Merge pull request #30 from cclauss/patch-2

Update README.rst to recommend pipx run or install

.github/workflows/test.yml

@@ -13,6 +13,16 @@ jobs:
       matrix:
         python-version: ["3.6", "3.7.7", "3.8", "3.9", "3.10", "3.11"]
         os: [ubuntu-20.04, macos-latest, windows-latest]
+        exclude:  # Apple Silicon ARM64 does not support Python < v3.8
+          - python-version: "3.6"
+            os: macos-latest
+          - python-version: "3.7.7"
+            os: macos-latest
+        include:  # So run those legacy versions on Intel CPUs
+          - python-version: "3.6"
+            os: macos-13
+          - python-version: "3.7.7"
+            os: macos-13
 
     steps:
       - uses: actions/checkout@v4

README.rst

@@ -12,7 +12,7 @@ Overview
 
 This package provides a Python script and pytest plugin to help convert Nose-based tests into pytest-based
 tests. Specifically, the script transforms ``nose.tools.assert_*`` function calls into raw assert statements, 
-while preserving format of original arguments as much as possible. For example, the script:
+while preserving the format of original arguments as much as possible. For example, the script:
 
 .. code-block:: python
 
@@ -31,35 +31,36 @@ transformed because there is no raw assert statement equivalent, or the equivale
 maintain. They are provided as functions in the pytest namespace via pytest's plugin system.
 
 
-Installation
--------------
-
-From a command shell run ::
-
-  pip install nose2pytest
-
-This puts an executable file in ``<python-root>/Scripts`` with *python-root* being the root folder of the 
-Python installation from which ``pip`` was run.
-
-
 Running
 ------------
 
-From a command shell, ::
+For a one-time conversion use the shell command ::
 
-  nose2pytest path/to/dir/with/python_files
+  pipx run --python 3.11 nose2pytest path/to/dir/with/python_files
   
 This will find all ``.py`` files in the folder tree starting at ``path/to/dir/with/python_files`` and 
 overwrite the original (assuming most users will be running this on a version-controlled code base, this is
 almost always what would be most convenient). Type ``nose2pytest -h`` for other options, such as ``-v``. 
 
 
+Installation
+-------------
+
+For doing multiple conversions use the shell command ::
+
+  pipx install --python 3.11 nose2pytest
+
+For each conversion use the shell command ::
+
+  nose2pytest path/to/dir/with/python_files
+
+
 Motivation
 ------------
 
 I have used Nose for years and it is a great tool. However, to get good test failure diagnostics with Nose you 
 ought to use the ``assert_*()`` functions from ``nose.tools``. Although they provide very good diagnostics, they 
-are not as convenient to use as raw assertions, since you have to decide before hand what type of assertion you 
+are not as convenient to use as raw assertions, since you have to decide beforehand what type of assertion you 
 are going to write: an identity comparison to None, a truth check, a falseness check, an identity comparison to another 
 object, etc. Just being able to write a raw assertion, and still get good diagnostics on failure as done by 
 pytest, is really nice. This is a main reason for using pytest for me. Another reason is the design of fixtures
@@ -72,7 +73,7 @@ manually, to get the same test coverage and results. A few gotchas:
 - test classes that have ``__init__`` will be ignored, those will have to be moved (usually, into class's 
   ``setup_class()``)
 - the ``setup.cfg`` may have to be edited since test discovery rules are slightly more strict with pytest
-- the order of tests may be different, but in general that should not matter
+- the order of tests may be different, but in general, that should not matter
 - all test modules are imported up-front, so some test modules may need adjustment such as moving some 
   code from the top of the test module into its ``setup_module()`` 
 
@@ -83,7 +84,7 @@ uses ``nose.tools.assert_*`` functions, yet with pytest you can use raw assertio
 these two approaches should a developer use? If you modify existing tests, should new assertions use raw assert? 
 Should the remaining test method, test class, or test module be updated? A test module can contain hundreds of 
 calls to ``nose.tools.assert_*`` functions, is a developer to manually go through each one to convert it? Painful and 
-error prone, in general not feasible to do manually. 
+error-prone, in general not feasible to do manually. 
 
 This is why I developed nose2pytest: I wanted to migrate my pypubsub project's test suite from Nose to pytest,
 but also have only pytest as a dependency, and have one obvious way to write assertions in the test suite.
@@ -92,14 +93,9 @@ but also have only pytest as a dependency, and have one obvious way to write ass
 Requirements
 -------------
 
-I expect nose2pytest script to run with Python >= 3.4, to correctly convert Python test suite >= 2.7, on any 
-OS supported by a version of python that has lib2to3 compatible with Python 3.4's lib2to3. I expect it to 
-succeed even with quite old versions of Nose (even prior to 1.0 which came out ca. 2010), and with the new 
-Nose2 test driver. 
-
-Note however that I have run the script only with Python 3.4, to convert Python 3.4 test suites based on 
-Nose 1.3.7 on Windows 7 Pro 64. If you have successfully used nose2pytest with other combinations, please 
-kindly let me know (via github). 
+I expect nose2pytest script to run with supported versions of CPython <= v3.11, on any OS supported by a version of
+Python that has lib2to3 compatible with fissix. I expect it to succeed even with quite old versions of Nose (even
+prior to 1.0 which came out ca. 2010) and with the new Nose2 test driver. 
 
