ayrisdev/crawl4ai
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: Temporarily disable fetching image file size in get_content_of_website_optimized

Browse Source 
Set the `image_size` variable to 0 in the `get_content_of_website_optimized` function to temporarily disable fetching the image file size. This change addresses performance issues and will be improved in a future update.

Update Dockerfile for linuz users
This commit is contained in:
unclecode
2024-07-31 13:29:23 +08:00
parent aa9412e1b4
commit 9e43f7beda
3 changed files with 19 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
crawl4ai/crawler_strategy.py
View File

@@ -137,10 +137,15 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
# self.service = Service(chromedriver_autoinstaller.install())
chromedriver_path = ChromeDriverManager().install()
self.service = Service(chromedriver_path)
self.service.log_path = "NUL"
self.driver = webdriver.Chrome(service=self.service, options=self.options)
# chromedriver_path = ChromeDriverManager().install()
# self.service = Service(chromedriver_path)
# self.service.log_path = "NUL"
# self.driver = webdriver.Chrome(service=self.service, options=self.options)
# Use selenium-manager (built into Selenium 4.10.0+)
self.service = Service()
self.driver = webdriver.Chrome(options=self.options)
self.driver = self.execute_hook('on_driver_created', self.driver)
if kwargs.get("cookies"):
@@ -292,7 +297,7 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
# Open the screenshot with PIL
image = Image.open(BytesIO(screenshot))
# Convert image to RGB mode
# Convert image to RGB mode (this will handle both RGB and RGBA images)
rgb_image = image.convert('RGB')
# Convert to JPEG and compress
@@ -304,11 +309,6 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
print(f"[LOG] 📸 Screenshot taken and converted to base64")
return img_base64
except Exception as e:
if self.verbose:
print(f"[ERROR] Failed to take screenshot: {str(e)}")
return ""
except Exception as e:
error_message = sanitize_input_encode(f"Failed to take screenshot: {str(e)}")
print(error_message)
@@ -321,7 +321,7 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
try:
font = ImageFont.truetype("arial.ttf", 40)
except IOError:
font = ImageFont.load_default(size=40)
font = ImageFont.load_default()
# Define text color and wrap the text
text_color = (255, 255, 255)
@@ -340,6 +340,6 @@ class LocalSeleniumCrawlerStrategy(CrawlerStrategy):
img_base64 = base64.b64encode(buffered.getvalue()).decode('utf-8')
return img_base64
def quit(self):
self.driver.quit()