add setuptools files for example package mentioned in Issue 248 by sosey · Pull Request #659 · pyOpenSci/python-package-guide

sosey · 2026-05-18T21:26:57Z

This adds example package files for setuptools, as mentioned in #248.
The examplePy package builds and tests pass with pytest.

willingc

Thanks @sosey for the contribution.

Tagging Leah for a final review 😄

lwasser · 2026-05-28T17:20:51Z

This looks great @sosey i'm going to see if @ucodery has time to review only because he started the package examples but if not I will review soon! Good to see you again at our sprints! i'm sorry we didn't' get to chat during the day.

sneakers-the-rat

looks good! with some minor comments and questions. i'm not sure of the context for this example or what the surrounding info is, so if my questions are covered by other material elsewhere already, feel free to ignore!

sneakers-the-rat · 2026-05-28T21:38:09Z

+
+def calc_annual_mean(df: pandas.DataFrame):
+    """Function to calculate the mean temperature for each year and the final mean"""
+    # TODO: make this a bit more robust so we can write integration test examples??


Suggested change

# TODO: make this a bit more robust so we can write integration test examples??

do we need this todo?

sneakers-the-rat · 2026-05-28T21:39:26Z

hyphen in filename, wouldn't import, looks like a duplicate of temporal.py? can this be removed?

I did this for some other example code and the hyphen was intentional. In sphinx we can use diff-file blocks to show, well diffs of how to modify code to fit best practice, but that involves having both files present in the repo. Also, we want these example files linted and tested to make sure they are high-quality, however the -raw files are often intentionally bad style.

Not sure if that is that is what is going on here, as I don't see any documentation references to this file.

Ah, OK. Didn't know that, thanks for the context. I didn't diff them but they look the same? Maybe a leftover from another example or something.

I tried to capture the purpose of examples/ in https://github.com/pyOpenSci/python-package-guide/blob/main/CONTRIBUTING.md#annex, although it looks like I failed to mention anything about diffing...

sneakers-the-rat · 2026-05-28T21:39:38Z

+
+def calc_annual_mean(df: pandas.DataFrame):
+    """Function to calculate the mean temperature for each year and the final mean"""
+    # TODO: make this a bit more robust so we can write integration test examples??


Suggested change

# TODO: make this a bit more robust so we can write integration test examples??

similarly, do we need this todo?

sneakers-the-rat · 2026-05-28T21:47:40Z

+import math
+
+from examplePy import temperature
+
+
+def test_fahrenheit_to_celsius_positive():
+    """Test F to C calculation for positive values"""
+    value = 95
+    expected_result = 35
+    result = temperature.fahrenheit_to_celsius(value)
+
+    assert result == expected_result
+
+
+def test_fahrenheit_to_celsius_negative():
+    """Test F to C calculation for negative values"""
+    value = -13
+    expected_result = -25
+    result = temperature.fahrenheit_to_celsius(value)
+
+    assert result == expected_result
+
+
+def test_fahrenheit_to_celsius_zero():
+    """Test F to C calculation for zero"""
+    value = 0
+    expected_result = -17.7778
+    result = temperature.fahrenheit_to_celsius(value)
+
+    # Test that the result is close to the expected value, within tolerances
+    assert math.isclose(result, expected_result, abs_tol=0.0001)
+
+
+def test_celsius_to_fahrenheit_positive():
+    """Test C to F calculation for positive values"""
+    value = 100
+    expected_result = 212
+    result = temperature.celsius_to_fahrenheit(value)
+
+    assert result == expected_result
+
+
+def test_celsius_to_fahrenheit_negative():
+    """Test C to F calculation for negative values"""
+    value = -20
+    expected_result = -4
+    result = temperature.celsius_to_fahrenheit(value)
+
+    assert result == expected_result
+
+
+def test_celsius_to_fahrenheit_zero():
+    """Test C to F calculation for zero"""
+    value = 0
+    expected_result = 32
+    result = temperature.celsius_to_fahrenheit(value)
+
+    assert result == expected_result


