Changes to Unicode support in LabVIEW 8.6

[tst]

There seems to have been a change in the way LabVIEW handles alternative text input between 8.5.1 and 8.6.

In 8.6, if I try to use type something in Hebrew (alt+shift in Windows changes the input language), the words come out in the wrong order (the first word appears on the left instead of on the right). The actual order of the letters in the word appears correctly (right to left), although sometimes the first letter in the text gets stuck on the left size.

If I try to use Hebrew in a label, I get a message telling me that I can't type a Unicode string into labels and enums.

In all of these cases, copying a string from Notepad and pasting it into LabVIEW displays it correctly (including in labels).

I have tried doing this with UseUnicode INI key set both to T and to F and it happens in both cases.

This SUCKS. Big time. Really.

I know that LabVIEW doesn't officially have support for right to left languages, but at least in previous versions it would sort of behave. I'm pretty sure this isn't what people who were expecting more Unicode support were thinking about.

Does anyone have any idea for a workaround? The only two I can currently think of is copying and pasting or writing a small VI which will reverse the text on demand, both of which are bad solutions.

If there's a patch planned, fixing this will definitely get my vote as something that deserves going in there.

P.S. I'm not sure why the Hebrew input is treated as Unicode. If I'm not mistaken, changing the input language in Windows should still result in ASCII characters and in previous versions it does (the Hebrew chars in the code page LabVIEW uses in 7.0 start at ASCII E0 and in 8.6 it starts at D005 with the next letter being D105).

P.P.S Did I mention this sucks?

[AristosQueue (NI)]

I'm going to suggest something really strange as a workaround, but, trust me, I have historical precedent for suggesting this...

Try editing the keyboard shortcuts for LabVIEW and change Quick Drop to be something other than ctrl+spacebar. 

Why might this help the problem? When I developed the  alignment grid, I thought that ctrl+# was a very intuitive shortcut key for toggling the grid on and off (since the hash symbol looks like a grid). What we found was that when the OS language was set to French, users could no longer type many of the extended grammer characters. The reason was that a critical scan code for the extended character sets overlapped with the scancode for ctrl+shift+3. In LV 8.6, we added ctrl+space, which is the first non-alphabetic shortcut key we've added since the ctrl+#. It might be that whatever keyboard scancode is used for Hebrew happens to overlap, and the code is being translated into a shortcut key for LV instead of a typed character.

I have no evidence that this is what is happening, but it is the only analagous situation that I know of, so I suggest it as something to try. 

[tst]

Doesn't help, but thanks.

It looks like typing was changed in 8.6 to be Unicode under certain conditions and I assume this change was deliberate, as LabVIEW specifically pops up a message when I try to edit a label. My hope is that there is an INI key or a workaround which whoever worked on changing this will know.

Just to simulate what this look like, in English it would look something like this - sentence a is This. Basically, it's sort of Yoda-speak. At other times, it can look like this - sentence a is hisT. At other times, it also shows giberrish and moves the cursor about 4 characters for each character typed. There doesn't seem to be any consistency to what's causing this.

[Babel]

Hi TST,

My group is going to reproduce the problem and discuss with LV R&D to see what they can do about it. I don't know why Hebrew would be treated as Unicode.

-Michel Farhi

Localization Manager

[tst]

I'm not sure if this is specifically Hebrew. I wouldn't be shocked if it happened with Arabic as well, but it might even happen with other languages. I'm still secretly hoping for some INI key which will change it.

P.S. Do you have a CAR, or should I open an official request?

[Babel]

Hi,

Our group will create a CAR when we have reproduced and I'll let you know the ID, so you can follow the issue with our AEs.

Thank you

-Michel Farhi

[tst]

Thanks. This could pose a real problem (it's really annoying), and the AEs might start hearing from other people soon too, so it would nice to know that there is (or will be) a solution.

[Babel]

I have reproduced the problem and have CARed it with a high severity: CAR # 124752.

I'll keep you informed when a developer has had time to debug the issue.

Thank you

-Michel Farhi

[tst]

Thanks.

[tst]

P.S. As a side point, do you have any AEs with Hebrew support installed or did you reproduce it using Arabic or something else?