-
-
Notifications
You must be signed in to change notification settings - Fork 18.4k
pandas 0.23.4 fails unit tests #23638
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
Comments
Maybe a duplicate of #23088 - is this in a clean environment? |
And to add to that, can you clarify how you did your install of |
Hi both, Many thanks for taking a look and apologies for not spotting #23088, although the error messages are somewhat different. It is a clean environment and it's an automated installation, so I can provide all the commands/steps used. However, it's a bit long. The short method below gives a similar result. I'm on a centos7 box. Am I doing it wrong?
After doing this, attempting to run tests results in: $ python -c 'import pandas; pandas.test()'
$ python -c 'import pandas; pandas.show_versions()'
|
Can you try building master and seeing what fails? The warnings like Can you run |
For the above build,
Interestingly, I get a similar error from a fresh python3.7 miniconda environment after doing a I've built and run master in a fresh directory using the same method as above, replacing the pandas build with:
I get:
And:
|
That's the full traceback for the unicode error? I was hoping for more. What's the output of |
Should I read anything into the fact that The recommended way to invoke pip is |
I'm afraid that's the full output. Don't read anything into py_pip2, it's just my terrible naming of a scratch directory. I've repeated the test by doing a fresh install and using Setting 35727 passed, 5143 skipped, 318 xfailed, 6 xpassed, 32 warnings in 678.87 seconds Whereas my build against 0.23.4 reports: 4 failed, 25058 passed, 4183 skipped, 78 xfailed, 26 xpassed, 163 warnings in 424.04 seconds (failures are something to do with the months of the year in different languages, don't know if this is interesting?) $ locale -a
failures from 0.23.4 tests:
|
Do you get the same errors on master with your normal locale? There was a change focused on fixing locale testing issues which gets released in 0.24 though I think it was only focused on Py27 (#22213) |
I get the same utf-8 unicode decode errors against both 0.23.4 and master. They all go away if I change the Once |
I'm not sure what's going wrong here then. Let us know if you're able to
debug it.
…On Tue, Nov 13, 2018 at 9:58 AM Mark Dixon ***@***.***> wrote:
I get the same utf-8 unicode decode errors against both 0.23.4 and master.
They all go away if I change the LANG environment variable from the
default en_GB.utf8 to en_GB.
Once LANG has been tweaked, unit tests are able to completion. 0.23.4 has
4 failures (clearly to do with locales), master has none.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#23638 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABQHIiIMsM0i9p_sKD-72XVTnnGOAfppks5uuuwJgaJpZM4YZ_Df>
.
|
@bodgerer any luck? |
@bodgerer were you able to find out anything further? |
@bodgerer This might not be what you're looking for, but you can just ignore testing certain vendor directories: $ pytest --ignore=../build/lib At least now you can perform testing on the code you write yourself. |
I installed pandas from Anaconda3-2018.12-Linux-x86_64.sh
yielded the following output in a Jupyter notebook:
The output of pd.show_versions() is as follows:
|
same issue, did you solved it? |
Those collection errors are fixed on master. You can wait until 0.24 is released (release candidate this week) or downgrade your pytest. |
Is it correct behavior ? Tried py2 (failed tests) and py3 (tests not started)
Python 2.7.5 pandas.test() finished with 2 failures
Python 3.6.6 pandas.test() halted interpreter
|
We've seen that locale issue elsewhere. Not sure if it's fixed on master. Can you debug the other failures? What's the other warning that's being raised? |
Original issue here is resolved. Other issues are orthogonal or even no longer supported (ex: Py2 issues) so closing as is. If anyone has anything else please open as a new issue |
pandas-dev#24760 pandas-dev#23638 There are some special characters encoded with 'window-1252' in lists created by 'locale -a'. The two know locales with this problem are Norwegian 'bokmål', and 'français'.
pandas-dev#24760 pandas-dev#23638 There are some special characters encoded with 'window-1252' in lists created by 'locale -a'. The two know locales with this problem are Norwegian 'bokmål', and 'français'.
pandas-dev#24760 pandas-dev#23638 There are some special characters encoded with 'window-1252' in lists created by 'locale -a'. The two know locales with this problem are Norwegian 'bokmål', and 'français'.
Problem description
Cannot seem to complete unit tests on a source install of pandas 0.23.4. I'm sure I'm doing something dumb, but cannot see what. Any ideas, please?
Code Sample, a copy-pastable example if possible
Output of
pd.show_versions()
pandas: 0.23.4
pytest: 3.9.3
pip: 18.1
setuptools: 40.5.0
Cython: 0.29
numpy: 1.15.3
scipy: 1.1.0
pyarrow: None
xarray: None
IPython: 7.1.1
sphinx: 1.8.1
patsy: None
dateutil: 2.7.5
pytz: 2018.7
blosc: None
bottleneck: 1.2.1
tables: 3.4.4
numexpr: 2.6.8
feather: None
matplotlib: 3.0.1
openpyxl: None
xlrd: None
xlwt: None
xlsxwriter: None
lxml: None
bs4: None
html5lib: None
sqlalchemy: None
pymysql: None
psycopg2: None
jinja2: 2.10
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None
The text was updated successfully, but these errors were encountered: