Error Installing with uv: UserWarning: Unknown distribution option: 'test_suite' | Error: pg_config executable not found.

[jim256]

dlt version

n/a

Describe the problem

I'm trying to use https://github.com/datacoves/dbt-coves, which has dlt as a dependency. When I try to install dbt-coves using uv, it fails due to an issue with dlt[redshift]:

uv init
uv add "dlt[redshift]>=1.0"

Fails with:

Using CPython 3.12.9 interpreter at: /usr/local/bin/python3.12
Creating virtual environment at: .venv
  × Failed to build `psycopg2cffi==2.9.0`
  ├─▶ The build backend returned an error
  ╰─▶ Call to `setuptools.build_meta:__legacy__.prepare_metadata_for_build_wheel` failed (exit status: 1)

      [stderr]
      /Users/<user name>/.cache/uv/builds-v0/.tmpf1hFEF/lib/python3.12/site-packages/setuptools/_distutils/dist.py:289:
      UserWarning: Unknown distribution option: 'test_suite'
        warnings.warn(msg)
      Error: pg_config executable not found.
      Please add the directory containing pg_config to the PATH.

      hint: This usually indicates a problem with the package or the build environment.
  help: If you want to add the package regardless of the failed resolution, provide the `--frozen` flag to skip locking and
        syncing.

Is this known / intentional?

Expected behavior

Installation succeeds without issue

Steps to reproduce

uv init
uv add "dlt[redshift]>=1.0"

Operating system

macOS

Runtime environment

Local

Python version

3.12.9

dlt data source

N/A

dlt destination

No response

Other deployment details

No response

Additional information

Mirror of issue in dbt-coves repo:
datacoves/dbt-coves#508

https://pypi.org/project/psycopg2cffi/ indicates that psycopg2cffi is about 4 years old, never left "Alpha" status, and only supports up to Python 3.8
Is this the same one dlt is using?

[rudolfix]

@jim256 this is very unusual. psycopg2cffi will be installed only on PyPy. This is a compiled python version. other possibility is that uv does not understand the marker we use... I'll kick this dependency out. postgres destination will not work on pypy anyway.