I frequently end up doing ~5 different manual actions when switching between using my laptop in dark environments and light environments, one of which is clicking Turn all styles off in the Stylish extension toolbar button to disable my dark user styles. I'd like to automate this somehow, probably a script bound to a hotkey. Is there a way to have the same effect as clicking Turn all styles off through the command line or some other way to accomplish this goal? I'm using Stylish in Chrome on Ubuntu.
My findings are that there isn't any way to change chrome extension settings from the command line and that the best way to do this is to set a key binding per #wOxxOm's comment and to use xdotool to send the keystrokes to chrome. There are a lot of gotchas in getting this to work. This is the final command I ended up using:
xdotool search --class 'google-chrome' windowactivate --sync \
key --window 0 --clearmodifiers 'ctrl+shift+5'
Chrome doesn't accept the keystrokes if it's not active, so use windowactivate --sync first. It seems that Chrome ignores XTEST events so use --window 0 to force xdotool to use XSendEvent instead. Finally, I had to use --clearmodifiers because I have Caps_Lock and Control_L swapped.
Related
I am trying to modify Chromium browser's behavior in Chromium OS.
When I execute Chromium browser, I can handover arguments with '--enable-features=' flag.
And I found a argument (#enable-virtual-keyboard) from Available list in chrome://flags.
Then I executed Chromium browser like below.
example instruction # /opt/google/chrome/chrome --incognito --enable-features=enable-virtual-keyboard
And when I check the flags in chrome://version, I could see the arguments.
Command Line: /opt/google/chrome/chrome --blahblah --enable-features=enable-virtual-keyboard
But actually the virtual keyboard is not working.
When I enable the argument (virtual keyboard) from chrome://flags manually. It works well and I could see the Virtual Keyboard.
What is wrong? I would appreciate it if you guys advise me.
--enable-features actually takes the name of a feature, not the name of a flag. The Chromium documentation explains the difference, but to simplify somewhat, a flag is sort of a user-visible interface to a feature or a command-line switch.
Anyway, to do what you want, you need to find out what feature or command-line switch the enable-virtual-keyboard flag toggles. Searching for enable-virtual-keyboard in Chromium Code Search brings us to these lines in about_flags.cc:
{"enable-virtual-keyboard", flag_descriptions::kVirtualKeyboardName,
flag_descriptions::kVirtualKeyboardDescription, kOsCrOS,
SINGLE_VALUE_TYPE(keyboard::switches::kEnableVirtualKeyboard)},
That last line tells us what we're looking for: that this flag is actually backed by a command-line switch, not a feature. Clicking on kEnableVirtualKeyboard shows us its definition:
const char kEnableVirtualKeyboard[] = "enable-virtual-keyboard";
So, adding --enable-virtual-keyboard to your command line should have the same effect as flipping the flag. (I should note, though, that there are a number of things that can break the virtual keyboard, so you may have to do other things to get it working.)
For the sake of completeness, if you'd found that it was actually backed by a feature, like the gesture-properties-dbus-service flag, for example:
{"gesture-properties-dbus-service",
flag_descriptions::kEnableGesturePropertiesDBusServiceName,
flag_descriptions::kEnableGesturePropertiesDBusServiceDescription, kOsCrOS,
FEATURE_VALUE_TYPE(chromeos::features::kGesturePropertiesDBusService)},
Then clicking kGesturePropertiesDBusService takes us to its definition which specifies the feature name:
const base::Feature kGesturePropertiesDBusService{
"GesturePropertiesDBusService", base::FEATURE_DISABLED_BY_DEFAULT};
...so in that case you'd pass --enable-features=GesturePropertiesDBusService to enable the flag from your command line.
I did the __disable-web-security in Terminal on Mac. Do I need to enable it again? Or does it enable by itself after restart?
If I need to enable it again, how do I do that? I searched everywhere, but didn't find.
Go to chrome://version and look at the Command Line. If the command line argument is still there, you should restart chrome and then launch it normally (without the flag).
A bit more context: "Flag" refers to two slightly different concepts in Chrome:
Any command line argument that starts with a -- (as opposed to an argument that names a URL or a file, for example).
The list of toggles available at chrome://flags.
Any arguments added manually to the command line will only persist until you quit Chrome. On the other hand, the toggles set at chrome://flags will persist until they are reset. (To further blur this, invoking chrome://restart will preserve the command line flags.)
This particular flag makes its presence fairly obvious, with a banner:
You are using an unsupported command-line flag: --disable-web-security. Stability and security will suffer.
So it should hopefully be clear whether the flag is present (and also why this is the sort of flag that can't be persisted at chrome://flags).
The flags set in #2 are delimited in Chrome's command line by the --flags-switches-begin…--flags-switches-end section. Arguments outside of that section are coming from somewhere other than chrome://flags (namely, from the command line, except in the case of Chrome OS and Android, which have their own startup procedures).
Developing a web app locally and I just can't get Chrome to actually go to this address, because auto search always kicks in (http://0.0.0.0:5000/api works though, for example).
Is there a way to overwrite this behaviour or force Chrome to try a http request first, before anything else?
