1. nvda-devel@groups.io
   Date   
Re: Finding components on an application from addon - asking for stategies and insights to do it

Marlon Brandão de Sousa
 

I use and appreciate your addon. This question was exactly because I wanted to send a pull request to it. Glad your are here.


I know VS Code is an electron application. What I do not know is how to, via NVDA, navigate through the accessibility tree to find and read what I need. I also do not know how to access the accessibility tree so that I know beforehand what I am looking for. This is what I am trying to figure out so that I can collaborate with addons that will work on vscode, discord and postman and other apps.


Do you have any insight?

On 12/06/2020 08:28, Pawel Urbanski wrote:
I've worked on an VS Code add-on. VSC is an Electron application. I
guess you could read though the code of Development Kit by Andy who
recently used Beautiful Soup library to parse HTML. The line number is
in teh status bar, which itself is hosted in some DIV container.
I will try to figure out if I could add line announcement to my add-on
by using textInfos offsets.

On 12/06/2020, Marlon Brandão de Sousa <splyt.lists@gmail.com> wrote:
Hello,

I have a need for an addom. Basically, in VSCode, I many times need to
know the line / column I am at the moment.

I know that performing now a read status line will give me this
information, but there are many other things that I don't need on the
status bar when quickly in needs to know the line / collumn positions.

Further more, this is a specific case for a generic question so that
others can also learn the strategy to achieve the goals I am trying to
reach at the moment.

Basically, when scripting an addon, we are trying to do one of three
things:
- Modify the way a given component is handled by NVDA so that it is read
correctly.
- Query information from parts of the application and present them in an
easier way for the addon user, so that they can gater needed information
that is visually easy to get but that without sight would require the
user to perform object navigation and to visit several parts of the UI
to get.
- Similarly to the above, perform clicks or other manipulations on the
UI to achieve actions that are common to the work flow of the
application but that are either inaccessible via keyboard or very hard
to perform in a productive way, many times making an efficient use of
that application not viable at all.

My need with VS Code can be classified at the item 2 of the above list.

Unfortunately I get pretty lost when achieving second and third items.
The first is achievable by using an overlay class thus overwiting the
component NVDA uses to handle a given component. About The second and
third would involve navigating through the window structure and
searching components by some kind of id and then querying information or
sending actions too that component.

In the old times when NVDA didn't exist other screen readers offered
higher level API's to do just this kind of thing. Speaking of JAWS, the
home row utility helped a lot when exploring an application internal
structure so that we could try to find an id, a class, a Window name and
such and then search for that identifier to get access to the wanted
component state. How can I do something similar via NVDA addon? Is there
any addon, any piece of the source code I should look into to start
figuring it out?


Thanks,

Marlon
Re: Finding components on an application from addon - asking for stategies and insights to do it

Pawel Urbanski
 

I've worked on an VS Code add-on. VSC is an Electron application. I
guess you could read though the code of Development Kit by Andy who
recently used Beautiful Soup library to parse HTML. The line number is
in teh status bar, which itself is hosted in some DIV container.
I will try to figure out if I could add line announcement to my add-on
by using textInfos offsets.

On 12/06/2020, Marlon Brandão de Sousa <splyt.lists@gmail.com> wrote:
Hello,

I have a need for an addom. Basically, in VSCode, I many times need to
know the line / column I am at the moment.

I know that performing now a read status line will give me this
information, but there are many other things that I don't need on the
status bar when quickly in needs to know the line / collumn positions.

Further more, this is a specific case for a generic question so that
others can also learn the strategy to achieve the goals I am trying to
reach at the moment.

Basically, when scripting an addon, we are trying to do one of three
things:
- Modify the way a given component is handled by NVDA so that it is read
correctly.
- Query information from parts of the application and present them in an
easier way for the addon user, so that they can gater needed information
that is visually easy to get but that without sight would require the
user to perform object navigation and to visit several parts of the UI
to get.
- Similarly to the above, perform clicks or other manipulations on the
UI to achieve actions that are common to the work flow of the
application but that are either inaccessible via keyboard or very hard
to perform in a productive way, many times making an efficient use of
that application not viable at all.

My need with VS Code can be classified at the item 2 of the above list.

Unfortunately I get pretty lost when achieving second and third items.
The first is achievable by using an overlay class thus overwiting the
component NVDA uses to handle a given component. About The second and
third would involve navigating through the window structure and
searching components by some kind of id and then querying information or
sending actions too that component.

