| import os |
| import requests |
| import json |
| import google.generativeai as genai |
|
|
| |
| SERPER_API_KEY = os.getenv('X_API_KEY') |
| GEMINI_API_KEY = os.getenv('GEMINI_API_KEY') |
| genai.configure(api_key=GEMINI_API_KEY) |
| model = genai.GenerativeModel('gemini-1.5-flash') |
|
|
| def search_articles(query:str): |
| """ |
| Searches for articles related to the query using Serper API. |
| Returns a list of dictionaries containing article URLs, headings, and text. |
| """ |
| articles = None |
| |
| url = "https://google.serper.dev/search" |
|
|
| payload = json.dumps({ |
| "q":query |
| }) |
| headers = { |
| 'X-API-KEY': SERPER_API_KEY, |
| 'Content-Type': 'application/json' |
| } |
|
|
| response = requests.request("POST", url, headers=headers, data=payload) |
| articles = response.text |
| return articles |
|
|
|
|
| def fetch_article_content(articles): |
| """ |
| Fetches the article content, extracting headings and text. |
| """ |
| content = "" |
| |
| data = json.loads(articles) |
|
|
| |
| if 'answerBox' in data: |
| if 'title' in data['answerBox']: |
| content += data['answerBox']['title']+"\n" |
| if 'snippet' in data['answerBox']: |
| content += data['answerBox']['snippet']+"\n" |
|
|
| |
| if 'organic' in data: |
| for result in data['organic']: |
| if 'title' in result: |
| content += result['title']+"\n" |
| if 'snippet' in result: |
| content += result['snippet']+"\n" |
|
|
| |
| if 'peopleAlsoAsk' in data: |
| for question in data['peopleAlsoAsk']: |
| if 'title' in question: |
| content += question['title']+"\n" |
| if 'snippet' in question: |
| content += question['snippet']+"\n" |
| return content.strip() |
|
|
|
|
| def generate_answer(content,query): |
| """ |
| Generates an answer from the concatenated content using GPT-4. |
| The content and the user's query are used to generate a contextual answer. |
| """ |
| |
| response = None |
| system_prompt = f"""You are a helpful assistant. Use the following context to answer the user's query. If the context doesn't contain relevant information, say so.\n |
| Below is the context : \n |
| {content}\n |
| Below is the user query: |
| {query}\n |
| Based on the user query above and the context given provide with highly accurate response for the user query . You should cover all points , each and every small concrete detail's in the context. |
| """ |
|
|
| response = model.generate_content(system_prompt) |
| return response.text |
| |
|
|
