google_search / app.py
mgokg's picture
Update app.py
8e85569 verified
raw
history blame
4.28 kB
import gradio as gr
import requests
from bs4 import BeautifulSoup
from gradio_client import Client
from urllib.parse import urljoin
import pandas as pd
from io import StringIO
import json
import groq
import os
custom_css = """
#md {
height: 400px;
font-size: 30px;
background: #202020;
padding: 20px;
color: white;
border: 1 px solid white;
}
"""
api_key = os.getenv('groq')
client = groq.Client(api_key=api_key)
def qwen(jsondata):
client = Client("Qwen/Qwen2.5-72B-Instruct")
result = client.predict(
query= f"return a valid json object \n {jsondata}",
history=[],
system="You are Qwen, created by Alibaba Cloud. You are a helpful assistant.",
api_name="/model_chat"
)
return result
def llm(message):
message = f"return a json object with the keys: name,email,phone,website \n the values can be found here, leave blank if value is not available:\n {message} \n return a json object only. no text, no explanaition"
try:
completion = client.chat.completions.create(
model="llama3-70b-8192",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": f"{message}"}
],
)
return completion.choices[0].message.content
except Exception as e:
return f"Error in response generation: {str(e)}"
def list_of_clubs(ort):
base_url = "https://vereine-in-deutschland.net"
all_links_text = []
initial_url = f"{base_url}/vereine/Bayern/{ort}"
try:
response = requests.get(initial_url)
response.raise_for_status()
soup = BeautifulSoup(response.content, 'html.parser')
# Determine the last page
link_element = soup.select_one('li.page-item:nth-child(8) > a:nth-child(1)')
last_page = 10
if link_element and 'href' in link_element.attrs:
href = link_element['href']
last_page = int(href.split('/')[-1])
# Loop through all pages and collect links
for page_number in range(1, last_page + 1):
page_url = f"{base_url}/vereine/Bayern/{ort}/p/{page_number}"
response = requests.get(page_url)
response.raise_for_status()
soup = BeautifulSoup(response.content, 'html.parser')
target_div = soup.select_one('div.row-cols-1:nth-child(4)')
if target_div:
texts = [a.text for a in target_div.find_all('a', href=True)]
all_links_text.extend(texts)
else:
print(f"Target div not found on page {page_number}")
except Exception as e:
return str(e), []
all_links_text = all_links_text[0::2]
return all_links_text
def process_ort(ort):
links_text = list_of_clubs(ort)
vereine = []
for verein in links_text:
headers = {
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
}
url = f"https://www.google.com/search?q=impressum {verein}"
response = requests.get(url, headers=headers)
soup = BeautifulSoup(response.content, 'html.parser')
impressum_div = soup.find('body')
contact_detailes = impressum_div.text
json_object = llm(contact_detailes)
#vereine.append(contact_detailes)
vereine.append(json_object)
valid_json = qwen(vereine)
# Convert JSON string to Python dictionary
data_dict = json.loads(valid_json)
Convert dictionary to DataFrame
df = pd.DataFrame(data_dict)
return df
return vereine
# Create the Gradio interface
with gr.Blocks(css=custom_css) as demo:
with gr.Row():
details_output = gr.DataFrame(label="answer", elem_id="md")
#details_output = gr.Textbox(label="Ausgabe", value = f"\n\n\n\n")
with gr.Row():
ort_input = gr.Textbox(label="prompt", placeholder="ask anything...")
with gr.Row():
button = gr.Button("Senden")
# Connect the button to the function
button.click(fn=process_ort, inputs=ort_input, outputs=details_output)
# Launch the Gradio application
demo.launch()