I've got two keyboards attached to my computer. A normal full-size keyboard and a numeric keypad. I want to "remap" the keys on the numeric keypad, but leave the full keyboard alone. So, when the user presses "5" on the keypad it would get remapped to the "Media Play" key, but if the same "5" was pressed on the keypad of the full keyboard, I'd get a "5". In essence, I want to turn that seperate numeric keypad into a media control device.
Unfortunately I'm not sure how to make this work. There is a "Raw Input" feature of Windows which allows to distinguish between keyboards, but that only allows to read the keys - not remap them. There are keyboard hooks (low level and high level) which can remap the keystrokes, but they don't distinguish between keyboards.
Can I make a combination of the two methods work? For that I'd have to be sure that the raw input message is processed before the keyboard hook kicks in. Is that so?
Also, I've read that Windows for some reason does not allow to use Raw Input and Keyboard Hooks in the same process. I could work around that by making 2 different processes, but that becomes more fragile because of communication latency (keyboard hooks have very limited time in which they must complete their work).
Is this doable at all?
In the end it turned out I needed a custom keyboard driver to accomplish this. Fortunately I found one ready for taking - Interception. The app is ready and works perfectly. :)
I wanted to expand on the answer by Vilx as I've revisited this post several times, and I've only just managed to remap my second keyboard so the keys act uniquely. This is done using Interception and Intercept together, and it's the only working solution that I've managed to get working so far.
I'm also looking into using a raspberry pi to remap a second keyboard. Although early days, I think it it's achievable: Plug keyboard into Pi, plug Pi into PC and write script to turn input keycodes into customized output commands / keycodes. Will update when I have tested.
My working Intercept solution is completely based on a Github tutorial by TaranVH which details the process the best so far. I've copied his solution below with some amends of my own.
DOWNLOADING INTERCEPTION:
Navigate to this page: https://github.com/oblitum/Interception
Click on "download the latest release."
Click on Interception.zip to download it.
Unzip the file to a folder on your computer - I recommend using a folder location that won't ever move. I put my Interception folder into C:\Program Files\ and I'll be using this location in my examples.
INSTALLING INTERCEPTION:
Open a terminal as administrator and type cd "C:\Program Files\Interception\command line installer"
Hit Enter
Now, type in the following line of text exactly as shown:
install-interception.exe /install
Hit Enter
Once the program installs sucessfully, you still must restart your computer in order to complete the installation.
(If you ever want to UNinstall interception, watch this): https://www.youtube.com/watch?v=Hn18vv--sFY
INSTALLING INTERCEPT - (This is different from intercepTION!)
Download Kamaz's intercept.exe zip from one of these locations:
http://octopup.org/img/code/interception/intercept.zip
https://github.com/TaranVH/2nd-keyboard/blob/master/Intercept/intercept.exe
Unzip it if it's a .zip file and locate intercept.exe
To keep these files all in one place, I moved intercept.exe into C:\Program Files\Interception\
Reboot your system.
SETTING UP INTERCEPT
Plug in your second keyboard and ensure that it is working normally.
Go to where you saved intercept.exe and double click to open.
Intercept will open as a terminal, and from there you should have basic instructions to start remapping your keyboard.
Type a to add a key.
On the SECONDARY keyboard, press the Q key once. (this is the key that you are remapping)
You will be prompted with Enter combo for this trigger, end with ESC.
Carefully execute the following keystrokes (it doesn't matter which keyboard): Win + E (That is: hold down the windows key while pressing the letter E. I've found you need to do this quickly and accurately, if you hold down too long, extra keypresses are added)
When I did it, this was shown in the terminal: [Left Windows]↓ [E]↓ [E]↑ [Left Windows]↑
Press Esc and you will be prompted to label the script
I labelled mine Q->WinE so I knew what it was doing.
Press Enter to accept the label
Press S to save the filter, or C to cancel if you made a mistake.
Add more filters or close the terminal.
USING INTERCEPT
Now you've set up a custom key command, you need to apply the filter to use it.
Open up intercept.exe
Press Y to apply the filters, you will get confirmation that the filters have been activated.
Now, when you press Q on your second keyboard, Intercept will change the command to Win+E which will open File Explorer.
Caveat:
Please read TaranVH githubs page on this https://github.com/TaranVH/2nd-keyboard/tree/master/Intercept as it provides some further details and troubleshooting.
Please take care to follow the instructions very exactly as there's very little error handling in this process
In this article http://vlaurie.com/computers2/Articles/remap-keyboard.htm is explained how to remap keys of a keyboard. Hope that helps.
Related
as the title indicates I need help with the on-screen keyboard properties.
I have to configure a script, which I will deploy through SCCM in 30 terminals, and I need to make the keyboard go on screen as soon as I start the pc.
I have already solved this by adding this key to the registry
HKLM\Software\Microsoft\Windows\CurrentVersion\Run Clavier_Visuel = "C:\Windows\System32\osk.exe"
The problem is that I can't find a place to set the parameters of the keyboard: Remove the X to close it, display the PAD number, set the language to French and make it fit the screen.
This can be done manually, but that's not the goal.
Another point is that different users will be logged into the computers, so I can't set the condition to single user, but it should be general for all.
If you can help me it would be great, since I have not found anything useful so far, a greeting!
I have to use USB barcode scanner in application running on remote desktop. This scanner is being recognised by Windows as keyboard. When I try to scan barcode however, every 3th or 4th character is being changed to another one. When I type numbers by hand, they are being correctly written. The question is: is there any way to programmatically produce "lag" on keyboard globally? I would like to create for example a half second gap between keyboard input events and a result on the screen. Some time ago I worked with Windows hooks, and I was able to change keyboard input globally, but I remember I had problems with delaying it. Is there any other way?
you can make an autohotkey script to listen to a keyboard input, wait and then press the same keyboard button.
once you have intalled AHK(autohotkey) , you can do a AHK script whit the intructions that are in the ahk documentation.