In the old times when NVDA didn't exist other screen readers offered
higher level API's to do just this kind of thing. Speaking of JAWS, the
home row utility helped a lot when exploring an application internal
structure so that we could try to find an id, a class, a Window name and
such and then search for that identifier to get access to the wanted
component state. How can I do something similar via NVDA addon? Is there
any addon, any piece of the source code I should look into to start
figuring it out?


Thanks,

Marlon
Finding components on an application from addon - asking for stategies and insights to do it

Marlon Brandão de Sousa
 

Hello,

I have a need for an addom. Basically, in VSCode, I many times need to know the line / column I am at the moment.

I know that performing now a read status line will give me this information, but there are many other things that I don't need on the status bar when quickly in needs to know the line / collumn positions.

Further more, this is a specific case for a generic question so that others can also learn the strategy to achieve the goals I am trying to reach at the moment.

Basically, when scripting an addon, we are trying to do one of three things:
- Modify the way a given component is handled by NVDA so that it is read correctly.
- Query information from parts of the application and present them in an easier way for the addon user, so that they can gater needed information that is visually easy to get but that without sight would require the user to perform object navigation and to visit several parts of the UI to get.
- Similarly to the above, perform clicks or other manipulations on the UI to achieve actions that are common to the work flow of the application but that are either inaccessible via keyboard or very hard to perform in a productive way, many times making an efficient use of that application not viable at all.

My need with VS Code can be classified at the item 2 of the above list.

Unfortunately I get pretty lost when achieving second and third items. The first is achievable by using an overlay class thus overwiting the component NVDA uses to handle a given component. About The second and third would involve navigating through the window structure and searching components by some kind of id and then querying information or sending actions too that component.

In the old times when NVDA didn't exist other screen readers offered higher level API's to do just this kind of thing. Speaking of JAWS, the home row utility helped a lot when exploring an application internal structure so that we could try to find an id, a class, a Window name and such and then search for that identifier to get access to the wanted component state. How can I do something similar via NVDA addon? Is there any addon, any piece of the source code I should look into to start figuring it out?


Thanks,

Marlon
Re: Auto language switching - Voice selection

Deenadayalan Moodley
 

Hi,

 

I want to add that I would like to see the ability to specify speech synth, and the voice to be used.  At this stage, if you have a number of voices, NVDA just picks the first one and uses it.  This is not always the best one for the job.

 

I did send a sample to NVDA where even things like Language tags in headings in tables is not being honoured.

 

Thanks.

 

From: nvda-devel@groups.io <nvda-devel@groups.io> On Behalf Of Julien Cochuyt
Sent: Thursday, 11 June 2020 17:38
To: nvda-devel@groups.io
Subject: [nvda-devel] Auto language switching - Voice selection

 

Dear all,

 

Does anyone know how to influence the selected voice when automatic language switching occurs?

In case this is synth-specific, the request comes from a customer using Microsoft OneCore as his primary synth.

Still, I'd be personally interested by info for other synth drivers, like eSpeak (my own daily horse) or SAPI 5.

I know there are some efforts made towards generalizing language detection and switching synth altogether (#279 / Dual Voice), but these go far beyond the simple configuration problem my customer is facing, which is comfortably reading his well tagged English/French/German documents.

 


Best regards,

 

Julien Cochuyt

Accessolutions

Joining Speech Chunks

Michael Chen
 

Hello All,


I've noticed with the new speech framework, the last spoken item in the NVDA log is now split into chunks of the spoken text. For example:


Speaking ['User Promoted Notification Area', 'tool bar']


I am used to using regular expressions in the speech dictionaries to customize commonly spoken text. In this example, I previously had a regular expression entry that would silence this text, but now with the "tool bar' chunk split off, I can nolonger match that part of the spoken text.


Is there any way to match the entire spoken list of chunks using the speech dictionary regular expressions?


Thanks for any help.


Sincerely,

Michael
Auto language switching - Voice selection

Julien Cochuyt
 

Dear all,

Does anyone know how to influence the selected voice when automatic language switching occurs?
In case this is synth-specific, the request comes from a customer using Microsoft OneCore as his primary synth.
Still, I'd be personally interested by info for other synth drivers, like eSpeak (my own daily horse) or SAPI 5.

I know there are some efforts made towards generalizing language detection and switching synth altogether (#279 / Dual Voice), but these go far beyond the simple configuration problem my customer is facing, which is comfortably reading his well tagged English/French/German documents.


