I used AutoHotKey for months with no issues, but now some (not all) of my shortcuts no longer work.
Any shortcut used to open a browser tab and go right to a site still work. For example, this…
#a::Run www.autohotkey.com
…will still bring a new tab into focus and load the AutoHotKey site when you press WIN+A. Similarly, everything I have to launch programs still works. But I was mostly using AHK to access accented letters or special characters on a laptop without a numpad. This…
!c::Send,ç
…used to produce a ç but now when you press ALT+c it generates this: ç
It doesn't seem to be tied exclusively to the ALT key. Even if I remap to avoid the ALT key I have the same issue. Below are (I believe) the affected bits in my *.ahk file.
Thanks to anyone who has ideas as to why this worked great and then one day stopped working! I would love to be able to use AutoHotKey again.
!a::Send,à
!b::Send,ß
!e::Send,è
#!e::Send,é
!i::Send,ï
!c::Send,ç
#!c::Send,Ç
!d::Send,·
!g::Send,?
!h::Send,?
!m::Send,—
!n::Send,ñ
!p::Send,¶
+!p::Send,þ
!s::Send,§
!x::Send,×
!'::Send,«
#!'::Send,»
!2::Send,²
!3::Send,³
!4::Send,¢
!7::Send,£
!=::Send,±
!/::Send,÷
+!/::Send,¿
!.::Send,…
What version of AutoHotkey are you using? Always use AutoHotkey and its documenatation from http://ahkscript.org/ (current uptodate version, new official website)! AutoHotkey and its documentation from autohotkey.com is outdated and you may have some problems using them!
Try to launch script with admin rights.
Turn of antivirus software and try if script works.
Check other software. Maybe some of them is using same shortcut?
Loading of UTF-8 macros containing national charactes doesn't work correctly if you have US keyboard active during the process. Switch to national keyboard, then reload the macro file.
For example, I'm using these locale/keyboards combinations and AHK script loads fine only with two:
When keyboard is switched to ENG-US, instead of success I get the following error (file is always the same, encoded in UTF-8):
---------------------------
AutoHotkey.ahk
---------------------------
Error at line 332.
Line Text: #ú::Send {Browser_Back}
Error: Invalid hotkey.
The script was not reloaded; the old version will remain in effect.
---------------------------
OK
---------------------------
With some other national characters, you don't get above error, but without a warning from AHK you get strange effects in interpretation of some UTF-8 characters.
AHK version: 1.1.16.05 (ahkscript.org)
Related
I am trying to figure out why VSCode sometimes decides to randomly insert "non-breaking spaces" U+00A0 instead of regular spaces, and how to stop this from happening?
This is a different type of space character which is causing issues, as seen in my previous question GitHub markdown not rendering and git detecting changes when no changes
This is happening to me on both Mac OS and Windows, and others have experienced this issue too. Although this question is about VSCode, this also happens on the online GitHub markdown editor.
How do I solve this problem?
Steps to reproduce
Edit a markdown .md file in VSCode for a while.
Put in lots of headings such as # my heading or ## my sub-heading.
If you have the markdown extension you can CTRL/CMD + SHIFT + P and Open preview to the side.
You should eventually notice that one of the headings doesn't render, and if you open the file in a hex editor such as HxD you'll notice that there is a non-breaking space instead of a regular space after the # character.
Possible causes
The backtick ` character on Windows can be combined with a letter on your keyboard for example a which results in à. Maybe the use of backticks is causing the space character to turn into a non-breaking space? However this wouldn't explain the problem on Mac OS, because backticks can't be combined with other characters.
I have this a lot too on Mac OS, but it turns out I am still holding the option key slightly (or perhaps shift on your keyboard) when typing the space after just having used the modifier key to type #
I've installed https://marketplace.visualstudio.com/items?itemName=viktorzetterstrom.non-breaking-space-highlighter&ssr=false#overview to spot my mistakes early.
