f5af
[2] https://github.com/nvaccess/nvda/issues/4115#issuecomment-2299084993
[3] https://github.com/nvaccess/nvda/blob/df6bd0770db12ff73905251054d12f01911030dd/source/NVDAObjects/IAccessible/__init__.py#L1861-L1874
Change-Id: Ifcd15527b4a3f5302168942f19248a0ba9a4d306
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172258
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Derive from the `IAccessible2_2` insterface [1] instead of
directly from (its base interface) `IAccessible2` and
add dummy implementations for the corresponding methods
that return `E_NOTIMPL` to indicate that there's no
real implementation yet.
An actual implementations for at least
`CMAccessible::get_relationTargetsOfType`
is planned for a separate upcoming commit.
Support for the interface can be checked e.g.
in Accerciser's Python console:
Before:
>>> from comInterfaces import IAccessible2Lib as IA2
>>> focus.IAccessibleObject
<POINTER(IAccessible2) ptr=0x9667f24 at 6841850>
>>> focus.IAccessibleObject.QueryInterface(IA2.IAccessible2)
<POINTER(IAccessible2) ptr=0x9667f24 at 6822120>
>>> focus.IAccessibleObject.QueryInterface(IA2.IAccessible2_2)
Traceback (most recent call last):
File "<console>", line 1, in <module>
File "C:\tools\cygwin\home\user\development\git\nvda\.venv\Lib\site-packages\comtypes\__init__.py", line 831, in QueryInterface
self.__com_QueryInterface(byref(iid), byref(p))
File "C:\tools\cygwin\home\user\development\git\nvda\source\monkeyPatches\comtypesMonkeyPatches.py", line 33, in __call__
return super().__call__(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
_ctypes.COMError: (-2147467262, 'No such interface supported', (None, None, None, 0, None))
With this commit in place:
>>> from comInterfaces import IAccessible2Lib as IA2
>>> focus.IAccessibleObject
<POINTER(IAccessible2) ptr=0x988c784 at 6363580>
>>> focus.IAccessibleObject.QueryInterface(IA2.IAccessible2)
<POINTER(IAccessible2) ptr=0x988c784 at c55d7b0>
>>> focus.IAccessibleObject.QueryInterface(IA2.IAccessible2_2)
<POINTER(IAccessible2_2) ptr=0x974d824 at 6847f80>
And calling one of the methods gives a "Not Implemented" COMError
as expected:
>>> ia2_2 = focus.IAccessibleObject.QueryInterface(IA2.IAccessible2_2)
>>> ia2_2.attribute("dummy")
Traceback (most recent call last):
File "<console>", line 1, in <module>
File "C:\tools\cygwin\home\user\development\git\nvda\.venv\Lib\site-packages\comtypes\_memberspec.py", line 482, in __call__
return self.fget(self.instance, *args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
_ctypes.COMError: (-2147467263, 'Not implemented', (None, None, None, 0, None))
[1] https://github.com/LinuxA11y/IAccessible2/blob/34f96c2a28f4974ad488703b41599a6c6059a7e0/api/Accessible2_2.idl
Change-Id: I77bed1f27353c4d08229ffe17afee03b813a4d4f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172256
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Use switch/case to explicitly map from UNO's
`AccessibleRelationType` to the IAccessible2 relation
type strings, instead of implicitly using the integer
values of the `AccessibleRelationType` as indices into
a map.
Let the caller allocate the string to return.
Use `sal_Int16` for the `AccessibleRelationType` param
to match the type of the actual integer values.
Change-Id: I8c094d763ca602131a578f20719d00cad53088d0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172257
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The methods have actual implementations, not just empty dummy
implementations by now, so this
// These methods only declare here, and their implementation bodies are empty now.
is misleading, maybe was true in an earlier stage of the process
of implementing IAccessible2 support?
Change-Id: I4a50f7199e18766947322d040cac52c0585c2567
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172255
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
As
commit 02c0e5f970dee4d064fcb26b378cbfe6f3a1bc05
Author: Mike Kaganski <mike.kaganski@collabora.com>
Date: Wed Aug 21 13:42:28 2024 +0500
Simplify BSTR construction from OUString a bit
did in other existing code, switch to using
`sal::systools::BStr::newBSTR` in this case newly
added in
commit dfd23124096b9428f2f7084a24de032750389c79
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Wed Aug 21 10:16:51 2024 +0100
wina11y: Retrieve default action description on demand
, merged shortly after existing code was adjusted.
Change-Id: Ic27a31ce22410d6d976295336bf97ada3881788b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172236
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
When retrieving the `IAccessible` for the relation
target's `XAccessible` and
`CMAccessible::get_IAccessibleFromXAccessible` initially
returns nullptr, insert the a corresponding object
and retrieve it then.