Suggested change

import math

from examplePy import temperature

def test_fahrenheit_to_celsius_positive():

"""Test F to C calculation for positive values"""

value = 95

expected_result = 35

result = temperature.fahrenheit_to_celsius(value)

assert result == expected_result

def test_fahrenheit_to_celsius_negative():

"""Test F to C calculation for negative values"""

value = -13

expected_result = -25

result = temperature.fahrenheit_to_celsius(value)

assert result == expected_result

def test_fahrenheit_to_celsius_zero():

"""Test F to C calculation for zero"""

value = 0

expected_result = -17.7778

result = temperature.fahrenheit_to_celsius(value)

# Test that the result is close to the expected value, within tolerances

assert math.isclose(result, expected_result, abs_tol=0.0001)

def test_celsius_to_fahrenheit_positive():

"""Test C to F calculation for positive values"""

value = 100

expected_result = 212

result = temperature.celsius_to_fahrenheit(value)

assert result == expected_result

def test_celsius_to_fahrenheit_negative():

"""Test C to F calculation for negative values"""

value = -20

expected_result = -4

result = temperature.celsius_to_fahrenheit(value)

assert result == expected_result

def test_celsius_to_fahrenheit_zero():

"""Test C to F calculation for zero"""

value = 0

expected_result = 32

result = temperature.celsius_to_fahrenheit(value)

assert result == expected_result

import math

import pytest

from examplePy import temperature

@pytest.mark.parametrize(

"value,expected",

[

pytest.param(95, 35, id="positive"),

pytest.param(-13, -25, id="negative"),

pytest.param(0, 17.7778, id="zero"),

]

)

def test_fahrenheit_to_celsius(value, expected):

"""Test F to C calculation"""

result = temperature.fahrenheit_to_celsius(value)

assert math.isclose(result, expected, abs_tol=0.0001)

@pytest.mark.parametrize(

"value,expected",

[

pytest.param(100, 212, id="positive"),

pytest.param(-20, -4, id="negative"),

pytest.param(0, 32, id="zero"),

]

)

def test_celsius_to_fahrenheit(value, expected):

"""Test C to F calculation"""

result = temperature.celsius_to_fahrenheit(value)

assert math.isclose(result, expected, abs_tol=0.0001)

just for the sake of demoing how one might write this in a real package/reducing duplication/etc. take or leave, not sure of the context/audience for this example.

sneakers-the-rat · 2026-05-28T21:51:52Z

this is fine, but as an aside on more generally how we structure example tests in guides, i've seen one of the major learning hurdles for learning how to write tests be "how to avoid magic number tests that are actually just testing that some value in a file doesn't change rather than function correctness." super common pattern that people will just throw a huge end-to-end thing in a test function and assert that some value that comes out of it remains a hardcoded value in the test body, rather than something where a known input is coupled to an expected output, and then they struggle with how to add additional cases to protect against regressions/having to tiptoe around the test data file to avoid trivial changes that break e.g. string comparisons.

sneakers-the-rat · 2026-05-28T21:52:34Z

+include-package-data = false
+
+[tool.setuptools_scm]
+write_to = "src/examplePy/_version.py"


maybe some comment on how this is used? like e.g. this might be imported in __init__ to set a __version__ or something?

sneakers-the-rat · 2026-05-28T21:53:05Z

+zip-safe = true
+include-package-data = false


comments on what these do, or links to docs?

sneakers-the-rat

sorry one more thing i am just sort of a big type annotations fan and think it's really good for newcomers to learn because they help you think through your code and let your IDE help you avoid bugs that can be very frustrating. also take or leave.

sneakers-the-rat · 2026-05-28T21:54:59Z

@@ -0,0 +1,26 @@
+def celsius_to_fahrenheit(celsius):


