1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
|
from abc import ABC, abstractmethod
from fractions import Fraction
from typing import Callable, Tuple
from PIL import Image
class Filter(ABC):
"""
A filter transforms a given input image to the desired output image. A filter can signal that it
is complete and will be removed from the pipeline.
:param str name: The name of the filter. The name is useful for debugging purposes.
"""
size: Tuple[int, int]
"The image size (width, height) in pixels that this filter transforms."
is_complete: bool
"Indicates if the filter is complete and should no longer be processed."
def __init__(self):
self.is_complete = False
@abstractmethod
def initialize(self, size: Tuple[int, int]):
"""Initializes the filter with the provided frame size. Since the construction
of the filter can happen before the size of the display is known, initialization
should be done here.
:param size: The filter image size
:type size: Tuple[int, int]
"""
pass
@abstractmethod
def transform(self, get_input: Callable[[], Image.Image], get_output: Callable[[int], Image.Image], input_changed: bool, time: Fraction) -> Tuple[Image.Image, int]:
"""
Transforms the given input image to the desired output image.
The default behaviour is to return the orignal image.
:param Callable[[], PIL.Image] get_input: A function that returns the input image to transform. Note that calling
this will create a copy of the input image, and it is safe to manipulate directly.
:param Callable[[int], PIL.Image] get_output: Provide the hashcode of the new frame and it will
return the output frame if it already exists. This avoids having to redraw an output frame that is already
cached.
:param bool input_changed: True if the input is different from previous run, False otherwise.
When true, you have to return an Image.
:param Fraction time: The current time in seconds, expressed as a fractional number since
the start of the pipeline.
:rtype: PIL.Image
:return: The transformed output image. If this filter did not modify the input, return None. This signals to the
pipeline manager that there was no change and a cached version will be moved to the next stage.
"""
pass
|