I just spent quite a lot of time trying to figure out why a "mv foo bar" command I copy and pasted from VScode was returning command not found. Turned out the space in between mv and foo was a non-breaking space which I only determined by pasting it into a unicode text converter. I would really like to know why this is happening and if it's a setting I can disable. I am using Windows 10 so I don't think it's just a Mac issue.
I have faced this trouble several times when copy/pasting code from Microsoft Skype or Microsoft OneNote. Both programs convert regular spaces into special Unicode spaces when pasting into it.
By the way, since version 1.63, Visual Studio Code automatically highlights these special spaces.
I ran into this too, but I'm sure I unwittingly inserted the non-breaking spaces using the spacebar alone. I found the hexdump plugin useful for diagnosing the problem. I still don't know how to avoid it.
I am trying to use org-mode and whenever I press Crtl+return emacs does C-J instead. I know this because when I use c-h k and then press crtl+return, emacs shows:
"It is bound to C-j."
Is there anyway I can make it register ctrl+return so I can insert headings?
I am using emacs23 through the terminal on ubuntu 12.04 x64. When I do this on my windows machine it does insert a new heading.
Possibly. As #phils says, C-Enter is usually not a valid sequence for the vast majority of terminal emulators. This is because when you press a key, or key combination, the terminal sends a sequence of characters to the relevant application. Pressing a sends a, whereas Enter sends CR - carriage return (otherwise known as ^M, \r or \015).
What sequences are actually implemented in modern terminals is mostly just a hang-over from the earlier, physical, terminals that they emulate (notably the VT-100). These machines didn't allow the use of Ctrl in conjunction with every key, such as the function keys, for example, so C-F1 was not possible and didn't have an escape sequence defined. Similarly, there is no control sequence for C-Enter, so when you press it, the terminal ignores the Ctrl and just sees the Enter.
You may, however, be able to define your own escape sequences in your terminal, which you can then bind to the correct key combinations in emacs. See here, for example. I was able to use this method to be able to use C-Enter. Unfortunately, this is still a bit of a pain, as it requires a lot of configuration and you may still be unable to use some sequences (C-M-% doesn't work in konsole for me). As with #phils, I've taken to using the GUI, but I'd love to have better integration with my terminal.
That's not a valid sequence for most terminals, I'm afraid, so they simply can't send it to Emacs.
If you run GUI Emacs instead, the key bindings will work.
Thomas Dickey's xterm is the only terminal software I know of which can (I think) support all the normal Emacs bindings. If you're okay with compiling that, then you could try it with the following .Xdefaults-(hostname) (or .Xresources if you re-merge after editing) to get basic support working:
XTerm*metaSendsEscape: true
XTerm*modifier: meta
XTerm*modifyOtherKeys: 1
You actually need XTerm*modifyOtherKeys: 2 to fully extend the sequences (otherwise you still can't type the likes of C-M-%), but by default that setting will break most key sequences, and right now I couldn't tell you how to configure things correctly for that setting (the only example I've seen -- xterm-extras.el as mentioned on the wiki and available in the easymacs download -- didn't work for me).
I mostly use GUI Emacs, but I'd love to get this working properly, so if anyone has had success with xterm-extras.el or similar, please do speak up.
If you're willing to install an Emacs package and configure your terminal, I wrote an Emacs package which can teach Emacs and terminals how to properly recognize all PC keyboard keys and modifier key combinations:
https://github.com/CyberShadow/term-keys
Its default configuration enables encoding the CtrlReturn combination, which allows Emacs to distinguish it from CtrlJ.
I have set up a small AutoHotkey script that uses hotstrings to allow me to type a few emoji and other special characters:
#SingleInstance force
#Hotstring * ?
::–::–
::†::†
::π::𝜋
::&shrug;::¯\_(ツ)_/¯
::&tm;::™
::&c;::©
::&r;::®
::&tableflip;::(╯°□°）╯︵ ┻━┻
::&music;::♫
::&piano;::🎹
::&cmark;::✓
::&xmark;::✗
::&hourglass;::⌛
The script is also available on PasteBin.
