I am trying to open the firefox browser from the robot framework. I have downloaded and extracted the gecko driver file. Additionally, I also add path to environment variable. But When I run test, then it returns an error with message:
"SessionNotCreatedException: Message: Expected browser binary location, but unable to find binary in default location, no 'moz:firefoxOptions.binary'
capability provided, and no binary flag set on the command line".
This is my code
Can anyone help me? Many thanks!
Related
I'm using GOLAND and import a package by go mod vendor.
It works good with compiling and building.
I can find types under this package other than the problematic file
I can't find any types defined in the problematic file
I've tried invalidate caches and restart the IDE, however it still not works.
I've find out why. Golang is failed to recognize file larger than 4.19M. Can change the file size limit accroding to https://youtrack.jetbrains.com/issue/GO-5029
PDSOE version 4.5.2, OE 11.7.1 - 64-bit - Win 10 64
I have an Webspeed (.w) opened in my PSDOE.
If I simply right click over this code and choose Check Syntax in the context menu (Ctrl+Shift+C)
it will return me an error from an include file used by this .w regarding the get-cookie():
Syntax Check:
** Unable to understand after -- "GET-COOKIE". (247)
In my .w, I have all the necessary Webspeed include files - including proto.i that will have the forward get-cookie function signature and all that.
So far so good, I could have thought of Propath and all that, but two things make things more complicated now:
If I choose to compile the code instead of checking syntax, it will compile just fine - which tells me that all my include files and propath are correctly configured and all set, a .r is produced
If from PDSOE, with my code still opened, I go in Eclipse \ OpenEdge \ Tool \ Procedure Editor , it will open a Progress Editor with the same code in place right, and from this procedure editor, if I simply right click and check syntax, it will return no errors at all.
I downloaded OE 12.3 and tried the same thing, got the same behavior.
I checked all the propath and it was right (code compiles)
I tried restarting the AVM, no difference
I tried a shared AVM, no difference
I am running out of ideas specially after I got the same issue on 12.3.
Would you happen to have any other idea or could thing of anything else that could explain the check syntax behavior in PDSOE?
Thanks!
So your project configuration sounds o.k.. You should open a call with Progress tech support IMHO.
If your .w file is a mapped web object and not a CGI wrapper, you might have little luck with support for those on OpenEdge 12.3 - as PASOE does not support running them. There's an enhancement request open for votes: https://openedge.ideas.aha.io/ideas/OPENEDGE-I-753?utm_source=idea_comment_mailer&utm_medium=email&utm_campaign=subscribers
On OpenEdge 11.7.1 this should still be supported though.
CGI wrappers should work fine on OpenEdge 12.3 though.
I am unable to have my screenshot automatically embed itself into the RobotFramework log file after my test is run. I am using:
SikuliX 1.1.4
RobotFramework 3.0.4 JAR distribution (placed in C:\Users[user]\AppData\Roaming\Sikulix\Extensions
I am able to run my simple test and also have a report/log kicked out - along with a screenshot. The only trouble is that the screenshot is not visible in the log file when viewed in a browser (tried Chrome and FF).
This is my code as written in Sikuli:
runScript("""
robot
*** Settings ***
Library Screenshot
*** Test Cases ***
Try to take a screenshot
Take Screenshot mypic 80%
""")
My folder where the logs are outputted looks like this after the test is run:
./log.html
./mypic_1.jpg
./output.xml
./report.html
./test2.robot
So far, all I have come across when searching the net is that the embed should 'just happen automatically' when using the 'Take Screenshot' keyword.
I have not tried this scenario with RobotFramework by itself - I think that may be my next step, but hoped someone would have an easy solution since I am wanting to use these two packages together. I am wanting to leverage Sikuli to help with my GUI intensive tests and RF for its' reporting and screenshotting capability.
I also had a similar problem with Sikurix 1.1.4 (with RF 3.0.0)
When overwriting the Robot directory of Robot Framework 3.0.4 in "C:\Users\xxxx\AppData\Roaming\Sikulix\Lib\robot", images are now embedded in Log.html.
Even though the comment made by temp temp didn't solve my issue right away, it did put me on the right track. After copying over the pip installed version of RF as mentioned in the comment, I removed the RF JAR file from my Extensions folder (not even sure why I tried this). Once I removed the JAR file, the image started embedding itself in the log. So it seems the combination of the RF version included with Sikuli combined with the RF JAR file was causing my issue.
when running the application from inside the mingw console, it executes without any issue. But when I try to execute the same exe file in the debug folder of windows explorer, it says QtCored4.dll was not found. I even added the bin folder location to the environment variables, but then that produces another error saying : The procedure entry point _Z5qFreePv could not be located in the dynamic link lib... .So, can anyone give me a hint on how to solve this problem? Thanks.
As nax83 pointed out above, the problem was the path variable ambiguity due to multiple version of qt installed on my system. I removed all other qt-related values from my path variable and then explicitly specified all the paths to one single version. Recompiled and built the application, and there was my application, running without any jerks and problems.
I hate to ask the same question others have asked in Stackoverflow, but I still can't figure out why Installaware 7.0 R2 has this problem.
I need to build a installable CD for the previous version of my software. So, I am using what came with RAD Studio XE to do this. After creating a new Win32 setup for installation and customizing it, I built and test ran the project. In the middle of the installation, it raised an error, "Error Folder Path 'C:' contains an invalid character." So, I went back and created another win32 project and without doing any changes to anything I compiled, built and test ran the project. It worked without any error. After spending hours changing one thing at a time and testing it over and over again until I hit the error message again, I finally found the offending property in the Installaware. The error is raised thereafter once you change the Target folder textbox default value which is $PROGRAMFILE$\$TITLE$\ or Shortcut folder textbox default value which is $TITLE$. They only way to get passed this error is by not changing those default folder paths and allowing the user to change the folder path during installation.
It sort of annoying especially when you spend thousands of dollar purchasing these software from Embarcadero and Codegear. Is there a fix for this? Does anybody know?
These variables are resolved automatically to full folder paths. If you delete them or set an incorrect value, your package will not be able to resolve the installation path (hence the error). So the path edit controls should have valid default values.
If you don't want to allow the user to change your installation path, you can try deleting the dialog which offers this option. I'm not sure if that version of InstallAware supports it though.
If you don't like InstallAware, there are some good free or commercial alternatives which may help you.