The standard dict class in Python 2 supports a method has_key(key). That method is not supported in Python 3.

  has_key(...)
      D.has_key(k) -> True if D has a key k, else False

Implement the method when on Python 2.

We are currently using the "classic" Sphinx theme. This theme is clearly associated with Python 2.

Since Python 2 went EOL, we should change to use the Python 3 standard theme ("python-docs-theme" package on Pypi).

I am comparing various implementations of case-insensitive dictionaries in Python, and have found that this one uses str.lower() rather than str.casefold(). See e.g. pydicti, which uses casefold() is available, and falls back to lower() if not available. If both fail, the key is used verbatim.

The advantage of casefold is that is has better unicode support than lower.

I'm packaging your module as an rpm package so I'm using the typical PEP517 based build, install and test cycle used on building packages from non-root account.

• python3 -sBm build -w --no-isolation
• because I'm calling build with --no-isolation I'm using during all processes only locally installed modules
• install .whl file in </install/prefix> using installer module
• run pytest with $PYTHONPATH pointing to sitearch and sitelib inside </install/prefix>
• build is performed in env which is cut off from access to the public network (pytest is executed with -m "not network")

• Python version: 3.9.18
• Operating System (type+version): Linux x86/64
• pytest 8.1.1

Please let me know if you need more details or want me to perform some diagnostics.

Neither the latest or stable versions of the RTD created docs include the iter..() and view..() methods of NocaseDict.
I triggered a new build for both stable and latest, so that is not the root cause.

A local make builddoc does include them.

The sys.getsizeof(obj) function returns the memory size of obj in Bytes. It does that by calling __sizeof__() on the object and adding the GC overhead if the object is GC-managed.

The rules for whether a user-defined class like NocaseDict has to implement __sizeof__() are not documented in the Python docs.

Here is a comparison between dict and NocaseDict that suggests that not implementing __sizeof__() is incorrect. On the other hand, dictionaries are referencing their key and value objects, so if multiple dictionaries reference the same objects it would not make too much sense to attribute their sizes to the dictionaries.

import sys
from nocasedict import NocaseDict
nd = NocaseDict()
d = dict()
print("len dict NocaseDict")
for x in range(0, 25):
    print(x, sys.getsizeof(d), sys.getsizeof(nd))
    key = 'key' + str(x)
    nd[key] = x
    d[key] = x

resulting in:

len dict NocaseDict
0 232 48
1 232 48
2 232 48
3 232 48
4 232 48
5 232 48
6 360 48
7 360 48
8 360 48
9 360 48
10 360 48
11 640 48
12 640 48
13 640 48
14 640 48
15 640 48
16 640 48
17 640 48
18 640 48
19 640 48
20 640 48
21 640 48
22 1176 48
23 1176 48
24 1176 48
25 1176 48

Actual behavior

Testcases that specify no expected warning tolerate warnings that occur nevertheless.

Expected behavior

Testcases that specify no expected warning should verify that no warning occurs and fail otherwise.

Execution environment

• nocasedict version: 1.0.0
• Python version: and
• Operating System (type+version): any

In order to better be able to fix yet unknown issues and in case the use of NocaseDict in pywbem reveals a need for changes, this first release should be done by going back to version number 0.5.0 and dev status "5 - Beta", I think.

DISCUSSION

From test run https://github.com/pywbem/nocasedict/runs/5707324455?check_suite_focus=true#step:13:15

************* Module setup
setup.py:12:0: W0402: Uses of a deprecated module 'distutils' (deprecated-module)

README:

• Add examples for hashable and keyableby.
• Add a statement about compatibility to standard dict.

Docs:

• Add example for case-insensitive access to docstring of NocaseDict class.
• Improve wording "preserved order".
• Add a statement about compatibility to standard dict.

Currently, NocaseDict is derived from object, but it should ideally be derived from dict. The issue is that it does not want to inherit the dict data, at least not the current implementation. This can probably be handled though with some changes in the implementation.

DISCUSSION: Do we want NocaseDict to be derived from dict, for type checking?

