Update fetch_url_util.py
Browse files- fetch_url_util.py +14 -11
fetch_url_util.py
CHANGED
|
@@ -6,46 +6,49 @@ from curl_cffi import requests
|
|
| 6 |
def fetch_image_from_url(url):
|
| 7 |
"""
|
| 8 |
Robustly fetches an image from a URL using TLS impersonation and
|
| 9 |
-
dynamic Referer headers
|
| 10 |
"""
|
| 11 |
-
if not url
|
| 12 |
return None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
-
# Extract base domain
|
|
|
|
| 15 |
parsed_url = urllib.parse.urlparse(url)
|
| 16 |
base_domain = f"{parsed_url.scheme}://{parsed_url.netloc}/"
|
| 17 |
|
| 18 |
-
# Standard browser headers
|
| 19 |
headers = {
|
| 20 |
-
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/
|
| 21 |
"Accept": "image/avif,image/webp,image/apng,image/svg+xml,image/*,*/*;q=0.8",
|
| 22 |
"Accept-Language": "en-US,en;q=0.9",
|
| 23 |
"Referer": base_domain,
|
| 24 |
"Cache-Control": "no-cache",
|
| 25 |
-
"Pragma": "no-cache",
|
| 26 |
}
|
| 27 |
|
| 28 |
try:
|
| 29 |
-
# '
|
| 30 |
response = requests.get(
|
| 31 |
url,
|
| 32 |
headers=headers,
|
| 33 |
impersonate="chrome120",
|
| 34 |
timeout=15,
|
| 35 |
-
|
| 36 |
)
|
| 37 |
|
| 38 |
response.raise_for_status()
|
| 39 |
|
| 40 |
-
# Open and verify the image
|
| 41 |
image_bytes = io.BytesIO(response.content)
|
| 42 |
img = Image.open(image_bytes)
|
| 43 |
-
img.load()
|
| 44 |
|
| 45 |
return img.convert("RGB")
|
| 46 |
|
| 47 |
except Exception as e:
|
| 48 |
print(f"Error fetching {url}: {str(e)}")
|
| 49 |
-
# You can raise a specific Gradio error here if preferred
|
| 50 |
import gradio as gr
|
|
|
|
| 51 |
raise gr.Error(f"Failed to fetch image: {str(e)}")
|
|
|
|
| 6 |
def fetch_image_from_url(url):
|
| 7 |
"""
|
| 8 |
Robustly fetches an image from a URL using TLS impersonation and
|
| 9 |
+
dynamic Referer headers. Automatically handles missing http/https prefixes.
|
| 10 |
"""
|
| 11 |
+
if not url:
|
| 12 |
return None
|
| 13 |
+
|
| 14 |
+
# 1. Handle missing protocol (e.g., pixiv.net -> https://pixiv.net)
|
| 15 |
+
url = url.strip()
|
| 16 |
+
if not url.startswith(("http://", "https://")):
|
| 17 |
+
url = "https://" + url
|
| 18 |
|
| 19 |
+
# 2. Extract base domain for the Referer header
|
| 20 |
+
# This is crucial for Pixiv and similar sites to prevent 403 Forbidden
|
| 21 |
parsed_url = urllib.parse.urlparse(url)
|
| 22 |
base_domain = f"{parsed_url.scheme}://{parsed_url.netloc}/"
|
| 23 |
|
|
|
|
| 24 |
headers = {
|
| 25 |
+
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36",
|
| 26 |
"Accept": "image/avif,image/webp,image/apng,image/svg+xml,image/*,*/*;q=0.8",
|
| 27 |
"Accept-Language": "en-US,en;q=0.9",
|
| 28 |
"Referer": base_domain,
|
| 29 |
"Cache-Control": "no-cache",
|
|
|
|
| 30 |
}
|
| 31 |
|
| 32 |
try:
|
| 33 |
+
# Changed 'follow_redirects' to 'allow_redirects' for curl_cffi compatibility
|
| 34 |
response = requests.get(
|
| 35 |
url,
|
| 36 |
headers=headers,
|
| 37 |
impersonate="chrome120",
|
| 38 |
timeout=15,
|
| 39 |
+
allow_redirects=True
|
| 40 |
)
|
| 41 |
|
| 42 |
response.raise_for_status()
|
| 43 |
|
|
|
|
| 44 |
image_bytes = io.BytesIO(response.content)
|
| 45 |
img = Image.open(image_bytes)
|
| 46 |
+
img.load()
|
| 47 |
|
| 48 |
return img.convert("RGB")
|
| 49 |
|
| 50 |
except Exception as e:
|
| 51 |
print(f"Error fetching {url}: {str(e)}")
|
|
|
|
| 52 |
import gradio as gr
|
| 53 |
+
# This will show a red toast notification in the UI
|
| 54 |
raise gr.Error(f"Failed to fetch image: {str(e)}")
|