Best regards,

Julien Cochuyt
Accessolutions

Re: WXPython Submodule Issue When Running From Source

Nathaniel Schmidt
 

Hi Joseph,

Works now. Thanks.

On 10 Jun 2020, at 2:15 am, Joseph Lee <joseph.lee22590@gmail.com> wrote:

Hi,
Because wxPython includes C extension code (pyd file), you'll need to run Python 3.7 specific and 32-bit wxPython 4.0.3.
Cheers,
Joseph

-----Original Message-----
From: nvda-devel@groups.io <nvda-devel@groups.io> On Behalf Of Nathaniel Schmidt
Sent: Thursday, June 4, 2020 1:59 AM
To: nvda-devel@groups.io
Subject: [nvda-devel] WXPython Submodule Issue When Running From Source

Hi all,

I am not sure how simple or difficult this problem is to fix but I am trying to run NVDA from source, and the following problem has been encountered when executing pythonw nvda.pyw from the command line. Is anyone able to help? Also sorry if there is already a topic on this
issue: I have been off the list temporarily. I thought I would post here to see if anyone else is having the issue before creating one on
GitHub:

From log:
---begin---
INFO - __main__ (18:25:49.265) - MainThread (23044):
Starting NVDA version source-master-44d8a31 CRITICAL - __main__ (18:25:49.940) - MainThread (23044):
core failure
Traceback (most recent call last):
File "nvda.pyw", line 214, in <module>
import core
File "core.py", line 31, in <module>
import nvwave
File "nvwave.py", line 15, in <module>
import wx
File "c:\GitInit\njsch\nvda\include\wxPython\wx\__init__.py", line 17, in <module>
from wx.core import *
File "c:\GitInit\njsch\nvda\include\wxPython\wx\core.py", line 12, in <module>
from ._core import *
ModuleNotFoundError: No module named 'wx._core'
--end---

At face-value, this sounds to me like a dependency-specific issue with WXPython's code.
Is anyone else having this issue? Has the classic WXPython submodule been updated recently?

Also, please note that I have set my cloned njsch/nvda/master branch to remote-track nvaccess/nvda/master.

Should I pip uninstall wxpython? I don't think I installed it before successfully creating my python virtual environment for the NVDA source distribution. Although under the circumstances, I don't really see how this would make a difference, since the log history indicates that the nvwave module is looking in the Git submodule-included dependency directory anyway.

Kind regards,

Nathaniel Schmidt
Re: WXPython Submodule Issue When Running From Source

Nathaniel Schmidt
 

Hi Lukasz,

I have both editions installed but I did use the wrong one via the system environment variable...

Nathaniel

On 10 Jun 2020, at 4:31 am, Lukasz Golonka <lukasz.golonka@mailbox.org> wrote:

Hello,

It looks like you've launched NVDA with the wrong version of Python -
NVDA can be started only with 3.7 32-bit.

--
Regards
Lukasz

On Thu, 4 Jun 2020 18:59:15 +1000
"Nathaniel Schmidt" <schmidty2244@gmail.com> wrote:

Hi all,

I am not sure how simple or difficult this problem is to fix but I am
trying to run NVDA from source, and the following problem has been
encountered when executing pythonw nvda.pyw from the command line. Is
anyone able to help? Also sorry if there is already a topic on this
issue: I have been off the list temporarily. I thought I would post
here to see if anyone else is having the issue before creating one on
GitHub:

From log:
---begin---
INFO - __main__ (18:25:49.265) - MainThread (23044):
Starting NVDA version source-master-44d8a31
CRITICAL - __main__ (18:25:49.940) - MainThread (23044):
core failure
Traceback (most recent call last):
File "nvda.pyw", line 214, in <module>
import core
File "core.py", line 31, in <module>
import nvwave
File "nvwave.py", line 15, in <module>
import wx
File "c:\GitInit\njsch\nvda\include\wxPython\wx\__init__.py", line
17, in <module>
from wx.core import *
File "c:\GitInit\njsch\nvda\include\wxPython\wx\core.py", line 12, in <module>
from ._core import *
ModuleNotFoundError: No module named 'wx._core'
--end---

At face-value, this sounds to me like a dependency-specific issue with
WXPython's code.
Is anyone else having this issue? Has the classic WXPython submodule
been updated recently?

Also, please note that I have set my cloned njsch/nvda/master branch
to remote-track nvaccess/nvda/master.

Should I pip uninstall wxpython? I don't think I installed it before
successfully creating my python virtual environment for the NVDA
source distribution. Although under the circumstances, I don't really
see how this would make a difference, since the log history indicates
that the nvwave module is looking in the Git submodule-included
dependency directory anyway.

