Release 6.0.0 to allow users to update to Python 3.12?

[hartwork]

Hi! 👋

(This is similar to issue #749 for the release of VCR.py 5.1.0.)

Since release 5.1.0 I believe we have merged the following pull requests to master (in order of pull request ID):

• Drop boto 2 support #731
• Drop simplejson support (fixes #751) #752
• Enable testing on Python 3.12 & PyPy 3.9/3.10 #754
• Configure Read the Docs to install the library #757
• build(deps): bump sphinx-rtd-theme from 1.2.2 to 1.3.0 #765
• build(deps): bump actions/checkout from 3 to 4 #766
• docs: Fix list formatting in changelog for v5.1.0 #771
• Finish up on Python 3.12 support #775
• Fix HTTPX support #784
• Fix CI / Block urllib3 >=2 for PyPy (alternative to #786) #787
• Fix: Httpx Human Readable Responses #788
• tests: Fix imports to tests/assertions.py (fixes #773) #789
• Use ruff to check code format intead of black #790
• Migrate linting from a tox environment to a GitHub Actions action #791
• Run tests without tox #794
• build(deps): bump actions/setup-python from 4 to 5 #795
• build(deps): bump sphinx-rtd-theme from 1.3.0 to 2.0.0 #797
• Show ruff diff errors #799
• Start using pre-commit in CI #800
• Fix ResourceWarning unclosed socket and re-enable and fix tornado tests #801

I believe it would be important to do a soon new release 6.0.0 to allow users to upgrade to Python 3.12 which VCR.py has been de-facto blocking (including one of my own projects, case hartwork/wnpp.debian.net#806).

For anyone wondering why 6.0.0 rather than 5.2.0, we agreed at #731 (comment) that merge of now-merged #731 should go along with a bump in major version.

@kevin1024 would you be up for cutting a release 6.0.0 from latest master shortly?
(Latest commit is 6a31904 as of the moment.)

Thanks and best, Sebastian

CC @jairhenrique

[jairhenrique]

This pr needs to be merged before this version #788

[hartwork]

@jairhenrique just to understand: "needs" as in "would be beneficial" or as in "hard blocker"? Could you elaborate?

[jairhenrique]

@hartwork I think it's a hard blocker because of this comment #784 (comment). More people may need to recreate the cassettes if this change doesn't go through.

[hartwork]

@jairhenrique thanks for elaborating, I understand! 👍

@parkerhancock if you could rebase your pull request #788 against latest master that would rock and greatly increase the chances of a soon release 6.0.0. Thanks in advance! 🙏

[hartwork]

@kevin1024 I believe all release blockers are resolved now.

CC @jairhenrique

[anton-daneyko-ultramarin]

What is the highest Python version that you recommend to use ATM?

[hartwork]

What is the highest Python version that you recommend to use ATM?

@anton-daneyko-ultramarin probably 3.11? Could you elaborate or give more context for the question?

[anton-daneyko-ultramarin]

@hartwork
Some context: I do not get any errors running my test suite if I use Python 3.12, however I do get some errors when I try to use PyCharm debugger with my test suite, so 3.12 is not very convenient for me for this reason (I appreciate the debugger). So I bumped it down to Python 3.11.6 and started to get some TypeError: cannot pickle 'generator' object errors. This prompted me to go to VCR.py issues page and read this thread from which I assume there are some problems with Python 3.12 (which I am not triggering with my test suite).

I am about to give 3.10 a try, but decided to ask you instead. I assumed it's something easy to answer.

[hartwork]

@anton-daneyko-ultramarin thanks for elaborating! Could you check if the pickle issue is the same as #737? If not, please create a new issue with details. Thanks in advance!

[vEpiphyte]

I'm running a big regression test with the latest vcrpy development head. I'll update this comment when it is done ( likely in the morning EST ) if i run across any issues :)

Update: 👍 My big regression test did not uncover any sort of issues with the current head 713cb36d35108a21416744a393f3c947b86a0efe

[dmelo]

Any update on when we'll have the new release?

[hartwork]

@kevin1024 we need you here to move forward. Do you have a minute?

[eugene-s]

Hey guys!

Are there any plans to release that stuff soon? I need fixes for httpx.

[hartwork]

@kevin1024 we need you here to move forward. Do you have a minute?

[graingert]

I'd like to get #801 in before 6.0

[kevin1024]

OK! lemme cut a release.

[graingert]

Wait I broke the tornado generator decorator in #801 fixed by #811

[kevin1024]

No worries, ping me when fixed and I'll do another release.

[anton-daneyko-ultramarin]

@anton-daneyko-ultramarin thanks for elaborating! Could you check if the pickle issue is the same as #737? If not, please create a new issue with details. Thanks in advance!

I did not have the chance to look into it back then and now I am getting bitten by this error again. I get a TypeError: cannot pickle 'generator' object error when using urllib3 v2.1.0 and the error goes away with urllib3 v1.26.18. The vcrpy version is 5.1.0. The stack trace leads to before_record_request in vcr/config.py. I probably won't be able to dig into it now as well, but I thought it might be helpful to leave some breadcrumbs for a future investigation or maybe @hartwork would be able to spot something fishy immediatly:

Tail of a stack trace

/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/site-packages/requests/adapters.py:486: in send
    resp = conn.urlopen(
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/site-packages/urllib3/connectionpool.py:790: in urlopen
    response = self._make_request(
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/site-packages/urllib3/connectionpool.py:536: in _make_request
    response = conn.getresponse()
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/site-packages/vcr/stubs/__init__.py:257: in getresponse
    if self.cassette.can_play_response_for(self._vcr_request):
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/site-packages/vcr/cassette.py:263: in can_play_response_for
    request = self._before_record_request(request)
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/site-packages/vcr/config.py:223: in before_record_request
    request = copy.deepcopy(request)
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/copy.py:172: in deepcopy
    y = _reconstruct(x, memo, *rv)
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/copy.py:271: in _reconstruct
    state = deepcopy(state, memo)
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/copy.py:146: in deepcopy
    y = copier(x, memo)
/home/acme-corp/.conda/envs/apa-proxy/lib/python3.10/copy.py:231: in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
x = <generator object Client.upload_data.<locals>.json_stream at 0x7f792ac86030>
memo = {140158380145104: <[AttributeError("'Request' object has no attribute 'method'") raised in repr()] Request object at 0...', 'uri': 'http://localhost:8080/websrv/portfolios/test_cancel_task/dates?enforce-partial-import=false'}}
_nil = []
    def deepcopy(x, memo=None, _nil=[]):
        """Deep copy operation on arbitrary Python objects.
    
        See the module's __doc__ string for more info.
        """
    
        if memo is None:
            memo = {}
    
        d = id(x)
        y = memo.get(d, _nil)
        if y is not _nil:
            return y
    
        cls = type(x)
    
        copier = _deepcopy_dispatch.get(cls)
        if copier is not None:
            y = copier(x, memo)
        else:
            if issubclass(cls, type):
                y = _deepcopy_atomic(x, memo)
            else:
                copier = getattr(x, "__deepcopy__", None)
                if copier is not None:
                    y = copier(memo)
                else:
                    reductor = dispatch_table.get(cls)
                    if reductor:
                        rv = reductor(x)
                    else:
                        reductor = getattr(x, "__reduce_ex__", None)
                        if reductor is not None:
>                           rv = reductor(4)
E                           TypeError: cannot pickle 'generator' object

[hartwork]

@anton-daneyko-ultramarin let me repeat, please create a new issue, your report is out of place here. Thank you!