googletest: need a way to resume from the TEST that crashed
Requested by Timur Iskhodzhanov <timurrrr@google.com>:
Consider we have a large number of tests in one executable (e.g.
Chromium unit_tests or ui_tests)
Use case:
One's local change has introduced a failure in a test X.Y which is
executed somewhere in the middle of the test run
(e.g. the whole binary takes 10 minutes, the failure happens after 5
minutes of running).
If he/she is sure the tests that are executed before X.Y are not
affected by the fix - he/she should be able to skip them and run only
X.Y and the following tests.
This feature will be especially useful for those who develop Valgrind-
like tools and test them using binaries based on googletest.
In this case it's quite common that some particular test is failing
when run under the tool and when it's fixed we should run only those
tests that follow X.Y
By "failure" I meant a crash or non-googletest assertion.
So the execution stops when X.Y fails.
Example: Chromium code uses CHECKs inside its code and as soon as they
fail, it prints out the stack trace and aborts the execution.
Another example: if there is a bug in Valgrind-like tool, it may crash
in an internal assertion when running X.Y and stop the test execution.
Original issue reported on code.google.com by zhanyong...@gmail.com
on 7 Sep 2010 at 7:07
About this issue
- Original URL
- State: closed
- Created 9 years ago
- Reactions: 1
- Comments: 29 (27 by maintainers)
Commits related to this issue
- Publish bazelbuild/rules_go@v0.37.0 (#311) Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im> — committed to hanneskaeufler/googletest by publish-to-bcr-bot[bot] 2 years ago
The attachment “crashsafe.py” no more accessible can anyone share it?