 The pytest package namespace will be extended with ``assert_`` functions that are not converted by the script
 only if, err, you have pytest installed!
@@ -207,7 +203,7 @@ Limitations
 
 - The script does not convert ``nose.tools.assert_`` import statements as there are too many possibilities. 
   Should ``from nose.tools import ...`` be changed to ``from pytest import ...``, and the implemented 
-  conversions removed? Should an ``import pytest`` statement be added, and if so, where? If it is added after
+  conversions be removed? Should an ``import pytest`` statement be added, and if so, where? If it is added after
   the line that had the ``nose.tools`` import, is the previous line really needed? Indeed the ``assert_``
   functions added in the ``pytest`` namespace could be accessed via ``pytest.assert_``, in which case the 
   script should prepend ``pytest.`` and remove the ``from nose.tools import ...`` entirely. Too many options, 
@@ -229,30 +225,17 @@ Limitations
     import bogo.assert_true
     bogo.assert_true(...)  # should this one be converted? 
   
-  The possiblities are endless so supporting this would require such a large amount of time that I 
+  The possibilities are endless so supporting this would require such a large amount of time that I 
   do not have. As with other limitations in this section
 
 - Nose functions that can be used as context managers can obviously not be converted to raw assertions. 
   However, there is currently no way of preventing nose2pytest from converting Nose functions used this way. 
   You will have to manually fix.
-
-- The lib2to3 package that nose2pytest relies on assumes python 2.7 syntax as input. The only issue that 
-  this has caused so far on code base of 20k lines of python 3.4 *test* code (i.e. the source code does not 
-  matter, as none of the test code, such as import statements, is actually run) are keywords like ``exec`` 
-  and ``print``, which in Python 2.x were statements, whereas they are functions in Python 3.x. This means 
-  that in Python 3.x, a method can be named ``exec()`` or ``print()``, whereas this would lead to a syntax
-  error in Python 2.7. Some libraries that do not support 2.x take advantage of this (like PyQt5). Any 
-  occurrence of these two keywords as methods in your test code will cause the script to fail converting 
-  anything. 
-
-  The work around is, luckily, simple: do a global search-replace of ``\.exec\(`` for ``.exec__(`` in your 
-  test folder, run nose2pytest, then reverse the search-replace (do a global search-replace of ``\.exec__\(`` 
-  for ``.exec(``).
-
+
 - ``@raises``: this decorator can be replaced via the regular expression ``@raises\((.*)\)`` to 
   ``@pytest.mark.xfail(raises=$1)``,
   but I prefer instead to convert such decorated test functions to use ``pytest.raises`` in the test function body.
-  Indeed, it is easy to forget the decorator, and add code after the line that raises, but this code will never 
+  Indeed, it is easy to forget the decorator and add code after the line that raises, but this code will never 
   be run and you won't know. Using the ``pytest.raises(...)`` is better than ``xfail(raise=...)``. 
 
 - Nose2pytest does not have a means of determining if an assertion function is inside a lambda expression, so
@@ -281,16 +264,17 @@ should be able to run both a unittest2pytest converter, then the nose2pytest con
 Solution Notes
 ---------------
 
-I don't think this script would have been possible without lib2to3, certainly not with the same functionality since 
-lib2to3, due to its purpose, preserves newlines, spaces and comments. The documentation for lib2to3 is very 
-minimal, so I was lucky to find http://python3porting.com/fixers.html.
+I don't think this script would have been possible without lib2to3/fissix, certainly not with the same
+functionality since lib2to3/fissix, due to their purpose, preserves newlines, spaces and comments. The
+documentation for lib2to3/fissix is very minimal, so I was lucky to
+find http://python3porting.com/fixers.html.
 
-Other than figuring out lib2to3 package so I could harness its 
-capabilities, some aspects of code transformations still turned out to be tricky, as warned by Regobro in the 
-last paragraph of his  `Extending 2to3 <http://python3porting.com/fixers.html>`_ page. 
+Other than figuring out lib2to3/fissix package so I could harness its capabilities, some aspects of code
+transformations still turned out to be tricky, as warned by Regobro in the last paragraph of his
+`Extending 2to3 <http://python3porting.com/fixers.html>`_ page. 
 