The same is already done in most places calling
`CMAccessible::get_IAccessibleFromXAccessible`, e.g.
`CAccTableCell::get_columnHeaderCells` or
`CEnumVariant::Next`.
This makes sure that a proper `IAccessible` is
returned instead of triggering an error if no
such object has been created yet.
This is the case e.g. for off-screen paragraph
objects in Writer that can still be accessed
on the UNO a11y level via the
`AccessibleRelationType::CONTENT_FLOWS_TO` relation.
However, when testing with the test docs from tdf#91739,
the issue even occured much earlier, with on-screen
paragraphs already that weren't available via the
FLOWS_TO relation from their preceding paragraphs.
With this commit in place, having NVDA read out
the whole sample documents using the NVDA + Down shortcut
works now for the tdf#91739 sample docs, in the same way
it already does for Orca on Linux.
For more details, see also the previous analysis in
NVDA ticket [1].
[1] https://github.com/nvaccess/nvda/issues/4115#issuecomment-2299084993
Change-Id: I2ef06c6efa368f10dcf5d917d781dabc8ce9a65f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172215
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Instead of having a COM method `CMAccessible::GetUNOInterface`
to get the `CMAccessible::m_xAccessible` for another
`CMAccessible` via COM, simply cast the `IMAccessible`
to `CMAccessible` and access the member directly.
Change-Id: Ic91c6cde3eb1ebb2bfaabdeb0a5e38017233949c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172214
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
In `CMAccessible::GetNavigateChildForDM`:
* reduce scope of local variables and the union
used only in one of the cases
* don't reuse the `pChildXAcc` union member
at the end of the method for something that's
completely unrelated
Change-Id: I2169bbf5a7af3c806181ab284294514ff68a1506
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172213
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Despite the
// initially m_xAction and m_xContext are the same object
// but they may be different once AccObject::UpdateAction() is called?
comment in winaccessibility/source/UAccCOM/MAccessible.h,
`MAccessible::m_xAction` is supposed to point to the same
underlying object as `MAccessible::m_xContext`, see
`AccObject::UpdateAction` (which queries for the
`XAccessibleAction` interface and sets that).
Simplify this by just querying `MAccessible::m_xContext`
for the `XAccessibleAction` iface on demand in those
places where it's needed.
Change-Id: Id3d89213937c2c72a936947da69ca7fddc8f1b09
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172205
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Quoting from the `IAccessible::accDoDefaultAction` doc [1]
for the `varChild` param:
> Specifies whether the default action belongs to the object
> or one of the object's child elements.
If the default object of a child and not the object itself
should be triggered, there's no need for the object
itself to also support actions itself.
Therefore, only perform such a check in the
`CHILDID_SELF` case.
Also, return `S_FALSE` in case there's no action, as this is
not a case of "An unknown or generic error occurred.", which
is the meaning of `E_FAIL` [2].
as other places do
move that check in `CMAccessible::get_accDefaultAction`, if a child
object is specified instead of requesting
[1] https://learn.microsoft.com/en-us/windows/win32/api/oleacc/nf-oleacc-iaccessible-accdodefaultaction
[2] https://learn.microsoft.com/en-us/windows/win32/winauto/return-values
Change-Id: Idcd3ab8a415e73e496cf027506428518a9f4a203
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172184
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Instead of doing extra work to keep
`CMAccessible::m_pszActionDescription` up to date,
move the logic to retrieve the action description
from the underlying `XAccessibleAction` into
`CMAccessible::get_accDefaultAction` and do that
on demand.
This is similar to what was done for the accessible
description in this commit:
commit fcf4a26275d7503835f9aa23cb94938809840300
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Wed Jan 5 13:41:53 2022 +0000
tdf#146306 wina11y: Retrieve accessible desc on demand
As expected, the example mentioned in the commit
message of
Change-Id: I5569f70739135172ddb604f1f1bd958e32626ab0
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Wed Aug 21 09:49:51 2024 +0100
wina11y: Unify handling to set action description
still behaves the same as without this commit in place.
Change-Id: I7d1ea0ce302df955314c8e46c530cc24d30fcd4b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172183
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Drop `AccObject::UpdateDefaultAction` and let
`AccObject::UpdateActionDesc` take care of
setting the action description for all roles.
It's unclear to me
* why the action description would only be set
for particular roles
* why `AccObject::UpdateActionDesc` would call
`AccObject::UpdateDefaultAction` for particular
roles if it was basically setting the description
the same way that `AccObject::UpdateActionDesc`
does directly for other roles (except for
`AccObject::UpdateDefaultAction` not checking
that there's at least one action).
Also simplify the code a bit while at it.