Kind regards,

Nathaniel Schmidt
Re: WXPython Submodule Issue When Running From Source

Lukasz Golonka
 

Hello,

It looks like you've launched NVDA with the wrong version of Python -
NVDA can be started only with 3.7 32-bit.

--
Regards
Lukasz

On Thu, 4 Jun 2020 18:59:15 +1000
"Nathaniel Schmidt" <schmidty2244@gmail.com> wrote:

Hi all,

I am not sure how simple or difficult this problem is to fix but I am
trying to run NVDA from source, and the following problem has been
encountered when executing pythonw nvda.pyw from the command line. Is
anyone able to help? Also sorry if there is already a topic on this
issue: I have been off the list temporarily. I thought I would post
here to see if anyone else is having the issue before creating one on
GitHub:

From log:
---begin---
INFO - __main__ (18:25:49.265) - MainThread (23044):
Starting NVDA version source-master-44d8a31
CRITICAL - __main__ (18:25:49.940) - MainThread (23044):
core failure
Traceback (most recent call last):
File "nvda.pyw", line 214, in <module>
import core
File "core.py", line 31, in <module>
import nvwave
File "nvwave.py", line 15, in <module>
import wx
File "c:\GitInit\njsch\nvda\include\wxPython\wx\__init__.py", line
17, in <module>
from wx.core import *
File "c:\GitInit\njsch\nvda\include\wxPython\wx\core.py", line 12, in <module>
from ._core import *
ModuleNotFoundError: No module named 'wx._core'
--end---

At face-value, this sounds to me like a dependency-specific issue with
WXPython's code.
Is anyone else having this issue? Has the classic WXPython submodule
been updated recently?

Also, please note that I have set my cloned njsch/nvda/master branch
to remote-track nvaccess/nvda/master.

Should I pip uninstall wxpython? I don't think I installed it before
successfully creating my python virtual environment for the NVDA
source distribution. Although under the circumstances, I don't really
see how this would make a difference, since the log history indicates
that the nvwave module is looking in the Git submodule-included
dependency directory anyway.

Kind regards,

Nathaniel Schmidt
Re: WXPython Submodule Issue When Running From Source

Joseph Lee
 

Hi,
Because wxPython includes C extension code (pyd file), you'll need to run Python 3.7 specific and 32-bit wxPython 4.0.3.
Cheers,
Joseph

-----Original Message-----
From: nvda-devel@groups.io <nvda-devel@groups.io> On Behalf Of Nathaniel Schmidt
Sent: Thursday, June 4, 2020 1:59 AM
To: nvda-devel@groups.io
Subject: [nvda-devel] WXPython Submodule Issue When Running From Source

Hi all,

I am not sure how simple or difficult this problem is to fix but I am trying to run NVDA from source, and the following problem has been encountered when executing pythonw nvda.pyw from the command line. Is anyone able to help? Also sorry if there is already a topic on this
issue: I have been off the list temporarily. I thought I would post here to see if anyone else is having the issue before creating one on
GitHub:

From log:
---begin---
INFO - __main__ (18:25:49.265) - MainThread (23044):
Starting NVDA version source-master-44d8a31 CRITICAL - __main__ (18:25:49.940) - MainThread (23044):
core failure
Traceback (most recent call last):
File "nvda.pyw", line 214, in <module>
import core
File "core.py", line 31, in <module>
import nvwave
File "nvwave.py", line 15, in <module>
import wx
File "c:\GitInit\njsch\nvda\include\wxPython\wx\__init__.py", line 17, in <module>
from wx.core import *
File "c:\GitInit\njsch\nvda\include\wxPython\wx\core.py", line 12, in <module>
from ._core import *
ModuleNotFoundError: No module named 'wx._core'
--end---

At face-value, this sounds to me like a dependency-specific issue with WXPython's code.
Is anyone else having this issue? Has the classic WXPython submodule been updated recently?

Also, please note that I have set my cloned njsch/nvda/master branch to remote-track nvaccess/nvda/master.

Should I pip uninstall wxpython? I don't think I installed it before successfully creating my python virtual environment for the NVDA source distribution. Although under the circumstances, I don't really see how this would make a difference, since the log history indicates that the nvwave module is looking in the Git submodule-included dependency directory anyway.

