No, the =? means "use this if no value is already set".
That is why it is there, to make sure there is a default value. I can only assume it is supposed to be updated on release, but whoever came up with it forgot to document it...
Ok. So we need to do something about this, I'll have a look.
The timezone thing is a typical example of a dependency problem, as handling timezones has changed in develop, but it requires a new tzdata setup which the release OE doesn't have.
So far i saw that if i disable the workaround by setting the boolean "m_remote_control" to false the qwerty keyboard work but the initial issue is back, the one with numerical input text.
That is the issue at hand, yes.
I wonder if there was any other issue reported at that time, or if there exists any extra infos which can help me to fix this?
No, this is exactly the problem. How to make a standard remote use the SMS-style input, while the qwerty remote use the alphanumeric input?
You still ignore the fact that the VU+ drivers reports the remote has an "A" key, when it hasn't. It has nothing to do with "blame", but with pointing out what the root cause of the problem is. Addressing the root cause is a fix, the rest is a workaround.
If you can fix it and provide a pull request, I'm all for it, I hate workarounds of any sort.
It doesn't help that afaik nobody in the team has one of these remotes, which makes development and testing very difficult.
But what I don't want it to bother users with a question that isn't relevant for 99.99999% of all users. If there is no way to reliably detect the remote, it should then be an override in the keyboard/remote setup menu (autodetect/numeric/alphanumeric), with autodetect being the default.
For my end-work decades ago I had to write a parser and compiler in Pascal, running on a Motorola 6800, which was fed a Z80 assembler file by the prof to validate it worked.
The main question is would it run the application has it would run on the original hardware?