-
-
Notifications
You must be signed in to change notification settings - Fork 866
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
MAINT: Update conftest to check PyErr_Occurred #1113
Conversation
@dalcde Every test fails here, the errors are let Module = {};
Module._module = Module; // add this
self.Module = Module; |
When setting up a test, we run `Error.stackTraceLimit = Infinity` via
run_js, when pyodide has not yet been set up.
|
I can merge this after the tests from #1111 are tidied up |
Okay, will do. |
Oops should have merged first. |
Also, the |
For whatever inexplicable reason the tests seem to be genuinely failing |
I will try to figure out what's wrong tomorrow. |
0325a50
to
41c66bb
Compare
I also added a second test with the ( |
Now the docs is failing because |
I think there are some bugs where pyodide leaves the python error flag in an inconsistent state. These bugs invoke unspecified behavior in CPython. They often go unnoticed because the error flag is frequently cleared or overwritten. However, under the right circumstances they can show up, often due to changes that don't directly touch the code causing the problem. For instance, bug #1110 was exposed in PR #1098 because I added extra error checking code, even though bug #1110 is caused by problems in
python2js.c
andpython2js_buffer.c
which were both unchanged in #1098.I have seen similar symptoms to the ones caused by #1110 without hitting that particular codepath, so I strongly suspect there are still other similar bugs in the code. This will help by double checking whether the Python error indicator is in an invalid state at the end of each
selenium.run_js
call.