I am aware I can just curl it or whatever to see the response...
Go to chrome://omnibox/ and check [x] Prevent inline autocomplete
The answer to this has to be in chrome://chrome-urls
you should see something like the following:
The "full" set of settings is on the chrome://chrome-urls . Also chrome://flags is worth to check. As a side by enabling some experimental features from chrome://flags you can greatly enhance your browsers HTML5 support which can be checked at HTML 5 support .
else:
Clear browser historythen go to settings and under Privacy --> uncheck [] Use a prediction service to help complete searches and URLs typed in the address bar .
Another idea is to examine carefully if using linux the output of the following command for parameters:
ps -aux | grep google-chrome-stable
in my case the output tells me a lot about what parameters are used at launch by default:
/opt/google/chrome/chrome --type=renderer --disable-layer-squashing --enable-transition-compositing --enable-deferred-image-decoding --enable-display-list-2d-canvas --enable-distance-field-text --enable-encrypted-media --enable-experimental-canvas-features --enable-experimental-web-platform-features --enable-lcd-text --enable-one-copy --enable-overlay-scrollbar --enable-renderer-mojo-channel --enable-smooth-scrolling --enable-viewport-meta --enable-webgl-draft-extensions --enable-web-midi --enable-zero-copy --max-tiles-for-interest-area=512 --enable-plugin-power-saver --lang=en-US --force-fieldtrials=AutoReloadExperiment/FlagEnabled/AutoReloadVisibleOnlyExperiment/FlagEnabled/ChromeSuggestions/Default/DomRel-Enable/enable/EnhancedBookmarks/Default/ExtensionContentVerification/Enforce/ExtensionInstallVerification/None/GCM/Enabled/MaterialDesignNTP/Enabled_forced/OmniboxBundledExperimentV1/StandardR4/PasswordGeneration/Disabled/PrerenderFromOmnibox/OmniboxPrerenderEnabled/QUIC/FlagEnabled/SafeBrowsingIncidentReportingService/Default/SettingsEnforcement/no_enforcement/UMA-Dynamic-Binary-Uniformity-Trial/default/UMA-Population-Restrict/normal/UMA-Uniformity-Trial-1-Percent/group_09/UMA-Uniformity-Trial-10-Percent/group_02/UMA-Uniformity-Trial-100-Percent/group_01/UMA-Uniformity-Trial-20-Percent/group_04/UMA-Uniformity-Trial-5-Percent/group_16/UMA-Uniformity-Trial-50-Percent/group_01/UwSInterstitialStatus/OnButInvisible/VoiceTrigger/Install/WebRTC-IPv6Default/Default/ --enable-crash-reporter=9F2AFD26-85F1-40CB-991F-0980EF2C4D14 --enable-offline-auto-reload --enable-offline-auto-reload-visible-only --enable-offline-load-stale-cache --enable-app-window-controls --enable-embedded-extension-options --enable-experimental-extension-apis --enable-scripts-require-action --enable-nacl --enable-nacl-debug --enable-streamlined-hosted-apps --enable-web-based-signin --javascript-harmony --out-of-process-pdf --enable-delegated-renderer --enable-impl-side-painting --num-raster-threads=4 --enable-gpu-rasterization --channel=5035.27.2136067136
Even yet another idea is maybe to write a small widget using python's tkinter and the webrowser modules, it could even get its input from the clipboard.
this command could be a work around solution too:
python -m webbrowser -t "http://ip.ip.ip.ip:portport/file/"
Another option yet is to use The Omnibox API and embed a custom omnibox in a simple webpage. Here are the omnibox api ready made samples.
You can avoid this by adding a "/" at the end of the URL http://0.0.0.0:5000/api/
or
You can try to add a null search engine with a URL of http://%s and null keyword.
Go to the search engine settings:
Open Settings.
Click Manage search engines
At the bottom of the Other search engines section add a new search engine.
Using #Juan Buhagiar answer as a starting point, I added your URL as the URL of a default search engine:
Other search engines
| MyAPI | 0.0.0.0 | http://0.0.0.0:5000/api/venues/show/45/20/cafes?rubbish=%s |
That just worked. The only drawback is that you get a redundant query to your request:
http://0.0.0.0:5000/api/venues/show/45/20/cafes?rubbish=0.0.0.0%3A5000%2Fapi%2Fvenues%2Fshow%2F45%2F20%2Fcafes%2F
instead of plain:
http://0.0.0.0:5000/api/venues/show/45/20/cafes
So as long as it does not conflict with your own GET, you can just ignore it.
Simplest way - Add a bookmark with your desired address. This time it will not goto search. This may not be useful if you have to change variables regularly, which was not my case.
This is helpful if you do not want it to disable omnibar search completely since it is quite a useful feature.
I've installed a Javascript formatter into ST2. For some reason, the keyboard shortcut is not working suddenly. When I search in my keymaps for alt+ctrl+f I don't see anything. How can I figure out why this shortcut is no longer doing anything? You can see the poorly formatted js in the screenshot. If I call the command via Sh+Ctrl+P it works fine. Does the Ctrl+Alt+F on the right side in the screenshot indicate that that command is bound to that shortcut, or just that the package thinks that it is bound to that?
