Yes, I agree absolutely with your thoughts.
toggle quoted messageShow quoted text
In this case the problem started manifesting itself with all snapshots after 4 February. My computer is a 64 bit Lenovo running on a fully updated version of Windows 10. It makes no difference whether or not add-ons are enabled. I use Defender as my anti-virus program and nothing else unusual is running in the background. When I fall back to the snapshot of 4 February, all is fine again.
If this snapshot is the one which might be released, then I am seriously concerned, because I will then not be able to update on that computer.
On 2018/02/15 11:18 AM, Brian's Mail list account via Groups.Io wrote:
Does the master branch latest snap work, as this is the one which will become the next release, not the next snaps.
I really think we have seen a number of unexplained strange behaviours in the last few weeks. Whether this is just coincidence or not I have no idea. Some people are using windows 7 others 8 and 10, so maybe they are not connected, but maybe there os some extra ingredient here we are all missing. Some update to a piece of code running in the background, such as anti virus or service that is messing about with more recent versions of nvda.#
I think these really do need a member of the dev team to try to find out the exact environments and extents of problems in a system smitten by one of these weird effects. Thus far the only one I've seen is the inability to restart sometimes after an update, traced to the core of windows being processing removal of locked folders and files and these usually go away if a delay in restarting is don or a reboot of nvda. These issues afflict some other software so have to be a windows thing. In the case of the people with problems these seem to be.
A dll missing which is part of a Microsoft runtime package
Intermittent refusal to speak menus and parts of dialogues
Lack of virtual buffers oon browsers.
Some of these are very reminiscent of non registered dll files of inability to locate said files by some part of the system. As these files mostly only show up as a problem when using a screenreader, it may well be there are problems in the wider community, which is going undetected, perhaps due to a combination of windows updates or updates to other software in the background. Of course it could b that the specifics of calling code are not entirely correct in certain circumstances making the nvda program have issues.
I really do think that somebody needs to find a way to raise all these things as issues sooner rather than later as you say, in case there is a wider issue. If developers can simulate the environment in use that causes these problems, then fixing them is not so hard.
Just my penneth on it.
Sent via blueyonder.
Please address personal email to:-
email@example.com, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Christo de Klerk" <firstname.lastname@example.org>
Sent: Wednesday, February 14, 2018 5:25 PM
Subject: [nvda] nvdaHelperRemote persists with latest snapshot
Hello especially to the NVDA developers
A few days ago (a few snapshots ago) I wrote that when I updated NVDA to the new snapshot, I received an error message with the following label:
The text of the error message is:
Error registering focus winEvent hook
Then there is an Ok button. When I Ok the dialog, NVDA works, except that there is no browse mode buffer which makes web browsing impossible.
This problem persists with the very latest snapshot which I have just installed, so I had to fall back again to the last working snapshot which is from about 4 February. This happens only on my 64 bit machine.
The result of this is that I have not been able to use the last number of snapshots and I have a real concern that this problem may manifest itself once 2018.1 gets released which would mean that I will not be able to use that version on my 64 bit machine.
Does anyone have any idea what is happening here?