COMMENT/KS: It is confusing when it is not inherited and probably not very pythonic. Thought it over last night and we should not be calling it a dict if it does not inherit from dict.

The current implementation raises TypeError for these comparisons. That is consistent with dict, and also the message is consistent.

However, there are still TODOs asking for decision on whether to support the comparisons.
These TODOs should be removed.

See Actions run: https://github.com/pywbem/nocasedict/runs/7646491265?check_suite_focus=true

Opened issue: PyCQA/flake8#1650

Probably need to circumvent the issue by temporarily pinning flake8 to <5.0

Clean up copyright and authorship in ./nocasedict/_nocasedict.py

Packaging request: https://bugzilla.redhat.com/show_bug.cgi?id=1881285

Verify comnpleteness of methods relative to standard dict.

In order to verify that the error handling of NocaseDict is compatible to the standard dict, support could be added to run the NocaseDict tests against a standard dict.

Actual behavior

The UserWarning that is issued to indicate that the order of items in input to the NocaseDict initialization or to its update() method is not preserved has two issues:

1. The source code location is incorrect when issued as a result of NocaseDict object initialization: It does not reference the user code, but the NocaseDict init method.

2. When the input is a standard dict, the warning is issued even when that dict contains only one item.

Expected behavior

1. Correct location in user code.

2. Warning is issued only when the dict contains more than one item.

Execution environment

• nocasedict version: 1.0.0
• Python version: any
• Operating System (type+version): any

The current NocaseDict class has an attribute allow_unnamed_keys that controls whether it allows None as a key. By default, it does not allow Noneas a key.

The standard Python dict always allows None as a key.

Pywbem uses this to disallows unnamed keys in most cases, except for CIMInstanceName.keybindings where they are allowed.

Proposal: Remove the allow_unnamed_keys attribute from NocaseDict and always allow None as a key. The places in pywbem where unnamed keys should be disallowed need to be changed to check for that, instead of delegating the check to NocaseDict.

DISCUSSION

The tests against the standard dict fail on Python 3.7 when trying to reverse it because that was introduced only in py38 but the test assumes it was introduced in py37 along with ordering.

PR #79 (targeted for 1.0.3) disabled pylint issue R0801 temporarily.

The circumvention should be transitioned into a final solution once pylint issue pylint-dev/pylint#4118 is addressed.

The standard dict class in Python 2 supports a method fromkeys(). That method is not supported in Python 3.

  fromkeys(...)
      dict.fromkeys(S[,v]) -> New dict with keys from S and values equal to v.
      v defaults to None.

Implement the method when on Python 2.

The NocaseDict class currently raises TypeError if the dict keys do not have a lower() method. The nocaselist.NocaseList class raises AttributeError in that case, which is the better solution because in some cases TypeError is also raised for other issues, and this allows distinguishing the cases better.

Change to raise AttributeError instead, for missing lower() method on key objects.

Actual behavior

The installation of packages is done with pip install --upgrade. That does not upgrade already installed versions of indirect dependents. The Ubuntu used by Travis has quite a number of packages preinstalled in the virtualenv it provides, so these packages stay at versions that are not the latest, if they satisfy the minimum requirements of the other packages.

Expected behavior

Testing with latest package levels should upgrade all packages including indirect dependencies.

This can be achieved by adding the pip option --upgrade-strategy eager.

Execution environment

• nocasedict version: 1.0.1
• Python version: any
• Operating System (type+version): any

Actual behavior

Creating a NocaseDict with byte-string keys generates an error since #122 was merged (version 1.0.5):

from nocasedict import NocaseDict
NocaseDict([(b'MyKey', b'myValue')])

>>> AttributeError: 'bytes' object has no attribute 'casefold'

Expected behavior

Byte-string keys worked fine in versions 1.0.4 and lower due to use of .lower() (which exists on both byte-strings and unicode strings). Either the SemVer should be bumped, and the API changed to generate an error for byte-string keys, or a compatibility wrapper should be added.

