robotframework
RIDE 1.7.4B2 sometime will crash silently
1.7.4b2 running on Python 3.7.4. Win10
usually I have 2 RIDE instance running concurrently with different test suite with different tags.
I found the RIDE disappeared without any error notification several times after the long time run as check, the ride was launched 1 day or more before, and the test may have already run several cycle or just in first cycle. the test will last 10 ~ 30 hours. some keyword has FOR loop embedded FOR loop
and as I tried to get log.html by rebot in command, I will get following error:
E:\Users\Test\Temp\RIDE6i5apnr5.d>rebot --log log.html --flattenkeywords foritem output.xml [ ERROR ] Reading XML source 'output.xml' failed: ParseError: no element found: line 22586760, column 310
I guess the output.xml also has been impact by crash that not written properly.
Is this a known issue or how can I collect log for this?
Thanks, Shawn
@qch1842 My view on this use case of RIDE is the following:
- If you are expecting long runs of your tests, you should use the command line or script (or CI) to start them.
- RIDE is (should be) focused on the development of test cases.
- There may be Operating System mechanisms that can affect a GUI application, like memory management, disk space or lock files restrictions. RIDE itself may have memory leaks, or the unreported crashes like you report.
- In the cases you want to recover a broken output.xml, you should use the
-doption to generate the output.xml, log.html and report.html files in a safe directory for latter access. Not in the temporary random directory RIDE created.
I'd also like to add that rebot won't be able to process an output.xml file if it's too large. My recommendation is to keep the test 1 hour short and to repeatedly rerun the test (you can add timestamp to the report).
Also, in runner tab, please uncheck Show message log. The textctrl has no limit, so it will crash RIDE if too much text is printed to it.
