InvokeAI/invokeai/app/services/image_storage.py

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

159 lines
5.6 KiB
Python
Raw Normal View History

# Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654)
import datetime
import os
2023-04-04 01:05:15 +00:00
from glob import glob
2023-03-03 06:02:00 +00:00
from abc import ABC, abstractmethod
from enum import Enum
from pathlib import Path
from queue import Queue
2023-04-04 01:05:15 +00:00
from typing import Callable, Dict
2023-03-03 06:02:00 +00:00
from PIL.Image import Image
2023-04-04 03:08:42 +00:00
from invokeai.app.datatypes.image import ImageField, ImageType
2023-04-04 01:05:15 +00:00
from invokeai.app.services.item_storage import PaginatedResults
2023-04-03 04:34:07 +00:00
from invokeai.app.util.save_thumbnail import save_thumbnail
2023-03-03 06:02:00 +00:00
from invokeai.backend.image_util import PngWriter
class ImageStorageBase(ABC):
"""Responsible for storing and retrieving images."""
@abstractmethod
def get(self, image_type: ImageType, image_name: str) -> Image:
pass
2023-04-04 01:05:15 +00:00
@abstractmethod
def list(
self, image_type: ImageType, page: int = 0, per_page: int = 10
) -> PaginatedResults[ImageField]:
2023-04-04 01:05:15 +00:00
pass
# TODO: make this a bit more flexible for e.g. cloud storage
@abstractmethod
def get_path(self, image_type: ImageType, image_name: str) -> str:
pass
@abstractmethod
def save(self, image_type: ImageType, image_name: str, image: Image) -> None:
pass
@abstractmethod
def delete(self, image_type: ImageType, image_name: str) -> None:
pass
def create_name(self, context_id: str, node_id: str) -> str:
2023-03-03 06:02:00 +00:00
return f"{context_id}_{node_id}_{str(int(datetime.datetime.now(datetime.timezone.utc).timestamp()))}.png"
class DiskImageStorage(ImageStorageBase):
"""Stores images on disk"""
2023-03-03 06:02:00 +00:00
__output_folder: str
__pngWriter: PngWriter
2023-03-03 06:02:00 +00:00
__cache_ids: Queue # TODO: this is an incredibly naive cache
__cache: Dict[str, Image]
__max_cache_size: int
def __init__(self, output_folder: str):
self.__output_folder = output_folder
self.__pngWriter = PngWriter(output_folder)
self.__cache = dict()
self.__cache_ids = Queue()
2023-03-03 06:02:00 +00:00
self.__max_cache_size = 10 # TODO: get this from config
Path(output_folder).mkdir(parents=True, exist_ok=True)
# TODO: don't hard-code. get/save/delete should maybe take subpath?
for image_type in ImageType:
2023-03-03 06:02:00 +00:00
Path(os.path.join(output_folder, image_type)).mkdir(
parents=True, exist_ok=True
)
2023-04-03 04:34:07 +00:00
Path(os.path.join(output_folder, image_type, "thumbnails")).mkdir(
parents=True, exist_ok=True
)
def list(
self, image_type: ImageType, page: int = 0, per_page: int = 10
) -> PaginatedResults[ImageField]:
2023-04-04 01:05:15 +00:00
dir_path = os.path.join(self.__output_folder, image_type)
image_paths = glob(f"{dir_path}/*.png")
# just want the filenames
image_filenames = list(map(lambda i: os.path.basename(i), image_paths))
# we want to sort the images by timestamp, but we don't trust the filesystem
# we do have a timestamp in the filename: `{uuid}_{timestamp}.png`
sorted_paths = sorted(
# extract the timestamp as int and multiply -1 to reverse sorting
image_filenames, key=lambda i: int(os.path.splitext(i)[0].split("_")[-1]) * -1
)
all_images = list(
map(lambda i: ImageField(image_type=image_type, image_name=i), sorted_paths)
)
2023-04-04 01:05:15 +00:00
count = len(all_images)
page_of_images = all_images[page * per_page : (page + 1) * per_page]
page_count_trunc = int(count / per_page)
page_count_mod = count % per_page
page_count = page_count_trunc if page_count_mod == 0 else page_count_trunc + 1
2023-04-04 01:05:15 +00:00
return PaginatedResults[ImageField](
items=page_of_images,
page=page,
pages=page_count,
per_page=per_page,
total=count,
2023-04-04 01:05:15 +00:00
)
def get(self, image_type: ImageType, image_name: str) -> Image:
image_path = self.get_path(image_type, image_name)
cache_item = self.__get_cache(image_path)
if cache_item:
return cache_item
image = Image.open(image_path)
self.__set_cache(image_path, image)
return image
# TODO: make this a bit more flexible for e.g. cloud storage
def get_path(self, image_type: ImageType, image_name: str) -> str:
path = os.path.join(self.__output_folder, image_type, image_name)
return path
def save(self, image_type: ImageType, image_name: str, image: Image) -> None:
image_subpath = os.path.join(image_type, image_name)
2023-03-03 06:02:00 +00:00
self.__pngWriter.save_image_and_prompt_to_png(
image, "", image_subpath, None
) # TODO: just pass full path to png writer
2023-04-03 04:34:07 +00:00
save_thumbnail(
image=image,
filename=image_name,
path=os.path.join(self.__output_folder, image_type, "thumbnails"),
)
image_path = self.get_path(image_type, image_name)
self.__set_cache(image_path, image)
def delete(self, image_type: ImageType, image_name: str) -> None:
image_path = self.get_path(image_type, image_name)
if os.path.exists(image_path):
os.remove(image_path)
2023-03-03 06:02:00 +00:00
if image_path in self.__cache:
del self.__cache[image_path]
def __get_cache(self, image_name: str) -> Image:
return None if image_name not in self.__cache else self.__cache[image_name]
def __set_cache(self, image_name: str, image: Image):
if not image_name in self.__cache:
self.__cache[image_name] = image
2023-03-03 06:02:00 +00:00
self.__cache_ids.put(
image_name
) # TODO: this should refresh position for LRU cache
if len(self.__cache) > self.__max_cache_size:
cache_id = self.__cache_ids.get()
del self.__cache[cache_id]