Suggested change

def celsius_to_fahrenheit(celsius):

def celsius_to_fahrenheit(celsius: float) -> float:

sneakers-the-rat · 2026-05-28T21:55:12Z

+    return fahrenheit
+
+
+def fahrenheit_to_celsius(fahrenheit):


Suggested change

def fahrenheit_to_celsius(fahrenheit):

def fahrenheit_to_celsius(fahrenheit: float) -> float:

sneakers-the-rat · 2026-05-28T21:58:45Z

+from examplePy.temperature import fahrenheit_to_celsius
+
+
+def calc_annual_mean(df: pandas.DataFrame):


Suggested change

def calc_annual_mean(df: pandas.DataFrame):

def calc_annual_mean(df: pandas.DataFrame) -> tuple[pandas.DataFrame, pandas.Series]:

sneakers-the-rat · 2026-05-28T21:58:56Z

+from typing import Sequence
+


Suggested change

from typing import Sequence

appears unused?

ucodery

GitHub is having serious issues loading, well issues, right now so I can't read the original motivation for this issue. I did a general pass on it, but will probably come back to it next week too.

ucodery · 2026-05-30T03:27:23Z

+zip-safe = true
+include-package-data = false
+
+[tool.setuptools_scm]


If this project wants to use setuptools-scm then that needs to be in the build-system.requires

ucodery · 2026-05-30T03:27:36Z

+]
+
+
+[build-system]


Why did this section get moved down?

ucodery · 2026-05-30T03:32:31Z

+version_scheme = "release-branch-semver"
+
+
+[project.optional-dependencies]


I personally like and use optional-dependencies, but the new thing to do here is use [dependency-groups](https://packaging.python.org/en/latest/specifications/dependency-groups/) (not sure if you have a preference based on the tutorials @lwasser ?)

ucodery · 2026-05-30T03:36:44Z

I did this for some other example code and the hyphen was intentional. In sphinx we can use diff-file blocks to show, well diffs of how to modify code to fit best practice, but that involves having both files present in the repo. Also, we want these example files linted and tested to make sure they are high-quality, however the -raw files are often intentionally bad style.

Not sure if that is that is what is going on here, as I don't see any documentation references to this file.

ucodery · 2026-05-30T03:37:55Z

+def calc_annual_mean(df: pandas.DataFrame):
+    """Function to calculate the mean temperature for each year and the final mean"""
+    # TODO: make this a bit more robust so we can write integration test examples??
+    # Calculate the mean temperature for each year


Maybe this is okay for an example project. But I would block this as over-commented for a deployed library.

sosey added 2 commits May 18, 2026 17:07

add example setuptools files (issue pyOpenSci#248)

67a15b5

remove egg files that got committed

5b03302

lwasser added the pyconus-26 label May 18, 2026

lwasser added this to pyOpenSci sprints 2023-2026 May 19, 2026

willingc approved these changes May 19, 2026

View reviewed changes

willingc requested a review from lwasser May 20, 2026 18:41

lwasser moved this to pyconus-2026 in pyOpenSci sprints 2023-2026 May 22, 2026

lwasser requested a review from ucodery May 28, 2026 17:21

sneakers-the-rat reviewed May 28, 2026

View reviewed changes

ucodery reviewed May 30, 2026

View reviewed changes

	def celsius_to_fahrenheit(celsius):
	def celsius_to_fahrenheit(celsius: float) -> float:

	def fahrenheit_to_celsius(fahrenheit):
	def fahrenheit_to_celsius(fahrenheit: float) -> float:

		from examplePy.temperature import fahrenheit_to_celsius


		def calc_annual_mean(df: pandas.DataFrame):

		version_scheme = "release-branch-semver"


		[project.optional-dependencies]

		zip-safe = true
		include-package-data = false

		]


		[build-system]

Conversation

sosey commented May 18, 2026

Uh oh!

willingc left a comment

Choose a reason for hiding this comment

Uh oh!

lwasser commented May 28, 2026

Uh oh!

sneakers-the-rat left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

sneakers-the-rat left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ucodery left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants