Use sys.__stdout__ for terminal encoding check

[nihir27]

What does this PR do?

It's common to redirect sys.std*. Currently, if you've redirected sys.stdout using a custom class and that class does not have an encoding attribute set, trainer.test can fail with an AttributeError. Since the sys.stdout.encoding check here concerns terminals, I propose using sys.__stdout__ instead.

Does your PR introduce any breaking changes? If yes, please list them.

Before submitting

• Was this discussed/approved via a GitHub issue? (not for typos and docs)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes? (if necessary)
• Did you write any new necessary tests? (not for typos and docs)
• Did you verify new and existing tests pass locally with your changes?
• Did you list all the breaking changes introduced by this pull request?
• Did you update the CHANGELOG? (not for typos, docs, test updates, or minor internal changes/refactors)

PR review

Anyone in the community is welcome to review the PR.
Before you start reviewing, make sure you have read the review guidelines. In short, see the following bullet-list:

• Is this pull request ready for review? (if not, please submit in draft mode)
• Check that all items from Before submitting are resolved
• Make sure the title is self-explanatory and the description concisely explains the PR
• Add labels and milestones (and optionally projects) to the PR so it can be classified

Did you have fun?

Make sure you had fun coding 🙃

[awaelchli]

@nihir27 Thanks for sending the PR. Do you have a way to demonstrate this using a small runnable code example? I'm just generally curious.

[carmocca]

Given your motivation, wouldn't it be better to check hasattr(sys.stdout, "encoding")?

After all, sys.stdout is what we will use to print, and __stdout__ is just a reference to the original handle: https://docs.python.org/3/library/sys.html#sys.__stdout__

[nihir27]

hasattr(sys.stdout, "encoding") and sys.stdout.encoding is None is an option that prevents an AttributeError being raised but for the case where one is writing to a terminal as well as a log file (like in https://stackoverflow.com/a/14906787) you lose the handling of the ─ character.

[nihir27]

That particular pattern may be actually be rare and in that case I would favour just doing hasattr(sys.stdout, "encoding") and sys.stdout.encoding is None.

[nihir27]

@awaelchli For example, below would raise an AttributeError

import sys

class Logger(object):
    def __init__(self):
        self.terminal = sys.__stdout__
        self.log = open("logfile.log", "a")
   
    def write(self, message):
        self.terminal.write(message)
        self.log.write(message)  

    def flush(self):
        pass    

sys.stdout = Logger()

https://stackoverflow.com/a/14906787

[awaelchli]

@nihir27 Would you mind adding this example as a unit test case in the file tests/tests_pytorch/trainer/logging_/test_eval_loop_logging.py, similar to the tests there that use EvaluationLoop._print_results? Thanks!

[awaelchli]

@leoleoasd If you find the time, would you mind taking a look at the failing test test_native_print_results_encodings, which is affected given your changes?

[leoleoasd]

Are you mentioning the right person?

[awaelchli]

Nope 🤣 Looks like both of us have too many browser pages open. I meant to ping @nihir27 :)

[awaelchli]

@nihir27 One test is failing here:

    @pytest.mark.parametrize("encoding", ["latin-1", "utf-8"])
    def test_native_print_results_encodings(monkeypatch, encoding):
        import pytorch_lightning.loops.dataloader.evaluation_loop as imports
    
        monkeypatch.setattr(imports, "_RICH_AVAILABLE", False)
    
        out = mock.Mock()
        out.encoding = encoding
        with redirect_stdout(out) as out:
            EvaluationLoop._print_results(*inputs0)
    
        # Attempt to encode everything the file is told to write with the given encoding
        for call_ in out.method_calls:
            name, args, kwargs = call_
            if name == "write":
>               args[0].encode(encoding)
E               UnicodeEncodeError: 'latin-1' codec can't encode characters in position 0-119: ordinal not in range(256)

Could you take another look? Let us know if it doesn't work and we can help.

[Borda]

@awaelchli, what is the situation here? :)

[gitguardian]

⚠️ GitGuardian has uncovered 2 secrets following the scan of your pull request.

Please consider investigating the findings and remediating the incidents. Failure to do so may lead to compromising the associated services or software components.

🔎 Detected hardcoded secrets in your pull request

GitGuardian id	Secret	Commit	Filename
-	Generic High Entropy Secret	78fa3af	tests/tests_app/utilities/test_login.py	View secret
-	Base64 Basic Authentication	78fa3af	tests/tests_app/utilities/test_login.py	View secret

🛠 Guidelines to remediate hardcoded secrets

1. Understand the implications of revoking this secret by investigating where it is used in your code.
2. Replace and store your secret safely. Learn here the best practices.
3. Revoke and rotate this secret.
4. If possible, rewrite git history. Rewriting git history is not a trivial act. You might completely break other contributing developers' workflow and you risk accidentally deleting legitimate data.

To avoid such incidents in the future consider

• following these best practices for managing and storing secrets including API keys and other credentials
• install secret detection on pre-commit to catch secret before it leaves your machine and ease remediation.

---

<sup>🦉 GitGuardian detects secrets in your source code to help developers and security teams secure the modern development process. You are seeing this because you or someone else with access to this repository has authorized GitGuardian to scan your pull request.

Our GitHub checks need improvements? Share your feedbacks!</sup>