Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add more local pipeline operations #24

Merged
merged 2 commits into from
May 22, 2021
Merged

Add more local pipeline operations #24

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e6e72d7
Add more local pipeline operations
replomancer May 18, 2021
9cb4e9c
Add better tests for laziness of local operations
replomancer May 22, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 9 additions & 9 deletions pipeline_dp/pipeline_operations.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,14 +149,14 @@ def count_per_element(self, rdd, stage_name: str = None):
class LocalPipelineOperations(PipelineOperations):
"""Local Pipeline adapter."""

def map(self, col, fn, stage_name: str = None):
def map(self, col, fn, stage_name: typing.Optional[str] = None):
return map(fn, col)

def map_tuple(self, col, fn, stage_name: typing.Optional[str] = None):
return (fn(k, v) for k, v in col)

def map_values(self, col, fn, stage_name: str):
pass
def map_values(self, col, fn, stage_name: typing.Optional[str] = None):
return ((k, fn(v)) for k, v in col)

def group_by_key(self, col, stage_name: typing.Optional[str] = None):
def group_by_key_generator():
Expand All @@ -167,17 +167,17 @@ def group_by_key_generator():
yield item
return group_by_key_generator()

def filter(self, col, fn, stage_name: str):
pass
def filter(self, col, fn, stage_name: typing.Optional[str] = None):
return filter(fn, col)

def keys(self, col, stage_name: str):
pass

def values(self, col, stage_name: str):
pass
def values(self, col, stage_name: typing.Optional[str] = None):
return (v for k, v in col)

def sample_fixed_per_key(self, col, n: int, stage_name: str):
pass

def count_per_element(self, col, stage_name: str):
pass
def count_per_element(self, col, stage_name: typing.Optional[str] = None):
yield from collections.Counter(col).items()
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I like idea with Counter!

One question, in this line, is 'col' accessed the first time only when iterations over count_per_element() started?

What do you think about adding the 'lazyness' test:
Let 'exception_generator' is a generator which raise an exception on the first access to it

and then
def test_lazyness():
self.ops.map(exception_generator) # should be no exception
self.ops.map_tuple(exception_generator)
....

on implementing a new method it would be easy to add to the test and to ensure that the method is lazy

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Cool idea for testing!
I added a test like this and yes, count_per_element works as expected.

67 changes: 64 additions & 3 deletions tests/pipeline_operations_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,9 +46,8 @@ def setUpClass(cls):
cls.ops = LocalPipelineOperations()

def test_local_map(self):
some_map = self.ops.map([1, 2, 3], lambda x: x)
# some_map is its own consumable iterator
self.assertIs(some_map, iter(some_map))
self.assertEqual(list(self.ops.map([], lambda x: x / 0)),
[])

self.assertEqual(list(self.ops.map([1, 2, 3], str)),
["1", "2", "3"])
Expand All @@ -64,6 +63,17 @@ def test_local_map_tuple(self):
self.assertEqual(list(self.ops.map_tuple(tuple_list, lambda k, v: (
str(k), str(v)))), [("1", "2"), ("2", "3"), ("3", "4")])

def test_local_map_values(self):
self.assertEqual(list(self.ops.map_values([], lambda x: x / 0)),
[])

tuple_list = [(1, 2), (2, 3), (3, 4)]

self.assertEqual(list(self.ops.map_values(tuple_list, str)),
[(1, "2"), (2, "3"), (3, "4")])
self.assertEqual(list(self.ops.map_values(tuple_list, lambda x: x**2)),
[(1, 4), (2, 9), (3, 16)])

def test_local_group_by_key(self):
some_dict = [("cheese", "brie"), ("bread", "sourdough"),
("cheese", "swiss")]
Expand All @@ -72,6 +82,57 @@ def test_local_group_by_key(self):
("cheese", ["brie", "swiss"]),
("bread", ["sourdough"])])

def test_local_filter(self):
self.assertEqual(list(self.ops.filter([], lambda x: True)),
[])
self.assertEqual(list(self.ops.filter([], lambda x: False)),
[])

example_list = [1, 2, 2, 3, 3, 4, 2]

self.assertEqual(list(self.ops.filter(example_list, lambda x: x % 2)),
[1, 3, 3])
self.assertEqual(list(self.ops.filter(example_list, lambda x: x < 3)),
[1, 2, 2, 2])

def test_local_values(self):
self.assertEqual(list(self.ops.values([])),
[])

example_list = [(1, 2), (2, 3), (3, 4), (4, 8)]

self.assertEqual(list(self.ops.values(example_list)),
[2, 3, 4, 8])

def test_local_count_per_element(self):
example_list = [1, 2, 3, 4, 5, 6, 1, 4, 0, 1]
result = self.ops.count_per_element(example_list)

self.assertEqual(dict(result),
{1: 3, 2: 1, 3: 1, 4: 2, 5: 1, 6: 1, 0: 1})

def test_laziness(self):
def exceptions_generator_function():
yield 1 / 0

def assert_laziness(operator, *args):
try:
operator(exceptions_generator_function(), *args)
except ZeroDivisionError:
self.fail(f"local {operator.__name__} is not lazy")
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This function exists for tests to fail properly with AssertionError (assertNotRaises does not exists in unittest).

e.g. when filter implementation is broken:

        return list(filter(fn, col))

the fail looks like this:

tests/pipeline_operations_test.py:122: in assert_laziness
    self.fail(f"local {operator.__name__} is not lazy")
E   AssertionError: local filter is not lazy

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for explanation. I like that approach!


# reading from exceptions_generator_function() results in error:
self.assertRaises(ZeroDivisionError,
next, exceptions_generator_function())

# lazy operators accept exceptions_generator_function()
# as argument without raising errors:
assert_laziness(self.ops.map, str)
assert_laziness(self.ops.map_values, str)
assert_laziness(self.ops.filter, bool)
assert_laziness(self.ops.values)
assert_laziness(self.ops.count_per_element)


if __name__ == '__main__':
unittest.main()