-
Notifications
You must be signed in to change notification settings - Fork 283
Fix type error in decision analysis #784
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix type error in decision analysis #784
Conversation
Thank you for fixing this. Can you also extend the decision test to raise this exception without your fix? |
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## master #784 +/- ##
==========================================
+ Coverage 95.61% 95.63% +0.01%
==========================================
Files 40 41 +1
Lines 4041 4082 +41
Branches 793 806 +13
==========================================
+ Hits 3864 3904 +40
Misses 100 100
- Partials 77 78 +1
☔ View full report in Codecov by Sentry. |
I have extended the decisions test with a setup using a second cpp that is raising the issue, but only for clang (14) and not for gcc (10). The two failing test cases require an update of the json and html output reference. The "--update-reference" command did not work for me... I also do not have all the required build environments. Could you straigthen out these references yourself? |
The switch is written with a The change in the makefile is not working for us. Can you modify the test to use a enumeration in the same C-File without adding another compile unit? |
I tried different switch-variations (with break, return, additional brackets, ...) to make it come up with one cpp but it does not work. There have to be two cpp's. As far as I understood it, the code in one cpp is creating some actual coverage of the switch statement, the other one only references the containing class but does not call the relevant function, so it ends up with 0 coverage and we end up with an empty list. I have no real insights on how gcov works exactly but that is my best guess about whats happening after trying several setups half of the day. As I am a bit out of time/resources to investigate the problem any further, I can not continue working on it - at least for the next couple of weeks. Sorry about that. |
5f44370
to
8494dd9
Compare
daa1893
to
585bb88
Compare
c6eb6c8
to
2527c5b
Compare
Proposal for fix of #783
Avoiding TypeError exception if
self.coverage.lines
is empty during decision analysis