We have several TVs, each with a Raspberry Pi hooked up, so they can be used as information dashboards. The current setup stays on one tab the entire time.
I am trying to set it up so that it will automatically rotate through the chromium tabs every 15 seconds, but am not sure how to approach this.
The Pi's, as far as I know, only have a headless OS with chromium installed, which is started on boot in the xinitrc file.
I know there are some chrome plugins that can do this, but I don't think they can be installed on the command line.
Any advice would be appreciated
You can use xdotool with watch to do this:
# rotate tabs after 30 seconds.
watch -n30 "xdotool search --onlyvisible --class chromium key ctrl+Tab"
Related
I'm trying to configure puppeteer to launch Chrome on a specific screen (or workspace) on macOS. I've found the --window-workspace CLI argument but it doesn't work. I tried every number from 0 to 10 (the number of workspaces I have on 3 screens), and the Chrome window shows up in the same workspace.
/Applications/Google\ Chrome\ Canary.app/Contents/MacOS/Google\ Chrome\ Canary --window-workspace=3
Is there a different way of identifying the workspace?
This is especially useful when working with multiple monitors and debugging puppeteer with headless mode turned off, having the ability to designate a screen where the window shows up would better the developer UX.
I had built console core image for raspberry Pi3, and I am able to boot the Rpi3 successfully using SD card.
I have created an electron app which is able to launch on remote display over ssh.
However when I launch the application on Monitor connected through HDMI Cable, It gives following error
Can not open Display :0.0
I have seen many people asking this question for not able to launch on remote display, which works fine in my case.
Can anyone help in this?
With console core, you don't have an X server running to display on.
You can upgrade your system to include the X installation -- search for 'install pixel desktop' -- or you could just refresh the card with the lite or full image.
If you set the Pi to auto login to the pi account, you can start your application on login by adding it to the bottom of .xinitrc. You could also start it from a remote ssh, displaying on the local display, by setting DISPLAY=:0 in the environment before you start. You'll need to explore the world of X Windows authentication to make this work. See the man page for the xhost command, for instance.
(This is an expansion of LetoThe2nd's comment, which probably should have been an answer instead.)
Console core image means that there is no xserver running, and hence no display :0. Try getting started with core-image-x11 maybe, or whatever suits the RasPi.
I have a strange issue whit my emulate touch functionality on Google Chrome version 49.0.2623.110 m.
I'm using laptop + monitor (both Dell) and running Windows 8.
The issue is that while Chrome is on my secondary monitor and touch screen emulation is enabled I cannot see a touch screen pointer (black circle) nor any kind of pointer, but when I move my Chrome to my main laptop display it would show. I tried disabling laptop display while using external monitor and it is the same, I cannot see the pointer. This was not a problem with previous Chrome version.
Did anyone had similar issues with this version?
Thank you
EDIT:
I've change the laptop, updated to win10 and updated Chrome to 50.0.2661.87 m version and I still have the same issue (colleague of mine has the same problem as well).
A colleague of mine "solved" the problem by enabling Windows' Mouse pointer trails at minimum length. I find it too annoying to work with the setting enable, but it does work if you really need it.
Mouse settings
They did some changes in recent version of Dev Tools and some options are now gone. Ensure you have "Mobile" or "Desktop with touch" selected, as you can see on the screenshot.
I toggled into full screen mode on QGIS 2.2 on my Mac, and now the menu bar won't show and I can't toggle out of it. From the documentation, it looks like it should toggle with F11, but it's not working. I've seen people having similar trouble on Ubuntu and Windows 7, but nothing for OSX. Any ideas? I appreciate it.
Just spent close to an hour with this problem- QGIS 2.2 on mavericks. Really vexing! I tried reinstalling (after trashing the previous version as instructed in the install notes), tried removing the hidden .qgis2 folder in the home directory, tried disabling all spaces on the machine (I started with 8 desktops), tried all permutations of Apple-F, ctrl-F, apple-shift-F, etc...
Finally, and simply, the answer was this:
F11 is documented as the toggle key, but in my mostly stock setup this was bound in MissionControl to 'display the desktop'. I rebound that function to F1, and then with QGIS up, hit F11 (on my MBP I need to also hold the fn key otherwise it's 'volume down') and I was out of Full screen mode.
Hope it works for you!
I am running Selenium Webdriver locally on my windows home server (windows 7 professional actually) which logs to a site and sends some emails. It's a java program with a single thread running.
I can see Firefox opening up just fine when running through Eclipse and follow the clicks etc as it logs on without problems.
I have set up a Task Scheduler task to start it up every morning though and run the jar. The program runs fine but when Firefox opens it is almost invisible, I can see the grey frames of it but everything else is transparent and invisible and if you drag something in front of it the graphics get stuck.
When clicking on Properties on Task manager I can see that it started with firefox.exe - foreground whereas when I open Firefox normally it says -browser.
The program still does what it should do, it's just Firefox is not visible at all when running through java -jar X.jar etc.
Any ideas why?
Windows 7 Professional
WebDriver 2.32
Firefox 21
I suspect it has something to do with the anonymous Firefox Profiles created each time firefox runs. I usually delete them from Temp once a week.
P.S No idea if it matters but server is headless and I am accessing it through TeamViewer.
Issue solved by deleting firefox profiles and loading from scratch. Weird, not sure yet what caused this and if it happens again.
Perhaps I will create a specific firefox profile and use instead of a anonymous one created each time.