didactic-meme / pyfuncol Goto Github PK

Describe the solution you'd like
We should optimise chained transformations in a similar way as what Apache Spark does: they should be lazy and re-ordered so that transformations reducing the number of elements are always executed first.

Describe the solution you'd like
List operations like lazy_map, lazy_filter should return generators instead of lists.

Describe alternatives you've considered
An alternative would be adding a lazy argument to each operation, but it's less clean IMHO.

It would be great if forbiddenfruit was made optional. Since forbiddenfruit requires CPython and I think if it would be better if it could work on other versions of python.

Instead of:

import pyfuncol

my_list = [1, 2, 3]
print(my_list.map(lambda x: x * 2))

I can do:

from pyfuncol.list import *

my_list = [1, 2, 3]
print(map(my_list, lambda x: x * 2))

Maybe there can be a check to see if forbiddenfruit is installed. If it is, add all the functions to the builtin classes, otherwise do nothing.

Sorry if I misunderstand how this works.

Is your feature request related to a problem? Please describe.
Expensive computations that are repeated in a map do not leverage the fact that we already computed once the result.

Describe the solution you'd like
We can memoize pure functions on hashable inputs with something like:

cached_f = functools.cache(f)
return [cached_f(x) for x in self]

To not restrict the library's usages, it would probably make sense to create pure_map, pure_filter, etc.

Describe the bug
Applying transformations on built-in subclasses, like OrderedDict, returns a plain dict.

Expected behavior
Subclasses should be preserved.

Desktop (please complete the following information):

• OS: macOS Monterey 12.1
• Python version 3.9.7
• pyfuncol version 1.0

Additional context
Could be solved by doing type(self)(...) before returning the result.

Could be useful this link.