 - Multi-line arguments: Python accepts multi-line expressions when they are surrounded by parentheses, brackets 
-  or braces, but not otherwise. For example converting:
+  or braces, but not otherwise. For example, converting:
 
   .. code-block:: python
 
@@ -313,7 +297,7 @@ last paragraph of his  `Extending 2to3 <http://python3porting.com/fixers.html>`_
 
   So nose2pytest checks each argument expression (such as ``long_a +\n long_b``) to see if it has 
   newlines that would cause an invalid syntax, and if so, wraps them in parentheses. However, it is also important
-  for readability of raw assertions that parentheses only be present if necessary. In other words:
+  for the readability of raw assertions that parentheses only be present if necessary. In other words:
 
   .. code-block:: python
 
@@ -342,10 +326,10 @@ last paragraph of his  `Extending 2to3 <http://python3porting.com/fixers.html>`_
 
   So nose2pytest only tries to limit the addition of external parentheses to code that really needs it. 
 
-- Operator precedence: Python assigns a precedence to each operator; operators that are on the same level
+- Operator precedence: Python assigns precedence to each operator; operators that are on the same level
   of precedence (like the comparison operators ==, >=, !=, etc) are executed in sequence. This poses a problem 
   for two-argument assertion functions. Example: translating ``assert_equal(a != b, a <= c)`` to 
-  ``assert a != b == a <= c`` is incorrect, it must be converted to ``assert (a != b) == (a <= c)``. However
+  ``assert a != b == a <= c`` is incorrect, it must be converted to ``assert (a != b) == (a <= c)``. However,
   wrapping every argument in parentheses all the time does not produce easy-to-read assertions:
   ``assert_equal(a, b < c)`` should convert to ``assert a == (b < c)``, not ``assert (a) == (b < c)``. 
 
@@ -358,20 +342,20 @@ last paragraph of his  `Extending 2to3 <http://python3porting.com/fixers.html>`_
 Contributing
 ------------
 
-Patches and extensions are welcome. Please fork, branch, then submit PR. Nose2pytest uses `lib2to3.pytree`,
+Patches and extensions are welcome. Please fork, branch, and then submit PR. Nose2pytest uses `lib2to3.pytree`,
 in particular the Leaf and Node classes. There are a few particularly challenging aspects to transforming
 nose test expressions to equivalent pytest expressions:
 
 #. Finding expressions that match a pattern: If the code you want to transform does not already match one
-   of the uses cases in script.py, you will have to determine the lib2to3 pattern expression
+   of the uses cases in script.py, you will have to determine the lib2to3/fissix pattern expression
    that describes it (this is similar to regular expressions, but for AST representation of code,
    instead of text strings). Various expression patterns already exist near the top of
    nose2pytest/script.py. This is largely trial and error as there is (as of this writing) no good
    documentation.
-#. Inserting the sub-expressions extracted by lib2to3 in step 1 into the target "expression template". For
-   example to convert `assert_none(a)` to `assert a is None`, the `a` sub-expression extracted via the lib2to3
-   pattern must be inserted into the correct "placeholder" node of the target expression. If step 1 was
-   necessary, then step 2 like involves creating a new class that derives from `FixAssertBase`.
+#. Inserting the sub-expressions extracted by lib2to3/fissix in step 1 into the target "expression template".
+   For example to convert `assert_none(a)` to `assert a is None`, the `a` sub-expression extracted via the
+   lib2to3/fissix pattern must be inserted into the correct "placeholder" node of the target expression. If
+   step 1 was necessary, then step 2 like involves creating a new class that derives from `FixAssertBase`.
 #. Parentheses and priority of operators: sometimes, it is necessary to add parentheses around an extracted
    subexpression to protect it against higher-priority operators. For example, in `assert_none(a)` the `a`
    could be an arbitrary Python expression, such as `var1 and var2`. The meaning of `assert_none(var1 and var2)`
@@ -431,10 +415,10 @@ Maintenance
 
 - Clone or fork the git repo, create a branch
 - Install `pytest` and `nose` on your system: `python -m pip install pytest nose`
-- In root folder, run `pytest`
+- In the root folder, run `pytest`
 - Once all tests pass, install tox on your system: on Ubuntu, `python -m pip install tox`
 - Run tox: `tox`
-- Add a python version if latest python is not in `tox.ini`
+- Add a python version if the latest Python is not in `tox.ini`
 
 Notes for Ubuntu:
 - My experience today installing python 3.5 to 3.11 on Ubuntu 18 was surprisingly not smooth. I had to use these
@@ -448,8 +432,8 @@ Notes for Ubuntu:
   - note however that once the correct tox installed,
 
 
-Acknowledgements
-----------------
+Acknowledgments
+---------------
 
 Thanks to (AFAICT) Lennart Regebro for having written http://python3porting.com/fixers.html#find-pattern, and 
 to those who answered