How to get pytest to not delete successful test directories

I'm using pytest and it's been great. My tests use the tmp_path fixture to hold a bunch of outputs (log files, plots, etc.). I want to keep these so I have a record of what the performance of all tests are, not just pass/fail. That way as I make changes I can see whether things are faster or slower, diff the logs, etc. Currently pytest only keeps the directories for tests that fail. I've been through the documentation and can't find any flags that will convince it to do otherwise.

Here's a sample test:

from pathlib import Path


def test_file_to_keep(tmp_path):

    # write some outputs from the test.
    my_file_to_keep = Path(tmp_path) / "my_file.txt"
    with open(my_file_to_keep, 'w') as out_file:
        out_file.writelines("some good output.")
        
    assert my_file_to_keep.exists()

I want to be able to look at the tmp_path directory and examine the contents of my_file.txt.

The temporary directories created with tmp_path are kept after each test run even if the test passes.

Running your example code, I can go to a terminal and do:

marco@sol:/tmp/pytest-of-marco/pytest-current/test_file_to_keep0$ cat my_file.txt 
some good output.

Please note that the retention policy can be configured with the tmp_path_retention_count and tmp_path_retention_policy variables, as described in the documentation here.

Also please note that you don't need to create a Path object with Path(tmp_path), as tmp_path is a pathlib.Path object, as described on the documentation.

You could even simplify it with:

def test_file_to_keep(tmp_path):

    # write some outputs from the test.
    my_file_to_keep = tmp_path / "my_file.txt"
    my_file_to_keep.write_text("some good output.")

    assert my_file_to_keep.exists()