CodeCow/Scrapegraph-ai
1
0
Fork 0
mirror of https://github.com/VinciGit00/Scrapegraph-ai.git synced 2026-07-01 21:00:48 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

fix(ScreenshotScraper): impose dynamic imports

Browse Source
This commit is contained in:
Federico Aguzzi 2024-09-02 15:03:54 +02:00
parent 08afc9292e
commit b8ef93738e
2 changed files with 31 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
scrapegraphai/utils/screenshot_scraping/screenshot_preparation.py
View File

@ -5,7 +5,6 @@ import asyncio
from io import BytesIO from io import BytesIO
from PIL import Image, ImageGrab from PIL import Image, ImageGrab
from playwright.async_api import async_playwright from playwright.async_api import async_playwright
import cv2 as cv
import numpy as np import numpy as np
from io import BytesIO from io import BytesIO
@ -42,6 +41,12 @@ def select_area_with_opencv(image):
A tuple containing the LEFT, TOP, RIGHT, and BOTTOM coordinates of the selected area. A tuple containing the LEFT, TOP, RIGHT, and BOTTOM coordinates of the selected area.
""" """
try:
import cv2 as cv
except ImportError:
raise ImportError("The dependencies for screenshot scraping are not installed. Please install them using `pip install scrapegraphai[screenshot_scraper]`.")
fullscreen_screenshot = ImageGrab.grab() fullscreen_screenshot = ImageGrab.grab()
dw, dh = fullscreen_screenshot.size dw, dh = fullscreen_screenshot.size
@ -116,8 +121,12 @@ def select_area_with_ipywidget(image):
import matplotlib.pyplot as plt import matplotlib.pyplot as plt
import numpy as np import numpy as np
from ipywidgets import interact, IntSlider try:
import ipywidgets as widgets from ipywidgets import interact, IntSlider
import ipywidgets as widgets
except:
raise ImportError("The dependencies for screenshot scraping are not installed. Please install them using `pip install scrapegraphai[screenshot_scraper]`.")
from PIL import Image from PIL import Image
img_array = np.array(image) img_array = np.array(image)

33
scrapegraphai/utils/screenshot_scraping/text_detection.py
View File

@ -1,24 +1,29 @@
""" """
text_detection_module text_detection_module
""" """
from surya.ocr import run_ocr
from surya.model.detection.model import (load_model as load_det_model,
load_processor as load_det_processor)
from surya.model.recognition.model import load_model as load_rec_model
from surya.model.recognition.processor import load_processor as load_rec_processor
def detect_text(image, languages: list = ["en"]): def detect_text(image, languages: list = ["en"]):
""" """
Detects and extracts text from a given image. Detects and extracts text from a given image.
Parameters: Parameters:
image (PIL Image): The input image to extract text from. image (PIL Image): The input image to extract text from.
lahguages (list): A list of languages to detect text in. Defaults to ["en"]. List of languages can be found here: https://github.com/VikParuchuri/surya/blob/master/surya/languages.py lahguages (list): A list of languages to detect text in. Defaults to ["en"]. List of languages can be found here: https://github.com/VikParuchuri/surya/blob/master/surya/languages.py
Returns: Returns:
str: The extracted text from the image. str: The extracted text from the image.
Notes: Notes:
Model weights will automatically download the first time you run this function. Model weights will automatically download the first time you run this function.
""" """
try:
from surya.ocr import run_ocr
from surya.model.detection.model import (load_model as load_det_model,
load_processor as load_det_processor)
from surya.model.recognition.model import load_model as load_rec_model
from surya.model.recognition.processor import load_processor as load_rec_processor
except:
raise ImportError("The dependencies for screenshot scraping are not installed. Please install them using `pip install scrapegraphai[screenshot_scraper]`.")
langs = languages langs = languages
det_processor, det_model = load_det_processor(), load_det_model() det_processor, det_model = load_det_processor(), load_det_model()