Should tmpdir clean up after itself? #543
Comments
|
Original comment by Jurko Gospodnetić (BitBucket: jurko, GitHub: jurko): I think pytest already does that. It just keeps the temporary test data around for the last few (5?) test runs, but all the ones before that are removed automatically. I haven't checked the source code, but I think that's how pytest behaves for me - Windows 7 installation. IMO, retaining the temporary test data for the last few test runs is extremely useful when you need to debug a specific test failure. Just my 2 cents worth... Best regards, |
|
I get that some people would prefer to keep some old test run data around. However frequently I have big test data and would like to have it cleaned up (particularly if the tests all pass). Is there a way for one to force cleanup after a run? |
|
Currently not, but it is easy to add a new [pytest]
tmpdir_keep=0Currently the default is |
|
my pytest version is 4.3.1 but not work this is the content in my pytest.ini file: this is the test code: CONTENT = "content"
def test_create_file(tmp_path):
print('tmp_path"', tmp_path)
d = tmp_path / "sub"
d.mkdir()
p = d / "hello.txt"
p.write_text(CONTENT)
assert p.read_text() == CONTENT
assert len(list(tmp_path.iterdir())) == 1
assert 0 |
|
Hi @scottming, The |
|
Should these details -- that |
|
@lazarillo It's documented here: https://docs.pytest.org/en/stable/tmpdir.html#the-default-base-temporary-directory |
|
Thanks, @The-Compiler ! I read that section, but I guess I did not read carefully enough, as I did not notice the
sentence. |
pytest intentionally keeps around a limited number of the previous test temporary directories [1]. This is generally OK, but in our tests that generate initrds, we create a few very large files (both the initrd and kernel in a few copies), which quickly adds up. I had a particularly large initrd (because of some mkosi-initrd shenanigans), and I unded up with dozens of gigabytes of temporary files from the tests. Let's just nuke the dirs where we write kernel data. [1] pytest-dev/pytest#543
pytest intentionally keeps around a limited number of the previous test temporary directories [1]. This is generally OK, but in our tests that generate initrds, we create a few very large files (both the initrd and kernel in a few copies), which quickly adds up. I had a particularly large initrd (because of some mkosi-initrd shenanigans), and I unded up with dozens of gigabytes of temporary files from the tests. Let's just nuke the dirs where we write kernel data. [1] pytest-dev/pytest#543 (cherry picked from commit ef0a69e)
pytest intentionally keeps around a limited number of the previous test temporary directories [1]. This is generally OK, but in our tests that generate initrds, we create a few very large files (both the initrd and kernel in a few copies), which quickly adds up. I had a particularly large initrd (because of some mkosi-initrd shenanigans), and I unded up with dozens of gigabytes of temporary files from the tests. Let's just nuke the dirs where we write kernel data. [1] pytest-dev/pytest#543 (cherry picked from commit ef0a69e) (cherry picked from commit d93341f)
pytest intentionally keeps around a limited number of the previous test temporary directories [1]. This is generally OK, but in our tests that generate initrds, we create a few very large files (both the initrd and kernel in a few copies), which quickly adds up. I had a particularly large initrd (because of some mkosi-initrd shenanigans), and I unded up with dozens of gigabytes of temporary files from the tests. Let's just nuke the dirs where we write kernel data. [1] pytest-dev/pytest#543 (cherry picked from commit ef0a69e) (cherry picked from commit d93341f)
pytest intentionally keeps around a limited number of the previous test temporary directories [1]. This is generally OK, but in our tests that generate initrds, we create a few very large files (both the initrd and kernel in a few copies), which quickly adds up. I had a particularly large initrd (because of some mkosi-initrd shenanigans), and I unded up with dozens of gigabytes of temporary files from the tests. Let's just nuke the dirs where we write kernel data. [1] pytest-dev/pytest#543 (cherry picked from commit ef0a69e) (cherry picked from commit d93341f)
pytest intentionally keeps around a limited number of the previous test temporary directories [1]. This is generally OK, but in our tests that generate initrds, we create a few very large files (both the initrd and kernel in a few copies), which quickly adds up. I had a particularly large initrd (because of some mkosi-initrd shenanigans), and I unded up with dozens of gigabytes of temporary files from the tests. Let's just nuke the dirs where we write kernel data. [1] pytest-dev/pytest#543 (cherry picked from commit ef0a69e) (cherry picked from commit d93341f)
pytest intentionally keeps around a limited number of the previous test temporary directories [1]. This is generally OK, but in our tests that generate initrds, we create a few very large files (both the initrd and kernel in a few copies), which quickly adds up. I had a particularly large initrd (because of some mkosi-initrd shenanigans), and I unded up with dozens of gigabytes of temporary files from the tests. Let's just nuke the dirs where we write kernel data. [1] pytest-dev/pytest#543 (cherry picked from commit ef0a69e) (cherry picked from commit d93341f)
|
Adding comment because this is coming up in related Google searches. There are two options which are now used to configure tmpdir retention:
See retention reference and the issue which planned these updates. |
Originally reported by: Christian Theune (BitBucket: ctheune, GitHub: ctheune)
We noticed that tmpdir isn't cleaning up after itself.
Is there a reason why not? The docs don't mention it. Otherwise: I'd be happy to provide a patch to make tmpdir clean up after itself.
The text was updated successfully, but these errors were encountered: