One-off pipes

[YoraiLevi]

partial-like syntax for pipe construction
#93

[JulienPalard]

It works, but is it really usefull? If you have a nice, striking example, please write it in the README.md to document the feature (without documentation it will never be used).

[YoraiLevi]

It naturally extends the pattern provided by the existing curying examples.
https://github.com/JulienPalard/Pipe#partial-pipes
Right after the curying example.
Possible doc:
Create a partially initialized custom pipe in one line

some_iterable | Pipe(some_func, 1, 2, 3)

It's a little bit hard to add a specific example that fits the readme pattern of decorating functions with Pipe since this syntax is a shorthand for it...

[JulienPalard]

Playing around itertools there's a few "interesing" ones we could pick:

>>> list(range(10) | Pipe(itertools.accumulate, initial=1))
[1, 1, 2, 4, 7, 11, 16, 22, 29, 37, 46]
>>> list(range(10) | Pipe(itertools.batched, 2))
[(0, 1), (2, 3), (4, 5), (6, 7), (8, 9)]
>>> list(range(5) | Pipe(itertools.chain, range(5)))
[0, 1, 2, 3, 4, 0, 1, 2, 3, 4]
>>> list(range(5) | Pipe(itertools.combinations, r=2))
[(0, 1), (0, 2), (0, 3), (0, 4), (1, 2), (1, 3), (1, 4), (2, 3), (2, 4), (3, 4)]
>>> list(range(20) | Pipe(itertools.compress, selectors=[1, 0] * 10))
[0, 2, 4, 6, 8, 10, 12, 14, 16, 18]
>>> list(range(20) | Pipe(itertools.compress, selectors=[0, 1] * 10))
[1, 3, 5, 7, 9, 11, 13, 15, 17, 19]
>>> list(range(20) | Pipe(itertools.islice, 5, 10))
[5, 6, 7, 8, 9]

maybe based on this we can imagine a "good" example?

[YoraiLevi]

I think that all of the examples you've provided are good and demonstrate exactly my intentions with the syntax. However I still don't yet see how they fit the flow of the current documentation.

[JulienPalard]

Feels like it should be documented under # Constructing your own and before # Partial Pipes.

Feels like it goes near the end = Pipe(deque), maybe with an example using deque, like last_ten = Pipe(deque, maxlen=10)?

Just don't forget to run pytest in order to verify your examples (the test_readme test runs the README using the doctest module to validate all examples are true).

[YoraiLevi]

Added section describing the usage to readme, "One-off pipes"

python -m tox -q

py38: SKIP ⚠ in 0.22 seconds
py39: SKIP ⚠ in 0.02 seconds
py310: SKIP ⚠ in 0.03 seconds
================================================================= test session starts =================================================================
platform win32 -- Python 3.11.3, pytest-7.4.4, pluggy-1.3.0
cachedir: .tox\py311\.pytest_cache
rootdir: C:\Users\devic\OneDrive\Documents\Sources\Pipe
plugins: hypothesis-6.92.2
collected 8 items

README.md .                                                                                                                                      [ 12%]
tests\test_pipe.py .......                                                                                                                       [100%]

================================================================== 8 passed in 1.73s ================================================================== 
py311: OK ✔ in 28.3 seconds
py312: SKIP ⚠ in 0.02 seconds
py313: SKIP ⚠ in 0.03 seconds
flake8: OK ✔ in 9.75 seconds
All done! ✨ 🍰 ✨
2 files would be left unchanged.
black: OK ✔ in 11.22 seconds

--------------------------------------------------------------------
Your code has been rated at 10.00/10 (previous run: 10.00/10, +0.00)

pylint: OK ✔ in 22.86 seconds
Combined data file .tox\.coverage.py311.abcdef-123.9148.XeajDSyx
Name    Stmts   Miss Branch BrPart  Cover   Missing
---------------------------------------------------
TOTAL     129      0     82      0   100%

1 file skipped due to complete coverage.
  py38: SKIP (0.22 seconds)
  py39: SKIP (0.02 seconds)
  py310: SKIP (0.03 seconds)
  py311: OK (28.30 seconds)
  py312: SKIP (0.02 seconds)
  py313: SKIP (0.03 seconds)
  flake8: OK (9.75 seconds)
  black: OK (11.22 seconds)
  pylint: OK (22.86 seconds)
  coverage: OK (8.00 seconds)
  congratulations :) (80.64 seconds)

[JulienPalard]

Thanks!