Testing with the "Don't Save" button in the "Save Document?"
dialog seen when pressing Alt+F4 with a modified Writer doc
gives the expected result in NVDA's Python console:
>>> focus.role
<Role.BUTTON: 9>
>>> focus.name
'Don’t Save'
>>> focus.IAccessibleActionObject.nActions()
1
>>> focus.IAccessibleActionObject.description(0)
'press'
If any particular reason for not setting the action description
for particular roles should show up in the future, they can
still be excluded in one central place now.
Change-Id: I5569f70739135172ddb604f1f1bd958e32626ab0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172182
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
For the only in `AccObjectWinManager::InsertAccObj`,
there's already a reference to the `XAccessibleContext`,
so just call `XAccessibleContext::getAccessibleRole`
directly.
Change-Id: I9e0a96d67ac8550b4988de6c3cefc8d6e0b319ef
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172181
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The `AccObject` ctor asserts that `AccObject::m_pIMAcc`
is non-null, so there's no need to check here again.
Change-Id: Ieb80dfa4f7229b63dbf53ed77ae54f8d4b1f5211
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172180
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
The `AccObject` ctor already has an assert
that creating the COM object succeeded,
so be consistent and drop additional checks still
handling the theoretical case that it might have failed
for some reason. (Running out of memory could be one,
but we don't handle that case in other places that construct
simple objects, e.g. using the `new` expression, either.)
Change-Id: Ic71ca8a256f1e224f0050c2ee9b039d8823e577a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172179
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
This makes sure that typedefs like `REFGUID` and `LONG` are
known and can e.g. be processed by Qt Creator's code model.
The missing include was not causing build problems because
the source files including acccommon.h include stdafx.h
first themselves.
Change-Id: Ifd93cdc263471314f6ed777ba71b15f2bff35417
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172178
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
As the method name and comment suggests, this method
was doing more in the past, but that's no more the case.
The only thing it does is doing an assert, but
there's no need to have a separate method for this,
so just do that right away in the (so far) only caller.
Change-Id: Ief41c8d4ec350622eb043340a4b0bd1079d337a1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172177
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Use the known string length into account (and I definitely like to
not rely on the undocumented implementation detail that OUString is
null-terminated).
Change-Id: I0e7478d25e802f8f8e31f02cf8d77613c1325bc9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172170
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
There's already a
if(!pRAcc.is())
{
*accessible = nullptr;
return E_FAIL;
}
further up in that method, so no need to check again.
Change-Id: Ifb855dce067b3bff5a654917e79817da72e156be
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172153
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Instead of doing extra bookkeeping to keep
`CMAccessible::m_pszName` as a class member up
to date with the current value of the accessible
name, just retrieve the current a11y name directly
from the underlying `XAccessible` when it is requested
via the IAccessible2 API in `CMAccessible::get_accName`.
This is basically the accessible name equivalent of the
the following commit that did similarly for the
accessible description:
commit fcf4a26275d7503835f9aa23cb94938809840300
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Wed Jan 5 13:41:53 2022 +0000
tdf#146306 wina11y: Retrieve accessible desc on demand
Adapt 'MAccessible::get_accDescription' to directly
retrieve the accessible description on demand via the
corresponding XAccessible, rather than keeping
track of it in a class member.
This simplifies the handling and makes it
unnecessary to "manually" update the description
on 'accessibility::AccessibleEventId::DESCRIPTION_CHANGED'
events, since the new value will
be queried next time it is needed anyway.
This also fixes the problem that a significant
amount of time was spent generating accessible
descriptions for all newly inserted a11y objects when
entering values into Calc cells with the NVDA
screen reader in use, resulting in several
seconds of delay.
Querying up-to-date values from the underlying
UNO interfaces on demand instead of doing extra
manual bookkeeping in the winaccessibility code
may be possible for more a11y attributes in addition
to the accessible description handled in this commit,
but each one will have to be looked at separately.
This also allows to drop the
`AccContainerEventListener::HandleNameChangedEvent` and
`AccObjectContainerEventListener::HandleStateChangedEvent`
overrides of the base class implementations unnecessary,
as the only thing they did in addition was updating the
a11y name.
This commit has no effect on events sent on the platform
layer, those are still sent as before to notify about the
name change.
Change-Id: I4d1df1d0b73ce963cdaec91d642ff5bbf2b371bc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172148
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Drop `AccObject::SetName` et al. taking a string param
with the new a11y name to set in favor of using
`AccObject::UpdateName` et al. that updates the a11y name
by querying the current value from the underlying
`XAccessible`.
This makes it unnecessary to pass the new value
around when a `AccessibleEventId::NAME_CHANGED`
event is received, and the new name has to
already be set on the a11y object at the point in
time that the event is received.
Change-Id: I1510bde22ef6d1271f6aeaffe245441cd9e11f19
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172145
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Let `AccEventListener::GetRole` return
`AccessibleRole::UNKNOWN` in case there's no
accessible object instead of -1.
The latter doesn't map to any a11y role.
This doesn't make any difference in practice right now,
as none of the callers explicitly handles the UNKNOWN
role either, but could e.g. simplify switching
`AccessibleRole` from an integer type to an enum in the
future.
Change-Id: I8b0d78642621ce997d55d12905e31e7978962ec7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172124
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The ARIA spec prohibits setting an a11y name for paragraphs [1],
which may have been the reason why an empty name was always
explicitly set in `AccObject::UpdateName`, and that was just
extended to apply for the block quote role when that new role
was added in
commit f44eae4912e8f2ba34d93a71a609473df976a2cb
Author: Michael Weghorn <m.weghorn@posteo.de>
Date: Fri Oct 27 14:53:59 2023 +0200
tdf#135586 a11y: Add a new BLOCK_QUOTE a11y role
Wile it makes sense to always report an empty a11y
name for paragraphs, ensuring that is not the task of the
platform a11y bridge, but it should be done properly in
the underlying `XAccessibleContext` implementation already,
as this is not specific to Windows/IAccessible2.
At least paragraphs in Writer already report an empty
a11y name. If anything else doesn't, that should be
fixed there.
[1] https://w3c.github.io/aria/#paragraph
Change-Id: I7ffd42acc02fc41cf73174e60c32e8938161c4d3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172123
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
It's unclear to me why an a11y object of a TEXT_FRAME
role with a SCROLL_PANE parent should be treated specially
and use the a11y name of its parent.
But the a11y name set this way is unconditionally overwritten
right away in the lines below anyway.
Therefore, just drop the special handling.
Change-Id: I3cfe63e0882803524f51fe760d38880b30105317
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172122
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
The extra null check for the pointer is also redundant,
as there is a check for `xContext.is()` just above it.
Change-Id: I9c0d0ab95c3031d8560506691e78e7994c5dce95
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172121
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
The first thing that
`AccTopWindowListener::AddAllListeners` does is to
retrieve the accessible context and it returns if
it's not valid, so there's no need to do check that
already before calling the method.
Change-Id: Ie825a9694b5c94bc9f9f2917ba5ae85e50670a66
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172120
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: If373619e6b8203d2f2f88fac4d834d44cb6e346d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172119
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The "m" prefix is for class members. Using it for
local variable names is misleading.
Change-Id: I79d62f59ac1093fcd9160ea4d8ae097d32eb40d8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172118
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Change-Id: I192fe62a666c86654b0352d202675c33a72bf89e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172152
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Returning S_OK certainly won't throw.
Change-Id: Ided99c26efb42cba21889890fa37d652d1e118ad
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172151
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
This in particular ensures that try and catch blocks
are indented by an additional indentation level
as expected, increasing readability.
Change-Id: Ic2c3c1500c07df75e3b606ab957a44b220f52194
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/172150
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
This in particular ensures that try and catch blocks
are indented by an additional indentation level
as expected, increasing readability.
Change-Id: I82af846dfe9707895d146406b6ed79b5a1f8a467
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/171798
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
This in particular ensures that try and catch blocks
are indented by an additional indentation level
as expected, increasing readability.
Change-Id: I8cfe89e80b9d8713a69037f1f5bda9ee7eaf782a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/171797
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
As described in the IAccessible2 spec [1], -1 can be used as a
special text offset:
> Using IA2_TEXT_OFFSET_LENGTH (-1) as an offset in any of the
> IAccessibleText or IAccessibleEditableText methods is the same
> as specifying the length of the string.
Replace -1 by the text length *before* doing the
check whether end offset is smaller than the start offset.
Otherwise, trying to query the whole text of a Writer paragraph
containing the text "hello" in NVDA's Python console would
incorrectly trigger an error:
>>> focus.IAccessibleTextObject.text(0,-1)
Traceback (most recent call last):
File "<console>", line 1, in <module>
File "comtypes\__init__.pyc", line 856, in __call__
File "monkeyPatches\comtypesMonkeyPatches.pyc", line 32, in __call__
_ctypes.COMError: (-2147467259, 'Unspecified error', (None, None, None, 0, None))
With this commit in place, it works as expected:
>>> focus.IAccessibleTextObject.text(0,-1)
'hello'
[1] https://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/_general_info.html#_specialOffsets
Change-Id: I489a42270a56178cc8ee0564eec3dc82e15969c4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/168853
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Change-Id: Ia7abdd7fe1fbc5b1a6eaecf7d842852bc3815e6e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/167201
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
Tested-by: Jenkins
|
|
Change-Id: I9659627ce90d6e23bbb3c27e01c365f1167b39ad
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/167171
Tested-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Reviewed-by: Caolán McNamara <caolan.mcnamara@collabora.com>
|