It all works in my text editor:
However, some replacements don't work everywhere, for example in Telegram Desktop:
Both emoji were originally copied from Telegram Desktop, so they are definitely supported in it.
Why isn't this working as expected?
I'm running AutoHotkey Unicode 64-Bit in version 1.1.23.3 on Windows 10 (version 1511), Sublime Text 3 (3103) and Telegram Desktop 0.9.28.
I answered a similar question in the past: Autohotkey Replace 2 specific character with one
I don't really know what the problem was, but we managed to get around the problem by making use of the clipboard. Maybe the same method can be used for your scenario.
I have written a python script for my co-workers, and then created an autohotkey script to run it every time someone presses Ctrl+LShift+Y. Looks something like this:
^+y::Run helper.py
The python script is fine, but the ahk script doesn't work on all the computers. Sometimes it works fine, and sometimes you get this error:
^+y does not exist in current keyboard layout
Now, searching the web this seems to be a problem with multi-language keyboards (we're using both Hebrew and English), because different languages means a different layouts (I guess?). I also found someone explaining that to solve this you need to use scan codes instead of the usual ^ and + and so on (I'd link to it but I cannot seem to find it now).
This all vaguely makes sense to me on a theoretical level, but when I want to realize it with actual code, I don't really know what to do. To me it seems as if this topic is hardly discussed (with the few exceptions being lacking in examples or hard to understand), so I'd love an answer that would include the following:
some simple way of determining the scan code for a key. This should preferably be a pythonic solution (and just out of curiosity, I'd love to know how to do this with linux as well). This is probably the easier part (but I think is an inherent part of a complete answer).
This is the important part: examples of how you implement that scan code in an autohotkey script, including edge-cases (if there are any).
Question 1
As you want to use the key with autohotkey, it makes sense to use autohotkey detect the key in the first place. Obviously this method works only on windows where autohotkey is running.
Write a Autohotkey script with this line and run it.
#InstallKeybdHook
Press the key you want to examine.
Open the script menu by right clicking the icon of the script in the right lower corner of your screen.
Select OPEN, then from the Menu "View / Key history and script info"
There is a line for each keypress.
First column is the VK (Virtual key) code, next is the scancode.
For example for CAPSLOCK the VK is 14 and the Scancode 03a
Question 2:
#InstallKeybdHook
VK14::
msgbox, you pressed capslock!
return
OR
#InstallKeybdHook
SC03a::
msgbox, you pressed capslock!
return
both work.
Note that you can combine two keys into a hotkey by combining them with & (but not 3)
#InstallKeybdHook
RShift & SC03a::
msgbox, you pressed Rshift capslock!
return
You can modify a Scancode with + and ^
#InstallKeybdHook
^+SC02C::
msgbox, you pressed Ctrl Shift and Y(maybe)!
return
Further info about this is on the page "List of Keys, Mouse Buttons, and Joystick Controls" of the autohotkey help file that comes with the default installation.
Please try to forgive this slightly off-programming-topic question. At least I'm getting this problem when porting my C++ code to Linux. The problem I'm having is very basic: some key combinations are not recognized by emacs. One result of that is that I can't 'undo' edits in Emacs.
I'm running Putty.exe from Windows to my Ubuntu machine. The Putty character set (Window -> Translation -> received data char set) is UTF-8, and all my Swedish characters show up neatly on the console. When in Emacs however, I can't get C-_ to work. Nothing happens. The key combination on my Swedish keyboard is
Ctrl Shift -
Any pointers are helpful!
Try with just Ctrl and "-". I do not use emacs, but I use that key for screen(1) since it is not used for anything else (and the default of Ctrl-a is not so useful). On windows (with my Norwegian keyboard) the keypress to trigger it is just Ctrl-. In konsole windows in KDE Ctrl- will decrease font size, so there I have to fall back to Ctrl Shift - (it used to be just Ctrl- there as well some years ago).
When I'm editing with Swedish keyboard settings I have gotten used to undo with C-x u for reasons I can't remember anymore.