Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import requests
|
3 |
+
from bs4 import BeautifulSoup
|
4 |
+
import nltk
|
5 |
+
from nltk.corpus import stopwords
|
6 |
+
from nltk.tokenize import word_tokenize
|
7 |
+
import string
|
8 |
+
|
9 |
+
nltk.download('punkt')
|
10 |
+
nltk.download('stopwords')
|
11 |
+
|
12 |
+
def get_text_from_url(url):
|
13 |
+
try:
|
14 |
+
response = requests.get(url, timeout=10)
|
15 |
+
soup = BeautifulSoup(response.text, 'html.parser')
|
16 |
+
text = soup.get_text()
|
17 |
+
return text
|
18 |
+
except:
|
19 |
+
return ""
|
20 |
+
|
21 |
+
def extract_keywords(text):
|
22 |
+
text = text.lower()
|
23 |
+
tokens = word_tokenize(text)
|
24 |
+
words = [word for word in tokens if word.isalnum()]
|
25 |
+
stop_words = set(stopwords.words('english'))
|
26 |
+
keywords = [word for word in words if word not in stop_words]
|
27 |
+
return set(keywords)
|
28 |
+
|
29 |
+
def compare_keywords(url_a, url_b):
|
30 |
+
text_a = get_text_from_url(url_a)
|
31 |
+
text_b = get_text_from_url(url_b)
|
32 |
+
|
33 |
+
if not text_a or not text_b:
|
34 |
+
return "❌ Failed to fetch one or both websites. Please check URLs."
|
35 |
+
|
36 |
+
keywords_a = extract_keywords(text_a)
|
37 |
+
keywords_b = extract_keywords(text_b)
|
38 |
+
|
39 |
+
diff_keywords = keywords_b - keywords_a
|
40 |
+
|
41 |
+
sorted_keywords = sorted(diff_keywords)
|
42 |
+
return "\n".join(sorted_keywords)
|
43 |
+
|
44 |
+
interface = gr.Interface(
|
45 |
+
fn=compare_keywords,
|
46 |
+
inputs=[
|
47 |
+
gr.Textbox(label="Your Website (A)"),
|
48 |
+
gr.Textbox(label="Competitor Website (B)")
|
49 |
+
],
|
50 |
+
outputs="text",
|
51 |
+
title="🔍 Competitor Keyword Finder",
|
52 |
+
description="Enter your website and a competitor's. This tool finds keywords the competitor uses but you don’t."
|
53 |
+
)
|
54 |
+
|
55 |
+
interface.launch()
|