Kind regards,

Nathaniel Schmidt
WXPython Submodule Issue When Running From Source

Nathaniel Schmidt
 

Hi all,

I am not sure how simple or difficult this problem is to fix but I am
trying to run NVDA from source, and the following problem has been
encountered when executing pythonw nvda.pyw from the command line. Is
anyone able to help? Also sorry if there is already a topic on this
issue: I have been off the list temporarily. I thought I would post
here to see if anyone else is having the issue before creating one on
GitHub:

From log:
---begin---
INFO - __main__ (18:25:49.265) - MainThread (23044):
Starting NVDA version source-master-44d8a31
CRITICAL - __main__ (18:25:49.940) - MainThread (23044):
core failure
Traceback (most recent call last):
File "nvda.pyw", line 214, in <module>
import core
File "core.py", line 31, in <module>
import nvwave
File "nvwave.py", line 15, in <module>
import wx
File "c:\GitInit\njsch\nvda\include\wxPython\wx\__init__.py", line
17, in <module>
from wx.core import *
File "c:\GitInit\njsch\nvda\include\wxPython\wx\core.py", line 12, in <module>
from ._core import *
ModuleNotFoundError: No module named 'wx._core'
--end---

At face-value, this sounds to me like a dependency-specific issue with
WXPython's code.
Is anyone else having this issue? Has the classic WXPython submodule
been updated recently?

Also, please note that I have set my cloned njsch/nvda/master branch
to remote-track nvaccess/nvda/master.

Should I pip uninstall wxpython? I don't think I installed it before
successfully creating my python virtual environment for the NVDA
source distribution. Although under the circumstances, I don't really
see how this would make a difference, since the log history indicates
that the nvwave module is looking in the Git submodule-included
dependency directory anyway.

Kind regards,

Nathaniel Schmidt
The read lloud voices in edge

Brian's Mail list account
 

I notice these are very good and seem to work well on windows 7. I don't suppose Microsoft offer any way to use these voices outside of Edge? It would be great if they did as a say all voice for example outside of the browser.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal E-mail to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
Newsgroup monitored: alt.comp.blind-users
Re: PowerShell not found error

Cyrille
 

Hello

I am posting this thread on the nvda-devel mailing list since it seems more adapted.

I tried to build on last parent commits master~4, master~2 and master~1. It worked well.

Then I have turned back to master and it builds well... Very strange.

Cyrille

Le 04/06/2020 à 15:51, Lukasz Golonka a écrit :
Hello Cyrille,

On Thu, 4 Jun 2020 11:08:05 +0200
"Cyrille via groups.io" <cyrille.bougot2=laposte.net@groups.io> wrote:

i Lukasz

I have installed scons as you did and it is now working on add-on builds.


Regarding NVDA build, I get the following error log:


scons: Reading SConscript files ...
TypeError: find_vc_pdir() missing 1 required positional argument: 'msvc_version'
:
  File "D:\Cyrille\DevP\nvdaNew2\SConstruct", line 236:
env32.SConscript('nvdaHelper/archBuild_sconscript',exports={'env':env32,'cli
entInstallDir':clientDir.Dir('x86'),'libInstallDir':sourceLibDir},variant_dir='b
uild/x86')
  File "c:\users\cyrille\appdata\local\programs\python\python37-32\lib\site-pack
ages\scons\SCons\Script\SConscript.py", line 597:
    return _SConscript(self.fs, *files, **subst_kw)
  File "c:\users\cyrille\appdata\local\programs\python\python37-32\lib\site-pack
ages\scons\SCons\Script\SConscript.py", line 286:
    exec(compile(scriptdata, scriptname, 'exec'), call_stack[-1].globals)
  File "D:\Cyrille\DevP\nvdaNew2\build\x86\archBuild_sconscript", line 190:
    win10localLib=env.SConscript('localWin10/sconscript',)
  File "c:\users\cyrille\appdata\local\programs\python\python37-32\lib\site-pack
ages\scons\SCons\Script\SConscript.py", line 597:
    return _SConscript(self.fs, *files, **subst_kw)
  File "c:\users\cyrille\appdata\local\programs\python\python37-32\lib\site-pack
ages\scons\SCons\Script\SConscript.py", line 286:
    exec(compile(scriptdata, scriptname, 'exec'), call_stack[-1].globals)
  File "D:\Cyrille\DevP\nvdaNew2\build\x86\localWin10\sconscript", line 46:
    find_vc_pdir(msvc),

Do you think it is linked or is it totally something else?
Huh, it turns out I also cannot build NVDA with latest master of Scons -
I had slightly older version installed previously. It is a change in
Scons which unfortunately breaks NVDA's Sconscripts.
Access violation with specific app and NVDA 2020.1

Noelia Ruiz
 

Hello:

When I try to start Fab Subtitler to generate live closed captions for
teletext (in real time) at job, I need to use Narrator since using
NVDA 2020.1 I get the following error dialogs with an OK button:

1. FAB Subtitler
Access violation at address 13010CD9 in module 'sttwin.exe'.
Read of address 13010CD9.

After pressing OK:
FAB Subtitler
Access violation at address 86522121 in module 'sttwin.exe'.
Read of address 86522121.

