16 tests failed about technical HOT 4 CLOSED

@xmatthias

Thanks for the hint about plugins. Some of the plugins caused failures.
All tests passed after these plugins were removed.

Thank you for your help!

from technical.

I'm not entirely sure what you're trying to do / run.

the output doesn't really look like default pytest output, and all shown "failures" appear to only be warnings.

Please try to run pytest regularily (pytest without any arguments) - and post the full output.

Aside from that - it's most likely some setup problem on your side.

tests run fine on all 3 supported system-types (linux, windows, macos) - which actions also show.

from technical.

These messages are printed during the FreeBSD tests of the port finance/py-technical.

pytest-7.4.3 is run.

Here is a complete test log.

from technical.

@yurivict is finance/py-technical an unofficial fork of technical? if yes, you may want to seek support there ?

you can install this module via pip (pip install technical) - which you'll see works just fine.

you should also test the following sequence (make sure to have ta-lib installed first):

git checkout https://github.com/freqtrade/technical.git
cd technical
python3 -m venv .venv
. .venv/bin/activate

pip install -r requirements-dev.txt

pytest

You'll see that tests pass - and i don't see why they shouldn't pass on freeBSD.
If they don't - then it's a free-bsd problem, as this sequence works fine on all other supported systems (linux, macOS, even windows).

since i know no CI provider offering a free freeBSD instance (at least GHA does not) - I'd not know how to test against freeBSD.
My assumption is that freeBSD is just another linux - if that's not the case, we will probably not be able to support it, unless the freeBSD project offers this for free to projects it tries to port.

I've no clue what that fork (or whatever that is) does that makes this fail - but one thing that i find odd is the number of (outdated plugins used by the output you've shown.

plugins: datadir-1.5.0, cov-4.1.0, randomly-3.12.0, timeout-2.1.0, time-machine-2.11.0, forked-1.6.0, rerunfailures-11.1.2, flaky-3.7.0, aspectlib-2.0.0, xdist-3.3.1, env-0.6.2, mock-3.10.0, flake8-1.1.1, anyio-4.0.0, hypothesis-6.88.1, asyncio-0.21.1, trio-0.8.0, pytest_httpserver-1.0.8, flask-0.15.1, subtests-0.11.0, black-0.3.12, checkdocs-2.10.1, mypy-0.10.3, black-multipy-1.0.1, enabler-2.3.1, benchmark-4.0.0

comparing this to the actual CI output, that's WAY more plugins than what should be used.

How it looks like in the technical CI:

plugins: cov-4.1.0, mock-3.12.0, random-order-1.1.0

source (the most recent run, but the ACTUAL runs don't matter).

Most, if not all failures come from plugins we don't use.

from technical.