Right-Click on your desktop.
Find "New" in the menu.
Click "AutoHotkey Script" inside the "New" menu.
Give the script a new name. Note: It must end with a .ahk extension. Ex. MyScript.ahk
Find the newly created file on your desktop and Right-Click it.
Click "Edit Script".
A window should have popped up, probably Notepad. If so, SUCCESS!
i::
sleep, 500 ;
Send, i
Return
(you can change the "i" for other keyboard keys).
a function would reduce the amount of code that you use, but i don't know how to do it in a AHK script
Any solutions out there for dealing with google script code-editor turning "forward delete" into a "kill line" binding? I'm used to the hidden cocoa/emacs derived navigation keys. Google decided to mess with control-d.
Clarifications:
This occurs when using the code editor. (edit-view?)
I would rather stop the google behavior than cope with it.
To understand more how fluid the control-key navigation can be
(and therefore how annoying when missing), try this...
set caps-lock to be the control key
open a cocoa browser like Safari and
start a script at https://script.google.com/
switch briefly to address bar
hit ctrl+a (goes to start of line)
hit ctrl+d (characters get deleted 1 at a time)
hit escape
switch to script area and try same thing (whole lines get munched!)
If you mean while you're working on your program in that editor, press the forward-delete key. On Apple's laptops and wireless keyboard, it's fn-delete (where by “delete”, I mean the backward delete, the key that is a.k.a. “backspace”).
You could try KeyRemap4Macbook.
Today I noticed that this problem no longer exists. So when editing a script file in the macro editor, emacs shortcuts WORK as expected. I cannot tell if they simply removed a bug or override, or actually patched it to work.
I'm building some code completion style text expansion functionality for text areas and inputs in a web application. I'm trying to work out a safe keyboard shortcut for invoking the completion proposals for browsers running on Mac OS X.
I've eliminated some potential candidates:
Command + Space - activates the Spotlight search field
Control + Space - activates the context menu in Firefox
Esc - cancels any background XMLHttpRequests in Firefox
That leaves Option + Space. I'm aware that may conflict with tools like Ubiquity, but that's something we don't expect our audience to be using.
Are there any conflicts I may have missed with Option+Space? Or do you have a better idea for a keyboard shortcut, and why?
A lot of Mac OS X apps use Option + Esc to do code completion or code hints.
Some kind of tab shortcut sounds like the best and most intuitive approach - users should be used to the 'type-tab-type-tab' workflow - maybe combine it with a modifier if it only wants to be optional.
As I recall, Visual Studio and TextMate both use tab to code-complete. In Visual Studio an intellisense menu pops up as you start typing and tab acts as the selection confirmation.
In TextMate, you start typing part of a command then hit tab to activate said command's associated "bundle". The associated bundle typically generates a code snippet and fills in any dynamic parts of the snippet as you type.
Come to think of it, most command lines work this way as well, auto-completing file names and paths when tab is pressed after typing a few characters.
EDIT: You say tabs are needed for field switching in a web interface, but you may want to try intercepting the keyboard event in the textfield, and check if they're started typing a macro. If they have, auto-complete and swallow the key by returning false; if not, simply let the command bubble through.
I discussed these ideas with our design team. One of them suggested using Control + Enter, which is what we decided to go with. This causes forms that only contain a single input field to automatically submit in Firefox, but we deal with that by preventing the default action caused by the event.
I re-assigned the Spotlight command to Command + Shift + Space. This is easily done via Prefrences.App
I am trying to learn Emacs and trying to find best keyboard layout for me. One thing is really annoying me. I have added following lines to .emacs
(global-set-key "\C-y" 'scroll-up)
(global-set-key "\M-y" 'scroll-down)
When I hold Control and press y a few times, it will page down on every press of y.
However, when I hold the Windows key (mapped as Meta) and press y a few times it will only page up on the first press of y and all subsequent presses of y I get the ‘y’ character inserted in the buffer.
Can the page up behave like page down? I want to hold Meta and keep pressing y to scroll multiple pages up.
I am using GNU Emacs 23.0.60.1 (i386-mingw-nt5.1.2600) of 2008-05-12 on LENNART-69DE564 (patched). It is Emacs with EmacsW32 patch. Is this problem with this Emacs? Problem with Meta key?
I tried original GNU Emacs (not patched) and it works OK with Alt. But my problem is not that I want to scroll without releasing any key. I release key y and press it multiple times but don't want to have to release Meta key. Same problem is described here:
http://groups.google.com/group/gnu.emacs.help/browse_thread/thread/f30f4b75a8b75b10
Problem is not in that I have changed key mapping. It looks like it is a bug in EmacsW32 version. Here is another description of the problem:
Unreleased Meta/Win modifier
Use C-v and M-v.
Don't change C-y, M-y default bindings.
Could this be a side affect of using the Windows key as Meta? I'm thinking this because in a non-Emacs situation if you press and hold the Windows key and another key for a short cut (Win+E for Explorer, Win+R for Run dialog, etc.) the desired action only triggers once, not multiple times if you keep holding it down.
I'd try reassigning Meta to Alt and see if the problem persists. If it doesn't, then I'm not sure what other option you have, since likely it's the OS only sending the Windows key press once to the app in focus.
You should use the patched EmacsW32 version, if you want the Windows key as Meta.
From the site about the patches:
"Changes that makes it possible to use the window keyboard keys as META in Emacs. Without this patch key sequences like E will always do what they by default does in windows, ie in this case open up Windows Explorer. (This patched is not used by default, you have to turn it on.)"