Questions:
1. Fab is an expensive proprietary application, so I think it's not
probable that a solution can be found for this specific and not very
used program. Should I create an issue to fix this in a general way or
to find some kind of workaround for other possible access violation?

2. Can I use some kind of configuration or command line parameters to
start Fab with NVDA without switching to Narrator for this single
task?
Fab website:
https://www.fab-online.com/subtitling/
Attached is a debug NVDA's log.
Thanks.

?
nvda.log

Latest alpha fails to download

Brian's Mail list account
 

When the new version is flagged and its initialised to download, there is a pause then it fails, there seems to be no indication of what the issue is, I guess the file is just not there??
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal E-mail to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
Newsgroup monitored: alt.comp.blind-users
Re: NVDA 2020 release schedule

Rich Caloggero
 

Hi all.

Sorry for the newbie-type question, but is there a place I can find more info on upcoming ideas / proposals for NVDA. In particular, I am interested in this "fast web" thing mentioned in the following message. Tried searching github issues, but couldn't find anyting relevant.
-- Rich


-- Rich

On 5/26/2020 2:36 PM, Bill Dengler wrote:
Hi Reef et al,
Any updates on the 2020.2 release? In particular:
I'd consider #11132 (Say-all repeats some lines) to be a blocking issue. Maybe consider reverting cancellable speech for 2020.2, then restore it again for 2020.3 if a fix can't be found to avoid blocking the release?
I'd like to see #11206 (Browse mode with auto focus focusable elements disabled: Focus focusable ancestor at caret when forcing focus mode with NVDA+space) make it into the release, assuming it's stable. This way there, new fast browse mode can be tested in 2020.2 (as the non-default) and #11190 (which enables fast browse mode by default) can be pushed to 2020.3, giving users lots of time to test and report issues as this new default represents a major change to NVDA functionality.
Windows Terminal has now made it to version 1.0 (stable release), and NVDA 2020.2 makes this application accessible (it currently isn't on previous NVDA builds). We should release this functionality to users as soon as practical.
Thanks,
Bill
-----Original Message-----
From: nvda-devel@groups.io <nvda-devel@groups.io> On Behalf Of Noelia Ruiz
Sent: 29 April 2020 10:28
To: nvda-devel@groups.io
Subject: Re: [nvda-devel] NVDA 2020 release schedule
Imo this may be documented if you find it useful, since the article about the release process was updated more than a year ago.
https://github.com/nvaccess/nvda/wiki/ReleaseProcess
Kind regards
2020-04-29 14:59 GMT+02:00, Reef Turner <reef@nvaccess.org>:
In the past we have aimed for 4 releases a year. The new approach will
likely result in a variable number of releases per year. The releases
will be smaller but more often. The release process takes a minimum of
6 weeks from first beta until release. That said, releases may be
delayed up by high priority issues. Realistically, I don't expect the
number of releases to change drastically.

-----Original Message-----
From: nvda-devel@groups.io <nvda-devel@groups.io> On Behalf Of Akash
Kakkar
Sent: Tuesday, 21 April 2020 8:37 PM
To: nvda-devel@groups.io
Subject: Re: [nvda-devel] NVDA 2020 release schedule

Hey Reef,
You said:
--We are aiming for a shorter release cycle, so we will likely start
the
2020.2 beta as soon as the 2020.1 release is made.-- So, will it be
for always or only for this time? and if it will be persistent, then
how many releases we can expect in a year?
Although, I'm interested in having more releases throughout the year.


On 4/21/20, Reef Turner <reef@nvaccess.org> wrote:
The 2020.1 RC is currently blocked waiting for:
https://github.com/nvaccess/nvda/pull/11040
We are aiming for a shorter release cycle, so we will likely start
the
2020.2 beta as soon as the 2020.1 release is made.
Re: Passing keys or entering text

Julien Cochuyt
 

Hi Bill,

To send arbitrary Unicode input, you can use brailleInput.handler.sendChars()


Best regards,

Julien Cochuyt
Accessolutions

Le jeu. 28 mai 2020 à 09:42, Bill Dengler <codeofdusk@...> a écrit :
Tangentially related, how would I type arbitrary text with NVDA? For example, I'd like to create a script to type – And — As they're not easy to type on my keyboard layout...

Bill
-----Original Message-----
From: nvda-devel@groups.io <nvda-devel@groups.io> On Behalf Of Tyler Spivey
Sent: 28 May 2020 03:39
To: nvda-devel@groups.io
Subject: Re: [nvda-devel] Passing keys or entering text

KeyboardInputGesture.fromName("shift+insert").send()

On 5/28/2020 12:25 AM, Sam Byrne wrote:
> Thanks Julien,
>
>
> Yes I knew of api.getClipData but how to actually get the info from
> NVDA memory to being entered on screen was where I got stuck. I was
> also unsure of how to write the actual keys that I want to pass when
> using passKey.
>
>
> Cheers,
>
> Sam
>
> On 28/05/2020 5:14 pm, Julien Cochuyt wrote:
>> Hi,
>>
>> Another commonly used standard paste gesture often supported in
>> terminals is shift+insert (Insert might need to be doubled to be
>> passed through depending on your NVDA configuration)
>>
>> To access the clipboard content programmatically with NVDA, you can
>> use api.getClipData()
>>
>> Best regards,
>>
>> Julien Cochuyt
>> Accessolutions
>>
>> Le jeu. 28 mai 2020 à 09:01, Sam Byrne <sam.byrne.90@...
>> <mailto:sam.byrne.90@...>> a écrit :
>>
>>     Hi All,
>>
>>
>>     I need some direction please on where to start for being able to
>>     perform
>>     standard keystrokes as part of my gestures or scripts I'm trying to
>>     write for a Putty app module. IE, I need to be able to paste the
>>     clipboard contents into a putty window, but as it is emulating a 3270
>>     mainframe window, Control+V keystroke isn't accepted. I've tried
>>     write
>>     () but that can only be used for adding content to files, not
>>     windows or
>>     screens it seems.
>>
>>
>>     How else could I write, type or import the clipboard text to the
>>     current
>>     cursor position? Please provide some basic example code if
>>     possible, as
>>     I really have no idea.
>>
>>
>>     I've done some googling, but anything with using keyboard and
>>     standard
>>     python appears overly complicated for what I'm trying to achieve.
>>
>>
>>     Thanks,
>>
>>
>>     Sam
>>
>>
>>
>>
>







Re: Passing keys or entering text

Rui Fontes
 

Do you know our FrequentText add-on?


You can find it at:

https://www.dropbox.com/s/631rs4t0xypc9u8/frequentText-1.0.nvda-addon?dl=1


Best regards,


Rui Fontes

NVDA portuguese team


Às 08:41 de 28/05/2020, Bill Dengler escreveu:

Tangentially related, how would I type arbitrary text with NVDA? For example, I'd like to create a script to type – And — As they're not easy to type on my keyboard layout...

Bill
-----Original Message-----
From: nvda-devel@groups.io <nvda-devel@groups.io> On Behalf Of Tyler Spivey
Sent: 28 May 2020 03:39
To: nvda-devel@groups.io
Subject: Re: [nvda-devel] Passing keys or entering text

KeyboardInputGesture.fromName("shift+insert").send()

On 5/28/2020 12:25 AM, Sam Byrne wrote:
Thanks Julien,


Yes I knew of api.getClipData but how to actually get the info from
NVDA memory to being entered on screen was where I got stuck. I was
also unsure of how to write the actual keys that I want to pass when
using passKey.


Cheers,

Sam

On 28/05/2020 5:14 pm, Julien Cochuyt wrote:
Hi,

Another commonly used standard paste gesture often supported in
terminals is shift+insert (Insert might need to be doubled to be
passed through depending on your NVDA configuration)

To access the clipboard content programmatically with NVDA, you can
use api.getClipData()

Best regards,

Julien Cochuyt
Accessolutions

Le jeu. 28 mai 2020 à 09:01, Sam Byrne <sam.byrne.90@gmail.com
<mailto:sam.byrne.90@gmail.com>> a écrit :

Hi All,


I need some direction please on where to start for being able to
perform
standard keystrokes as part of my gestures or scripts I'm trying to
write for a Putty app module. IE, I need to be able to paste the
clipboard contents into a putty window, but as it is emulating a 3270
mainframe window, Control+V keystroke isn't accepted. I've tried
write
() but that can only be used for adding content to files, not
windows or
screens it seems.


