<Ctrl+Shift+0> keyboard shortcut (still) doesn't work

[Niall Saunders]

I had forgotten about this one, I mentioned it a good while back.

With a colour image on screen there are a few keyboard shortcuts that allow you to view the different 'components' of that image.

For example <Ctrl+Shift+R> will let you view just the 'Red' channel. (and ...+G or ...+B shows you the Gn and Bu channels respectively)

However, although the on-screen menu suggests that you will get back to the 'normal' (i.e. RGB) view using <Ctrl+Shift+0> - I have never managed to get this to work.

Am I doing something stupid, or is there a 'bug'? (Using the mouse to click the appropriate area of the drop-down menu works, so the actual code - obviously - works)

Cheers,

[NKV]

back to the 'normal' (i.e. RGB) view using <Ctrl+Shift+0> - I have never managed to get this to work.

Win32, <Ctrl+Shift+0> working fine.

[Niall Saunders]

Win32, <Ctrl+Shift+0> working fine

Sorry, I should have added that it was not working on Vista64.

What about your 64-bit install, Sander? Does this work for you?

Nikolay - can you confirm whether the script depends on 'which' combination of 'left or right' CTRL and Shift keys are used? Can you also confirm whether the script depends on the Numeric Keypad '0'? (as far as my brief test confirmed earlier this morning, the '0' is a 'number zero', NOT a 'letter O', as the 'letter O' combination triggers an 'open file' type of dialogue).

Cheers,

[Nocturnal]

Hi Niall,

I never use these shortcuts but ctrl-shift-0 (zero) does not revert back to RGB display on my system (vista x64).

[Niall Saunders]

Thanks for at least checking Sander - so it looks like it must be a Vista-64 compilation issue (if it works in Win32)

Cheers,

[NKV]

Niall, I not have Numeric Keypad, but left and right (CTRL and Shift) work similar.
VAIO notebook Win7-32 -  doesn't work.
Desktop PS/2 Keyboard WinXP32 - OK.
Desktop wireless Keyboard Win7-64 - OK.

I think the bug depend on Keyboard. You can test your Keyboard via

Code: [Select]

   this.onKeyPress = function(value)
   {
      console.writeln(value);
   }Or just remove "//" in line 620 of current version Animation script.

[Niall Saunders]

Hi Nikolay,

OK - with your excellent 'DeBugger' script (i.e. Animation on Steriods ), I get the following O/P from the console when I enable line 620

Code: [Select]

Processing script file: C:/PCL64/src/scripts/Animation.js
Prepare images

Reading image:
C:/Users/Niall/Documents/_ex DSI/_Work In Progress/20100515and17 M57 LRGB/Lu Lights 7-0C 60s x60/calibrated/aligned/_a_cM27 IIP Lu Light 7-0C 60s 0001.fit
Reading FITS: 32-bit floating point, 1 channel(s), 748x577 pixels: 100%

Reading image:
C:/Users/Niall/Documents/_ex DSI/_Work In Progress/20100515and17 M57 LRGB/Lu Lights 7-0C 60s x60/calibrated/aligned/_a_cM27 IIP Lu Light 7-0C 60s 0002.fit
Reading FITS: 32-bit floating point, 1 channel(s), 748x577 pixels: 100%

Reading image:
C:/Users/Niall/Documents/_ex DSI/_Work In Progress/20100515and17 M57 LRGB/Lu Lights 7-0C 60s x60/calibrated/aligned/_a_cM27 IIP Lu Light 7-0C 60s 0003.fit
Reading FITS: 32-bit floating point, 1 channel(s), 748x577 pixels: 100%

HistogramTransformation: Processing view: Image68
Processing gray channel: Histogram transformation: 100%
0.0450 s
49
50
51
52
53
54
55
56
57
48
48
49
50
51
52
53
54
55
56
57
The list of keycodes was generated by pressing each key along the top row of my main keyboard, from '1' to '0'. I then pressed each key on the numeric keypad, from '0' to '9'. It is obvious that every key provided an appropriate response.

Then I repeated the same keypresses, in the same order, but with the Left-Shift key held down all the time. This is the result (just the keycodes this time)

Code: [Select]

-268435455
33
34
163
36
37
94
38
42
40
41
268435472
-268435455
268435968
-268435455
268435460
-268435455
268436480
-268435455
268435457
-268435455
12
-268435455
268435459
-268435455
268435712
-268435455
268435458
-268435455
268436224
-268435455

This time the first keycode corresponds to the <Left-Shift> key being pressed, and held.
The next ten keycodes are generated as I run along the top-row number keys, '1' to '0'

After keycode '41' I then get '268435472' when I press '0' on the numeric keypad, and '-268435455' when I release '0', where the code presumably detects that the <Left-Shift> key is still being pressed (just as happened right at the start of the sequence). From then on I get a 'new' keycode for each Numeric Keypad keypress, with the <Left-Shift> keycode repeated between each subsequent keypress.

It is clearly obvious that the keycodes are totally different on the Numeric keypad than they are on the main keypad - as would be expected. It is also clear that EVERY keypress was detected, with no problem

Then I repeated exactly the same sequence of keypresses, but this time with the <Left-Ctrl> key pressed and held, as follows :-

Code: [Select]

-268435454
49
50
51
52
53
54
55
56
57
48
48
49
50
51
52
53
54
55
56
57
Now I am starting to get confused The <Left-Ctrl> modifier key does not seem to 'modify' the keycodes at all - even though 'it' itself is actually detected right at the start of the sequence. They seem to be identical to those returned when no modifier key was pressed. A quick double-check showed that the <Right-Ctrl> modifier key had exactly the same effect as the <Left-Ctrl> modifier key - i.e. NO EFFECT

Now to the 'key' of the matter - the same test again, but this time with the <Left-Shift> key pressed first (and held down thereafter), followed by the <Left-Ctrl> key (same agin, kept pressed thereafter), then the two keyboard sequences, as follows :-

Code: [Select]

-268435455
-268435454
49
50
51
52
53
94
55
56
57
268435472
-268435455
268435968
-268435455
268435460
-268435455
268436480
-268435455
268435457
-268435455
12
-268435455
268435459
-268435455
268435712
-268435455
268435458
-268435455
268436224
-268435455
Now, this is where my little friends have to come to my assistance,

As you can see, there is NO ENTRY after the '55', '56', '57' keycode sequence. Somebody has STOLEN the '48' keycode

And, presumably this is why PixInsight is not responding to <Shift-Ctrl-0>

That particular keypress combination did NOT get captured by your script, and presumably does not get captured by PI 'core'. Even more confusing is the fact that <Shift-Ctrl-0_NUM> DOES get detected (generating keycode '268435455' according to the output of your script).

And,Juan, 'Yes' I am using a 'bog-standard' Logitech Wireless keyboard

There is just something 'horrible' going on. I am going to call a Priest

Cheers,

[NKV]

Somebody has STOLEN the '48' keycode

Niall, it's keyboard manufacture or keyboard driver stole your <Shift-Ctrl-0> 
I have same problem with my notebook. Also you can test your keyboard with any keyboard tester.
I used this one http://www.passmark.com/products/keytest.htm and found: combination <0+Shift+Ctrl> and <Shift-Ctrl-0> produce different result. 

So in my case the problem inside stupid keyboard driver.

Juan, maybe you can add combination <0+Shift+Ctrl> or <Shift+0+Ctrl> to next version of PI ?

[Niall Saunders]

Thanks Nikolay,

I will download and try that software this evening.

What bugs me is that this is a Logitech keuboard - allegedly the 'defacto' manufacturers of keyboards. Perhaps Lohitech should spend less time trying to fill your PC with 'back-door' software and more time just getting their drivers to work.

Maybe it's time for a new one - this keybaord has been bugging Juan for years now

Cheers,

[Nocturnal]

Mine is a pretty vanilla Microsoft 4000.

http://www.microsoft.com/hardware/mouseandkeyboard/productdetails.aspx?pid=043&active_tab=systemrequirements

[Juan Conejero]

There is just something 'horrible' going on. I am going to call a Priest

Hmm, haven't seen any of those here

Seriously now. I can confirm this problem on Vista x64. My keyboard is a wired Logitech. Ctrl+Shift+0 works with the same keyboard on XP 32. I really have no idea about why does this happen. I'll try with a different keyboard but I suspect this is a Vista x64 specific issue.

Thanks for reporting and sorry for not having a workaround. And please, don't change your keyboard as it has proven invaluable as a testing tool for PI