-
-
Notifications
You must be signed in to change notification settings - Fork 346
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
Failure in test-python-utils with v2.6.0a2 #1040
Comments
@santoshshanbhogue ... thank you for reporting this issue. Could you please provide some additional information about your specific setup? (both hardware and cantera configuration). I am asking as there are some apparent differences between your system and the GitHub OSX test environment (where those tests pass). |
Hardware info: Cantera configuration: System kernel info: Let me know if you need anything else |
I think the NumPy, Cython, and Python versions would all help too. |
I'm using Python 3.7.4 |
Interesting - I was mostly curious about Apple Silicon vs Intel processor. I don’t use a Mac and there’s nothing out of the ordinary, so I don’t have any ideas at the moment. |
I'm mostly interested in which C++ compiler C++ standard library version are being used, and what version of Boost is present. One of those is responsible for the strings that are being returned here. I think what's happening is that we've ended up setting some very specific expectations for the return values of |
I'm using g++ and compiling with --std=c++0x Apple clang version 12.0.0 (clang-1200.0.32.29) |
It's interesting that this is (I think) the version of Boost that first introduced the
p.s. You aren't actually using g++. As you can see from the version output mentioning clang, this is just a very confusing alias that Apple has introduced. |
Interesting, actually, no, the 1.55.0 version that I use does not have the I just upgraded to v1.76 (the latest) which has the I also just checked (https://cantera.org/compiling/dependencies.html#other-required-software) and it states that Boost > 1.54 will suffice By the way, is there some plan for development in MATLAB for which you need contributors? |
We aren't deliberately requiring a newer Boost version, and apparently the note I made in I'd like to keep this open, as there are apparently a couple of little hiccups here that warrant fixes. |
I looked into which version has |
Without Boost, the string versions of certain typenames are compiler-dependent. Make sure we use consitent names for all of the types where we end up checking for the names in the test suite. Also, correct comment indicating the minimum version of Boost that has the demangle function. Fixes Cantera#1040
Without Boost, the string versions of certain typenames are compiler-dependent. Make sure we use consistent names for all of the types where we end up checking for the names in the test suite. Also, correct comment indicating the minimum version of Boost that has the demangle function. Fixes Cantera#1040
Without Boost, the string versions of certain typenames are compiler-dependent. Make sure we use consistent names for all of the types where we end up checking for the names in the test suite. Also, correct comment indicating the minimum version of Boost that has the demangle function. Fixes #1040
Problem description
Two tests fail on my mac (there are others that fail, but they are comments in the source code that these are expected such as in
test_mixture.py::TestMixture::test_equilibrate2
)Steps to reproduce
run
scons test test-python-utils
Behavior
See the screen dump at the end of this message for more info
System information
Attachments
Additional context
The text was updated successfully, but these errors were encountered: