puhoshville / keepthis Goto Github PK

Is your feature request related to a problem? Please describe.
While working with keepthis for long time you may be interested in deleting old data in store. but we can see only unique hash of data without any information about person how published that data.

Describe the solution you'd like
While setting keepthis instance at start add option to provide username. use systems username as default value.

Describe alternatives you've considered
no

Additional context
this information will be observable from memcached gui client with option to delete some cache

Dependabot couldn't authenticate with https://pypi.python.org/simple/.

You can provide authentication details in your Dependabot dashboard by clicking into the account menu (in the top right) and selecting 'Config variables'.

View the update logs.

Remove requests_count attribute from KeepThis class.

Very good example - https://github.com/Delgan/loguru
We want to make it as simple as possible.
I see that will be enough to initialize KeepThis instance with memcached_host="localhost" and memcached_port=11211 by deafault. Or use ENV variables if those provided.

At the moment project support only json serializable objects. This is huge restriction for data scientists as numpy arrays and pandas dataframes are commonly used.

• numpy ndarray support in args
• pandas dataframe support in args
• numpy ndarray support in kwargs
• pandas dataframe support in kwargs

Describe the bug
No requirements installing on pip install keepthis

To Reproduce
Steps to reproduce the behavior:

1. pip install keepthis
2. In python terminal: import keepthis
3. See the error

>>> import keepthis
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/puhoshville/Documents/private/repos/venv/lib/python3.9/site-packages/keepthis/__init__.py", line 1, in <module>
    from keepthis.KeepThis import KeepThis
  File "/Users/puhoshville/Documents/private/repos/venv/lib/python3.9/site-packages/keepthis/KeepThis.py", line 4, in <module>
    import numpy as np
ModuleNotFoundError: No module named 'numpy'

Expected behavior

Can import package after installation.

windows-2016, python 3.5 build failed

================================== FAILURES ===================================
44
______________________________ test_hash_ndarray ______________________________
45

46
    def test_hash_ndarray():
47
        array = np.array([1, 2, 3, 4])
48
        result_hash = KeepThis._hash_ndarray(array)
49
        assert isinstance(result_hash, str)
50
>       assert result_hash == '46c4f0c1fb94a6327fafea6bb1ddf0dd4ddb09f77142e1afae176f96'
51
E       AssertionError: assert 'ebea61ffae91...9c659d6b09170' == '46c4f0c1fb94...2e1afae176f96'
52
E         - ebea61ffae916052811e215e5876b886429481d7c7a9c659d6b09170
53
E         + 46c4f0c1fb94a6327fafea6bb1ddf0dd4ddb09f77142e1afae176f96
54

55
tests\test_KeepThis.py:57: AssertionError
56
========================= 1 failed, 5 passed in 1.14s =========================
57

More info about failed pipeline: https://github.com/puhoshville/keep-this/runs/242493727

Why we got different from unix and macos hash for same object? And how can we avoid such problems?

Let context manager to handle connections to memcached. It will simplify logic and will be more "pythonic"

Example broken_pipe.py is not very clear for reader. We can describe this case easily by using records from desktop (in GIF)