Execution environment

• nocasedict version: 1.1.0
• Python version: 3.11
• Operating System (type+version): Linux

Actual behavior

See Travis run: https://travis-ci.org/github/pywbem/nocasedict/jobs/732205320

Testcase test1: Pip install from repo root directory: /home/travis/build/pywbem/nocasedict
Saving location of current virtualenv: /home/travis/virtualenv/python3.4.8
Before creating virtualenv: /home/travis/build/pywbem/nocasedict/tmp_installtest/virtualenvs/nocasedict_test_test1
Python version: Python 3.4.8 from /home/travis/virtualenv/python3.4.8/bin/python
Pip version: pip 19.1.1 from /home/travis/virtualenv/python3.4.8/lib/python3.4/site-packages/pip (python 3.4)
Creating virtualenv: /home/travis/build/pywbem/nocasedict/tmp_installtest/virtualenvs/nocasedict_test_test1
Error: Command failed with rc=1: virtualenv -p /home/travis/virtualenv/python3.4.8/bin/python  /home/travis/build/pywbem/nocasedict/tmp_installtest/virtualenvs/nocasedict_test_test1, output follows:
AttributeError: 'Path' object has no attribute 'write_bytes'
Makefile:585: recipe for target 'installtest' failed

Expected behavior

Success

Execution environment

• nocasedict version: 1.0.0
• Python version: 3.4
• Operating System (type+version): Ubuntu (Travis)

See TheKevJames/coveralls-python#252.
A possible mitigation for now is to pin coveralls to <3.0.0.

The current implementation of keys() etc. and iter..() is consistent with Python 2, but misses the view..() methods added in Python 2.7.

In Python 3, the iter..() and view..() methods were dropped and the keys() etc. methods return a view.

Proposal: Implement the behavior of the standard dict class for each of the major Python versions. If a common behavior across the major Python versions is needed, users can use the six package.

The current NocaseDict init method supports an iterable of pywbem CIM objects whose name attribute is used as dict key. That ability is very convenient for pywbem and needs to be retained.

Options are:

• NocaseDict continues to support the functionality, but in a cleaned up way. It would be an additional functionality on top of what the standard Python dict supports.
• NocaseDict no longer supports it, and the pywbem code handles it. That is possible because NocaseDict objects so far were not created by pywbem users.

COMMENT/ks: Actually it is used in pywbemcli but only in tests. The tests can be changed.; it is used to generate keybindings for test_instances.py and scopes for test_qualdecl.py

DISCUSSION

The current NocaseDict code supports a __hash__() method that calculates a hash value from the set of tuples of lower cased dict key and dict item value. That makes it a hashable object. Hashable objects can be used as members in sets or as keys in mappings (dicts).

NocaseDict objects are mutable, but the hash value is calculated under the assumption that the object does not change its value while used in a set or as a key, i.e. it is calculated just once for a particular object. For this reason, the mutable standard types in Python (e.g. dict) are not hashable. Changing the value while in a set or used as a key can have strange effects (there are forum threads full of that).

Right now, we document that the NocaseDict objects that are used in a set or as a mapping key must not change while being used that way. However, there is no enforcement about that.

Pywbem is using NocaseDict objects in sets (I believe, need to double check).

DISCUSSION: Should we continue supporting NocaseDict objects being hashable, or should we remove that functionality because it is considered too dangerous for the general user.

It would be nice to have stronger typing when using the package. Currently, Mypy does not find any types when importing from nocasedict:

A good first step would be to add the py.typed marker, and ensure that all function arguments and return values have type hints. The project could benefit from Mypy linting itself, but that is not required.

After switching to the Python 3 theme for the docs, there are some minor issues with it:

• It shows an empty "Copyright" text at the bottom right end of each page.
• It shows a link "Python" to www.python.org at the top left and bottom left end of each page.
• There is a reddish background color on some areas, e.g. on the Parameter or Raises sections of method/function descriptions.

Starting with Python 3.7, the standard dict is guaranteed to be ordered, so the use of OrderedDict can be replaced by the standard dict on py>=3.7.