How else could I write, type or import the clipboard text to the
current
cursor position? Please provide some basic example code if
possible, as
I really have no idea.


I've done some googling, but anything with using keyboard and
standard
python appears overly complicated for what I'm trying to achieve.


Thanks,


Sam
Re: Passing keys or entering text

Rui Fontes
 

Hello!


Have you imported the module?

Something like:

from keyboardHandler import KeyboardInputGesture



Rui Fontes

NVDA portuguese team


Às 09:58 de 28/05/2020, Sam Byrne escreveu:

Thanks Tyler, I'm going to need a bit more background though as I currently get:

NameError: name 'KeyboardInputGesture' is not defined


I'm still very new to this as explained earlier.

On 28/05/2020 5:38 pm, Tyler Spivey wrote:
KeyboardInputGesture.fromName("shift+insert").send()

On 5/28/2020 12:25 AM, Sam Byrne wrote:
Thanks Julien,


Yes I knew of api.getClipData but how to actually get the info from NVDA
memory to being entered on screen was where I got stuck. I was also
unsure of how to write the actual keys that I want to pass when using
passKey.


Cheers,

Sam

On 28/05/2020 5:14 pm, Julien Cochuyt wrote:
Hi,

Another commonly used standard paste gesture often supported in
terminals is shift+insert
(Insert might need to be doubled to be passed through depending on
your NVDA configuration)

To access the clipboard content programmatically with NVDA, you can
use api.getClipData()

Best regards,

Julien Cochuyt
Accessolutions

Le jeu. 28 mai 2020 à 09:01, Sam Byrne <sam.byrne.90@gmail.com
<mailto:sam.byrne.90@gmail.com>> a écrit :

     Hi All,


     I need some direction please on where to start for being able to
     perform
     standard keystrokes as part of my gestures or scripts I'm trying to
     write for a Putty app module. IE, I need to be able to paste the
     clipboard contents into a putty window, but as it is emulating a 3270
     mainframe window, Control+V keystroke isn't accepted. I've tried
     write
     () but that can only be used for adding content to files, not
     windows or
     screens it seems.


     How else could I write, type or import the clipboard text to the
     current
     cursor position? Please provide some basic example code if
     possible, as
     I really have no idea.


     I've done some googling, but anything with using keyboard and
     standard
     python appears overly complicated for what I'm trying to achieve.


     Thanks,


     Sam
Re: Passing keys or entering text

Sam Byrne
 

Thanks Tyler, I'm going to need a bit more background though as I currently get:

NameError: name 'KeyboardInputGesture' is not defined


I'm still very new to this as explained earlier.

On 28/05/2020 5:38 pm, Tyler Spivey wrote:
KeyboardInputGesture.fromName("shift+insert").send()

On 5/28/2020 12:25 AM, Sam Byrne wrote:
Thanks Julien,


Yes I knew of api.getClipData but how to actually get the info from NVDA
memory to being entered on screen was where I got stuck. I was also
unsure of how to write the actual keys that I want to pass when using
passKey.


Cheers,

Sam

On 28/05/2020 5:14 pm, Julien Cochuyt wrote:
Hi,

Another commonly used standard paste gesture often supported in
terminals is shift+insert
(Insert might need to be doubled to be passed through depending on
your NVDA configuration)

To access the clipboard content programmatically with NVDA, you can
use api.getClipData()

Best regards,

Julien Cochuyt
Accessolutions

Le jeu. 28 mai 2020 à 09:01, Sam Byrne <sam.byrne.90@gmail.com
<mailto:sam.byrne.90@gmail.com>> a écrit :

Hi All,


I need some direction please on where to start for being able to
perform
standard keystrokes as part of my gestures or scripts I'm trying to
write for a Putty app module. IE, I need to be able to paste the
clipboard contents into a putty window, but as it is emulating a 3270
mainframe window, Control+V keystroke isn't accepted. I've tried
write
() but that can only be used for adding content to files, not
windows or
screens it seems.


How else could I write, type or import the clipboard text to the
current
cursor position? Please provide some basic example code if
possible, as
I really have no idea.


I've done some googling, but anything with using keyboard and
standard
python appears overly complicated for what I'm trying to achieve.


Thanks,


Sam
401 - 420 of 45477