python-utils | A collection of useful python modules

I am getting the 502 Whoops, GitLab is taking too much time to respond error on my fresh Manual installation of Gitlab-ce 13.2.4 on CENTOS 8 (nginx/1.16.1).

I downloaded the rpm gitlab-ce-13.2.4-ce.0.el8.x86_64.rpm

• dnf -y install policycoreutils-python-utils
• rpm -Uvh gitlab-ce-13.2.4-ce.0.el8.x86_64.rpm
• /etc/gitlab/gitlab.rb >> external_url "https://git.mydomain.com"
• /etc/gitlab/gitlab.rb >> nginx['enable'] = false
• /etc/gitlab/gitlab.rb >> web_server['external_users'] = ['nginx']
• /etc/gitlab/gitlab.rb >> gitlab_rails['trusted_proxies'] = [ '172.16.1.0/24', '192.168.10.0/24']
• /etc/nginx/conf.d/git.mydomain.com.conf >> https://gitlab.com/gitlab-org/gitlab-foss/-/raw/master/lib/support/nginx/gitlab-ssl
• systemctl restart nginx
• gitlab-ctl reconfigure

My git.mydomain.com.conf

I am trying to get python-utils package and utils module work in my anaconda3. However, whenever I open my Anaconda Powershell and try to install the package it fails with the comment

EnvironmentNotWritableError: The current user does not have write permissions to the target environment. environment location: C:\ProgramData\Anaconda3

I searched for solutions and was advised that I update conda. However, when I ran the comment below

conda update -n base -c defaults conda

it also failed with EnvironmentNotWritableError showing.

Then I found a comment that says maybe my conda isn't installed at some places, so I tried

conda install conda

which got the same error.

Then I tried

conda install -c conda-forge python-utils

which also failed with the same error.

Maybe it's the problem with setting paths? but I don't know how to set them. All I know about paths is that I can type

sys.path

and get where Anaconda3 is running.

My pyglet application to view 3-d objects in STL format is acting funny. when I load a stl file, it works fine, but then when I try to draw it, it acts funny. code doesn't crash, but the test file I loaded for a cube doesn't look right. I think it might be joining all of the points in the draw code that I don't want to join. the test file should appear 30 pixels by 30 instead of the whole upper right corner:

a picture showing that a chunk of the display is the cube.

at 150 degrees, looks almost right...

the red lines are the wireframe, green dots vertices, and gray is a face of the cube.

here is my code:

I'm working on a piece of software just now that needs to run on a newer version of Redhat. It currently works fine on 6 and 7 but now needs to install and run on Redhat 8.

There is nothing really specific in the software that will stop it running on the newer OS version but i'm having a problem with the dependency requirements. The script that runs the software needs access to semanage (/usr/sbin/semanage). On Redhat 6 and 7 this was supplied by policycoreutils-python package and was given as a requirement argument when the rpm was built. From Redhat 8 onwards, the package has been renamed to policycoreutils-python-utils (see this link, section 8.6.2).

From rpm 4.13, there is support for boolean dependencies (see here) which would make this an easy fix as I could just do something like -R (policycoreutils-python or policycoreutils-python-utils) but changing the rpm-build package from its current 4.8 version is a major pain as it's used by another team for another piece of software. I also need to keep it a distro independent rpm so can't have different versions of the rpm for newer versions of the OS.

So, my questions:

1) Is there some other way to have a conditional requirement in the spec file with the older rpm 4.8 version?

2) If I was to get approval to upgrade rpm to latest 4.15, is there likely to be any issue on target systems that have older rpm installed e.g. not understanding the conditional?

Update: The rpm will be built once on a RH 6 box and should work on RH 7 & 8.

I am trying to make a WFS request and create a GeoDataFrame (from the geopandas library) with this data. However, when I run my code, an assert raises an error:

Why do seemingly simple/atomic conda installs result in fairly complex uninstalls??

I recently tried the following conda install

I'd like to create a python3.6 env with default anaconda packages. The manual and many online resources say the command is conda create -n py36 python=3.6 anaconda. On other computers that works, but on one particular computer it only installs very few packages. I can't figure out why. I've checked $HOME/anaconda2/env and there's nothing in there. (ubuntu 16.04, installed from anaconda, not miniconda)

I'm trying to execute a python script which pygam ( from pygam import LogisticGAM, LinearGAM ).

when I'm trying to execute this script  I have this error: ModuleNotFoundError: No module named 'pygam'

The problem that pygam is installed with both pip and pip3.

This the result when I try to re-install it using pip:

I could not import utils.denoising_utils even though I installed utils package. I tried from utils import *, which was working fine, but while I try to import utils.denoising_utils it throws an error.

I used :