I would suggest you try to use FindKeyConflicts. This plugin will:
Assist in finding key conflicts between various plugins. This plugin
will report back shortcut keys that are mapped to more than one
package. This does not guarantee that the listed plugins are
necessarily in conflict, as details, such as context, are ignored.
This is simply a tool to help assist what plugins may be conflicting.
Is there a way to disable and replace shortcut commands in Google Chrome. I want to use Chrome for a public computer that only can access one site. Because of this I want to disable keys like Ctrl+Tab, Ctrl+T, Alt+F4 and I want to change F11 to a command like Ctrl+Shift+Alt+J (example) to stop users from exiting full screen mode.
Settings on the network block everything but a specific domain but now I want to block the user from exiting the browser.
I know this question is quite old, but I found a solution that works for me (and probably many others too).
If you don't use the CTRL and ALT keys in the browser at all, you can disable them in the OS itself.
Under Linux, I used xmodmap -pke to find out which key is mapped to which code.
Knowing the keycodes, I used:
xmodmap -e "keycode 37 = "
xmodmap -e "keycode 105 = "
to disable both left and right CTRL keys (to prevent something like CTRL+W, CTRL+T, etc.), then with:
xmodmap -e "keycode 133 = "
xmodmap -e "keycode 134 = "
both left and right SUPER keys (Windows Keys) (to prevent opening the start menu and such fancy stuff)
and then finally with xmodmap -e "keycode 105 = " I disabled the ALT key (to prevent ALT+F4, etc.).
And optionally, we can disable F1 too, so that the Chrome Support Page doesn't open, with: xmodmap -e "keycode 67 = "
Finally, let Chromium or Chrome lock the rest down for us using Kiosk Mode:
chromium-browser --kiosk http://example.com/
or
google-chrome --kiosk http://example.com/
And right click is already disabled in kiosk mode, so we don't need to change anything there.
With all that done, the end user can only navigate with the mouse within the predefined webpage (And links leading to some other content, of course) and write stuff with the normal characters on the keyboard, but nothing more.
Reloading may be still possible (F5), but even that can be disabled with: xmodmap -e "keycode 71 = "
Caution: Please execute xmodmap -pke first to discover if your keyboard or OS have the same keymapping, or you may disable other normal keys without knowing.
Caution 2: Note that if you've done everyhing above and then launch Chrome or Chromium in Kiosk Mode, you can't get out anymore! Only physically pressing the power button or killing the application over SSH or Telnet will let you resume normal operation again.
To make those changes permanent, read the end of this guide:
https://stackoverflow.com/a/11219056/3525780
EDIT: To those who have problems disabling the F1, F5, etc. keys, use following as a workaround:
xmodmap -e "keycode 67 = Escape"
(Somehow those "F keys" need to be assigned to an already existing and assigned key)
Having recently encountered the same kiosk-type problem (and not being able disable all keys in Chrome) I eventually found a solution which I thought I would share:
Using node-webkit I created the following package.json file:
{
"name" : "mykiosk",
"window" : {
"fullscreen" : true,
"toolbar" : false
},
"main" : "http://the-one-and-only-allowed.url/"
}
Launch with: ./nw
All function keys are blocked. Ctrl+N/T do not create tabs. It is quite nice
One last javascript/onload trick to disable the right-click context menu:
window.oncontextmenu = function(ev) {
ev.preventDefault();
ev.stopPropogation();
return false;
}
Chrome has Kiosk Mode, but that won't prevent users from using OS keyboard shortcuts (like ALT+F4, which aren't part of Chrome. Windows handles those). To start it in Kiosk Mode, run it using these parameters:
chrome.exe --kiosk http://www.google.com
My public library actually did something pretty awesome: they installed an extremely minimal Debian build on their kiosks, and run Google Chrome on each one. There are no close buttons, and no desktop to get into, so this deters virtually all the CTRL+ALT+DELETE hackers out there. ALT+F4 doesn't work either, and closing the browser by right-clicking opens up another one instantly.
But they forgot to get rid of GRUB's 10 second timeout, which lets users (well, me) get into recovery mode -_-, so I'm working with them to get that fixed...
I'd seriously consider Linux, as you can install it really quickly on multiple computers and basically forget about viruses and security. But the downside is that there isn't a "Administrator Panel" for you to tweak things with. You'd have to whip out nano (sorry, can't get used to vim) and edit some config files.
For me using version 52.0.2743 the --kiosk tag didn't work, but the --app="http://www.example.com" did what I wanted. (Disabled chrome keyboard shortcuts so I could use shellinabox + nano without issue.)
Also works on Chrome Canary (for which the --kiosk tag also didnt' work).
Other possibly helpful links:
Chrome support how to make a Kiosk App: https://support.google.com/chrome/a/answer/3316168?hl=en
Kiosk App for Chrome: https://chrome.google.com/webstore/detail/kiosk/afhcomalholahplbjhnmahkoekoijban?hl=en