Bug: Installation option [full] is missing since v1.0.6

[detlefarend]

1 Describe the bug
The installation option [full] is missing in the current version 1.0.7. When using this option, the latest version providing it is installed - here 1.0.5.

2 To Reproduce
Steps to reproduce the behavior:

1. In a shell: pip install mlpro-int-sb3[full] --upgrade
2. Version 1.0.5 is installed instead of 1.0.7

3 Expected behavior
The latest version - here 1.0.7 - should be installed.

4 Screenshots

5 Additional Informations

5.1 Operating System

• Linux
• Windows
• Mac OS

[steveyuwono]

Hi Detlef, this probably happens because you use Python version 3.10. I followed the main MLPro package, which requires Python >= 11, as defined here: https://github.com/fhswf/MLPro-Int-SB3/blob/v1.0.7/setup.cfg.

If you are using 3.10, then it will directed to mlpro-int-sb3 1.0.5, and 3.9 is to 1.0.2.

[detlefarend]

Hi Detlef, this probably happens because you use Python version 3.10. I followed the main MLPro package, which requires Python >= 11, as defined here: https://github.com/fhswf/MLPro-Int-SB3/blob/v1.0.7/setup.cfg.

If you are using 3.10, then it will directed to mlpro-int-sb3 1.0.5, and 3.9 is to 1.0.2.

Hi Steve, I am on Python 3.12...

[steveyuwono]

Hi Detlef, this probably happens because you use Python version 3.10. I followed the main MLPro package, which requires Python >= 11, as defined here: https://github.com/fhswf/MLPro-Int-SB3/blob/v1.0.7/setup.cfg.
If you are using 3.10, then it will directed to mlpro-int-sb3 1.0.5, and 3.9 is to 1.0.2.

Hi Steve, I am on Python 3.12...

Hi Detlef, we ended up with mlpro-int-sb3 1.0.5 because pip resolved dependencies due to numpy version to an older mlpro v1.9.8. Hence, it hits a conflict. But now, sb3 supports numpy 2.0, but mlpro-int-sb3[full] still had a restriction blocking it. This restriction has already been fixed and merged to main and will be included in the next release.