Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -3,23 +3,20 @@ import requests
|
|
3 |
from bs4 import BeautifulSoup
|
4 |
from transformers import pipeline
|
5 |
|
6 |
-
#
|
7 |
summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")
|
8 |
|
9 |
def scrape_website(url):
|
10 |
"""Extracts text from a website with error handling"""
|
11 |
try:
|
12 |
-
headers = {'User-Agent': 'Mozilla/5.0'}
|
13 |
response = requests.get(url, headers=headers, timeout=10)
|
14 |
-
response.raise_for_status()
|
15 |
|
16 |
soup = BeautifulSoup(response.text, "html.parser")
|
17 |
-
|
18 |
-
# Extract text from common content-containing tags
|
19 |
text_elements = soup.find_all(['p', 'article', 'main', 'section'])
|
20 |
text = " ".join([e.get_text(strip=True, separator=' ') for e in text_elements])
|
21 |
-
|
22 |
-
return text if text.strip() else "No content found"
|
23 |
|
24 |
except Exception as e:
|
25 |
return f"Scraping Error: {str(e)}"
|
@@ -27,44 +24,97 @@ def scrape_website(url):
|
|
27 |
def summarize_website(url):
|
28 |
"""Handles the full summarization pipeline"""
|
29 |
try:
|
30 |
-
|
31 |
-
|
32 |
-
if "Error" in extracted_text:
|
33 |
-
return extracted_text
|
34 |
|
35 |
-
|
36 |
-
|
37 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
|
39 |
-
# Truncate text to model's max input length (1024 tokens for DistilBART)
|
40 |
-
max_input_length = 1000 # Conservative estimate for token count
|
41 |
-
truncated_text = extracted_text[:max_input_length]
|
42 |
-
|
43 |
-
# Generate summary
|
44 |
-
summary = summarizer(
|
45 |
-
truncated_text,
|
46 |
-
max_length=200,
|
47 |
-
min_length=50,
|
48 |
-
do_sample=False,
|
49 |
-
truncation=True # Ensure truncation is enabled
|
50 |
-
)
|
51 |
-
|
52 |
-
return f"**Summary:**\n\n{summary[0]['summary_text']}"
|
53 |
-
|
54 |
except Exception as e:
|
55 |
-
return f"Summarization Error: {str(e)}"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
56 |
|
57 |
-
#
|
58 |
-
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
69 |
|
70 |
-
|
|
|
|
|
|
|
|
|
|
|
|
3 |
from bs4 import BeautifulSoup
|
4 |
from transformers import pipeline
|
5 |
|
6 |
+
# Load summarization pipeline
|
7 |
summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")
|
8 |
|
9 |
def scrape_website(url):
|
10 |
"""Extracts text from a website with error handling"""
|
11 |
try:
|
12 |
+
headers = {'User-Agent': 'Mozilla/5.0'}
|
13 |
response = requests.get(url, headers=headers, timeout=10)
|
14 |
+
response.raise_for_status()
|
15 |
|
16 |
soup = BeautifulSoup(response.text, "html.parser")
|
|
|
|
|
17 |
text_elements = soup.find_all(['p', 'article', 'main', 'section'])
|
18 |
text = " ".join([e.get_text(strip=True, separator=' ') for e in text_elements])
|
19 |
+
return text.strip() if text.strip() else "No content found"
|
|
|
20 |
|
21 |
except Exception as e:
|
22 |
return f"Scraping Error: {str(e)}"
|
|
|
24 |
def summarize_website(url):
|
25 |
"""Handles the full summarization pipeline"""
|
26 |
try:
|
27 |
+
with gr.Column(variant="panel"):
|
28 |
+
gr.Markdown("## β‘ Processing...")
|
|
|
|
|
29 |
|
30 |
+
extracted_text = scrape_website(url)
|
31 |
+
|
32 |
+
if "Error" in extracted_text:
|
33 |
+
return f"β {extracted_text}"
|
34 |
+
|
35 |
+
if len(extracted_text.split()) < 50:
|
36 |
+
return "β οΈ Error: Insufficient content for summarization (minimum 50 words required)"
|
37 |
+
|
38 |
+
max_input_length = 1000
|
39 |
+
truncated_text = extracted_text[:max_input_length]
|
40 |
+
|
41 |
+
summary = summarizer(
|
42 |
+
truncated_text,
|
43 |
+
max_length=200,
|
44 |
+
min_length=50,
|
45 |
+
do_sample=False,
|
46 |
+
truncation=True
|
47 |
+
)
|
48 |
+
|
49 |
+
return f"## π Summary\n\n{summary[0]['summary_text']}"
|
50 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
51 |
except Exception as e:
|
52 |
+
return f"β Summarization Error: {str(e)}"
|
53 |
+
|
54 |
+
# Custom CSS for mobile optimization
|
55 |
+
css = """
|
56 |
+
@media screen and (max-width: 600px) {
|
57 |
+
.container {
|
58 |
+
padding: 10px !important;
|
59 |
+
}
|
60 |
+
.input-box textarea {
|
61 |
+
font-size: 16px !important;
|
62 |
+
}
|
63 |
+
}
|
64 |
+
"""
|
65 |
|
66 |
+
# Mobile-optimized interface with Blocks API
|
67 |
+
with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Website Summarizer") as app:
|
68 |
+
gr.Markdown("# π AI Website Summarizer")
|
69 |
+
gr.Markdown("Paste any website URL below to get an instant AI-powered summary!")
|
70 |
+
|
71 |
+
with gr.Row():
|
72 |
+
url_input = gr.Textbox(
|
73 |
+
label="Website URL",
|
74 |
+
placeholder="Enter full URL (https://...)",
|
75 |
+
lines=1,
|
76 |
+
max_lines=1,
|
77 |
+
elem_id="input-box"
|
78 |
+
)
|
79 |
+
|
80 |
+
with gr.Row():
|
81 |
+
submit_btn = gr.Button("Generate Summary π", variant="primary")
|
82 |
+
clear_btn = gr.Button("Clear π")
|
83 |
+
|
84 |
+
output = gr.Markdown()
|
85 |
+
|
86 |
+
# Example section
|
87 |
+
gr.Examples(
|
88 |
+
examples=[
|
89 |
+
["https://en.wikipedia.org/wiki/Large_language_model"],
|
90 |
+
["https://www.bbc.com/news/technology-66510295"]
|
91 |
+
],
|
92 |
+
inputs=url_input,
|
93 |
+
label="Try these examples:",
|
94 |
+
examples_per_page=2
|
95 |
+
)
|
96 |
+
|
97 |
+
# Progress indicator
|
98 |
+
progress = gr.Textbox(visible=False)
|
99 |
+
|
100 |
+
# Event handlers
|
101 |
+
submit_btn.click(
|
102 |
+
fn=summarize_website,
|
103 |
+
inputs=url_input,
|
104 |
+
outputs=output,
|
105 |
+
api_name="summarize"
|
106 |
+
)
|
107 |
+
|
108 |
+
clear_btn.click(
|
109 |
+
fn=lambda: ("", ""),
|
110 |
+
inputs=None,
|
111 |
+
outputs=[url_input, output],
|
112 |
+
queue=False
|
113 |
+
)
|
114 |
|
115 |
+
# Mobile-friendly configuration
|
116 |
+
app.launch(
|
117 |
+
server_name="0.0.0.0",
|
118 |
+
server_port=7860,
|
119 |
+
favicon_path="https://www.svgrepo.com/show/355037/huggingface.svg"
|
120 |
+
)
|