import gradio as gr import groq import os import tempfile import uuid from dotenv import load_dotenv from langchain.text_splitter import RecursiveCharacterTextSplitter from langchain.vectorstores import FAISS from langchain.embeddings import HuggingFaceEmbeddings import fitz # PyMuPDF import base64 from PIL import Image import io import requests import json import re from datetime import datetime, timedelta # Load environment variables load_dotenv() client = groq.Client(api_key=os.getenv("GROQ_TECH_API_KEY")) embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2") # Directory to store FAISS indexes FAISS_INDEX_DIR = "faiss_indexes_tech" if not os.path.exists(FAISS_INDEX_DIR): os.makedirs(FAISS_INDEX_DIR) # Dictionary to store user-specific vectorstores user_vectorstores = {} # Custom CSS for Tech theme custom_css = """ :root { --primary-color: #4285F4; /* Google Blue */ --secondary-color: #34A853; /* Google Green */ --light-background: #F8F9FA; --dark-text: #202124; --white: #FFFFFF; --border-color: #DADCE0; --code-bg: #F1F3F4; --code-text: #37474F; --error-color: #EA4335; /* Google Red */ --warning-color: #FBBC04; /* Google Yellow */ } body { background-color: var(--light-background); font-family: 'Google Sans', 'Roboto', sans-serif; } .container { max-width: 1200px !important; margin: 0 auto !important; padding: 10px; } .header { background-color: var(--white); border-bottom: 1px solid var(--border-color); padding: 15px 0; margin-bottom: 20px; border-radius: 12px 12px 0 0; box-shadow: 0 1px 2px rgba(0,0,0,0.05); } .header-title { color: var(--primary-color); font-size: 1.8rem; font-weight: 700; text-align: center; } .header-subtitle { color: var(--dark-text); font-size: 1rem; text-align: center; margin-top: 5px; } .chat-container { border-radius: 8px !important; box-shadow: 0 1px 3px rgba(0,0,0,0.1) !important; background-color: var(--white) !important; border: 1px solid var(--border-color) !important; min-height: 500px; } .message-user { background-color: var(--primary-color) !important; color: var(--white) !important; border-radius: 18px 18px 4px 18px !important; padding: 12px 16px !important; margin-left: auto !important; max-width: 80% !important; } .message-bot { background-color: #F1F3F4 !important; color: var(--dark-text) !important; border-radius: 18px 18px 18px 4px !important; padding: 12px 16px !important; margin-right: auto !important; max-width: 80% !important; } .input-area { background-color: var(--white) !important; border-top: 1px solid var(--border-color) !important; padding: 12px !important; border-radius: 0 0 12px 12px !important; } .input-box { border: 1px solid var(--border-color) !important; border-radius: 24px !important; padding: 12px 16px !important; box-shadow: 0 1px 2px rgba(0,0,0,0.05) !important; } .send-btn { background-color: var(--primary-color) !important; border-radius: 24px !important; color: var(--white) !important; padding: 10px 20px !important; font-weight: 500 !important; } .clear-btn { background-color: #F1F3F4 !important; border: 1px solid var(--border-color) !important; border-radius: 24px !important; color: var(--dark-text) !important; padding: 8px 16px !important; font-weight: 500 !important; } .pdf-viewer-container { border-radius: 8px !important; box-shadow: 0 1px 3px rgba(0,0,0,0.1) !important; background-color: var(--white) !important; border: 1px solid var(--border-color) !important; padding: 20px; } .pdf-viewer-image { max-width: 100%; height: auto; border: 1px solid var(--border-color); border-radius: 8px; box-shadow: 0 1px 2px rgba(0,0,0,0.05); } .stats-box { background-color: #E8F0FE; padding: 10px; border-radius: 8px; margin-top: 10px; } .tool-container { background-color: var(--white); border-radius: 8px; box-shadow: 0 1px 3px rgba(0,0,0,0.1); padding: 15px; margin-bottom: 20px; border: 1px solid var(--border-color); } .code-block { background-color: var(--code-bg); color: var(--code-text); padding: 12px; border-radius: 8px; font-family: 'Roboto Mono', monospace; overflow-x: auto; margin: 10px 0; border-left: 3px solid var(--primary-color); } .repo-card { border: 1px solid var(--border-color); padding: 15px; margin: 10px 0; border-radius: 8px; background-color: var(--white); } .repo-name { color: var(--primary-color); font-weight: bold; font-size: 1.1rem; margin-bottom: 5px; } .repo-description { color: var(--dark-text); font-size: 0.9rem; margin-bottom: 10px; } .repo-stats { display: flex; gap: 15px; color: #5F6368; font-size: 0.85rem; } .repo-stat { display: flex; align-items: center; gap: 5px; } .qa-card { border-left: 3px solid var(--secondary-color); padding: 10px 15px; margin: 15px 0; background-color: #F8F9FA; border-radius: 0 8px 8px 0; } .qa-title { font-weight: bold; color: var(--dark-text); margin-bottom: 5px; } .qa-body { color: var(--dark-text); font-size: 0.95rem; margin-bottom: 10px; } .qa-meta { display: flex; justify-content: space-between; color: #5F6368; font-size: 0.85rem; } .tag { background-color: #E8F0FE; color: var(--primary-color); padding: 4px 8px; border-radius: 4px; font-size: 0.8rem; margin-right: 5px; display: inline-block; } """ # Function to process PDF files def process_pdf(pdf_file): if pdf_file is None: return None, "No file uploaded", {"page_images": [], "total_pages": 0, "total_words": 0} try: session_id = str(uuid.uuid4()) with tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) as temp_file: temp_file.write(pdf_file) pdf_path = temp_file.name doc = fitz.open(pdf_path) texts = [page.get_text() for page in doc] page_images = [] for page in doc: pix = page.get_pixmap() img_bytes = pix.tobytes("png") img_base64 = base64.b64encode(img_bytes).decode("utf-8") page_images.append(img_base64) total_pages = len(doc) total_words = sum(len(text.split()) for text in texts) doc.close() text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200) chunks = text_splitter.create_documents(texts) vectorstore = FAISS.from_documents(chunks, embeddings) index_path = os.path.join(FAISS_INDEX_DIR, session_id) vectorstore.save_local(index_path) user_vectorstores[session_id] = vectorstore os.unlink(pdf_path) pdf_state = {"page_images": page_images, "total_pages": total_pages, "total_words": total_words} return session_id, f"✅ Successfully processed {len(chunks)} text chunks from your PDF", pdf_state except Exception as e: if "pdf_path" in locals() and os.path.exists(pdf_path): os.unlink(pdf_path) return None, f"Error processing PDF: {str(e)}", {"page_images": [], "total_pages": 0, "total_words": 0} # Function to generate chatbot responses with Tech theme def generate_response(message, session_id, model_name, history): if not message: return history try: context = "" if session_id and session_id in user_vectorstores: vectorstore = user_vectorstores[session_id] docs = vectorstore.similarity_search(message, k=3) if docs: context = "\n\nRelevant information from uploaded PDF:\n" + "\n".join(f"- {doc.page_content}" for doc in docs) # Check if it's a GitHub repo search if re.match(r'^/github\s+.+', message, re.IGNORECASE): query = re.sub(r'^/github\s+', '', message, flags=re.IGNORECASE) repo_results = search_github_repos(query) if repo_results: response = "**GitHub Repository Search Results:**\n\n" for repo in repo_results[:3]: # Limit to top 3 results response += f"**[{repo['name']}]({repo['html_url']})**\n" if repo['description']: response += f"{repo['description']}\n" response += f"⭐ {repo['stargazers_count']} | 🍴 {repo['forks_count']} | Language: {repo['language'] or 'Not specified'}\n" response += f"Updated: {repo['updated_at'][:10]}\n\n" history.append((message, response)) return history else: history.append((message, "No GitHub repositories found for your query.")) return history # Check if it's a Stack Overflow search if re.match(r'^/stack\s+.+', message, re.IGNORECASE): query = re.sub(r'^/stack\s+', '', message, flags=re.IGNORECASE) qa_results = search_stackoverflow(query) if qa_results: response = "**Stack Overflow Search Results:**\n\n" for qa in qa_results[:3]: # Limit to top 3 results response += f"**[{qa['title']}]({qa['link']})**\n" response += f"Score: {qa['score']} | Answers: {qa['answer_count']}\n" if 'tags' in qa and qa['tags']: response += f"Tags: {', '.join(qa['tags'][:5])}\n" response += f"Asked: {qa['creation_date']}\n\n" history.append((message, response)) return history else: history.append((message, "No Stack Overflow questions found for your query.")) return history # Check if it's a code explanation request code_match = re.search(r'/explain\s+```(?:.+?)?\n(.+?)```', message, re.DOTALL) if code_match: code = code_match.group(1).strip() explanation = explain_code(code) history.append((message, explanation)) return history system_prompt = "You are a technical assistant specializing in software development, programming, and IT topics." system_prompt += " Format code snippets with proper markdown code blocks with language specified." system_prompt += " For technical explanations, be precise and include examples where helpful." if context: system_prompt += " Use the following context to answer the question if relevant: " + context completion = client.chat.completions.create( model=model_name, messages=[ {"role": "system", "content": system_prompt}, {"role": "user", "content": message} ], temperature=0.7, max_tokens=1024 ) response = completion.choices[0].message.content history.append((message, response)) return history except Exception as e: history.append((message, f"Error generating response: {str(e)}")) return history # Functions to update PDF viewer def update_pdf_viewer(pdf_state): if not pdf_state["total_pages"]: return 0, None, "No PDF uploaded yet" try: img_data = base64.b64decode(pdf_state["page_images"][0]) img = Image.open(io.BytesIO(img_data)) return pdf_state["total_pages"], img, f"**Total Pages:** {pdf_state['total_pages']}\n**Total Words:** {pdf_state['total_words']}" except Exception as e: print(f"Error decoding image: {e}") return 0, None, "Error displaying PDF" def update_image(page_num, pdf_state): if not pdf_state["total_pages"] or page_num < 1 or page_num > pdf_state["total_pages"]: return None try: img_data = base64.b64decode(pdf_state["page_images"][page_num - 1]) img = Image.open(io.BytesIO(img_data)) return img except Exception as e: print(f"Error decoding image: {e}") return None # GitHub API integration def search_github_repos(query, sort="stars", order="desc", per_page=10): """Search for GitHub repositories""" try: github_token = os.getenv("GITHUB_TOKEN", "") headers = {} if github_token: headers["Authorization"] = f"token {github_token}" params = { "q": query, "sort": sort, "order": order, "per_page": per_page } response = requests.get( "https://api.github.com/search/repositories", headers=headers, params=params ) if response.status_code != 200: print(f"GitHub API Error: {response.status_code} - {response.text}") return [] data = response.json() return data.get("items", []) except Exception as e: print(f"Error in GitHub search: {e}") return [] # Stack Overflow API integration def search_stackoverflow(query, sort="votes", site="stackoverflow", pagesize=10): """Search for questions on Stack Overflow""" try: params = { "order": "desc", "sort": sort, "site": site, "pagesize": pagesize, "intitle": query } response = requests.get( "https://api.stackexchange.com/2.3/search/advanced", params=params ) if response.status_code != 200: print(f"Stack Exchange API Error: {response.status_code} - {response.text}") return [] data = response.json() # Process results to convert Unix timestamps to readable dates for item in data.get("items", []): if "creation_date" in item: item["creation_date"] = datetime.fromtimestamp(item["creation_date"]).strftime("%Y-%m-%d") return data.get("items", []) except Exception as e: print(f"Error in Stack Overflow search: {e}") return [] def get_stackoverflow_answers(question_id, site="stackoverflow"): """Get answers for a specific question on Stack Overflow""" try: params = { "order": "desc", "sort": "votes", "site": site, "filter": "withbody" # Include the answer body in the response } response = requests.get( f"https://api.stackexchange.com/2.3/questions/{question_id}/answers", params=params ) if response.status_code != 200: print(f"Stack Exchange API Error: {response.status_code} - {response.text}") return [] data = response.json() # Process results for item in data.get("items", []): if "creation_date" in item: item["creation_date"] = datetime.fromtimestamp(item["creation_date"]).strftime("%Y-%m-%d") return data.get("items", []) except Exception as e: print(f"Error getting Stack Overflow answers: {e}") return [] def explain_code(code): """Explain code using LLM""" try: system_prompt = "You are an expert programmer and code reviewer. Your task is to explain the provided code in a clear, concise manner. Include:" system_prompt += "\n1. What the code does (high-level overview)" system_prompt += "\n2. Key functions/components and their purposes" system_prompt += "\n3. Potential issues or optimization opportunities" system_prompt += "\n4. Any best practices that are followed or violated" completion = client.chat.completions.create( model="llama3-70b-8192", # Using more capable model for code explanation messages=[ {"role": "system", "content": system_prompt}, {"role": "user", "content": f"Explain this code:\n```\n{code}\n```"} ], temperature=0.3, max_tokens=1024 ) explanation = completion.choices[0].message.content return f"**Code Explanation:**\n\n{explanation}" except Exception as e: return f"Error explaining code: {str(e)}" def perform_repo_search(query, language, sort_by, min_stars): """Perform GitHub repository search with UI parameters""" try: if not query: return "Please enter a search query" # Build the search query with filters search_query = query if language and language != "any": search_query += f" language:{language}" if min_stars and min_stars != "0": search_query += f" stars:>={min_stars}" # Map sort_by to GitHub API parameters sort_param = "stars" if sort_by == "updated": sort_param = "updated" elif sort_by == "forks": sort_param = "forks" results = search_github_repos(search_query, sort=sort_param) if not results: return "No repositories found. Try different search terms." # Format results as markdown markdown = "## GitHub Repository Search Results\n\n" for i, repo in enumerate(results, 1): markdown += f"### {i}. [{repo['full_name']}]({repo['html_url']})\n\n" if repo['description']: markdown += f"{repo['description']}\n\n" markdown += f"**Language:** {repo['language'] or 'Not specified'}\n" markdown += f"**Stars:** {repo['stargazers_count']} | **Forks:** {repo['forks_count']} | **Watchers:** {repo['watchers_count']}\n" markdown += f"**Created:** {repo['created_at'][:10]} | **Updated:** {repo['updated_at'][:10]}\n\n" if repo.get('topics'): markdown += f"**Topics:** {', '.join(repo['topics'])}\n\n" if repo.get('license') and repo['license'].get('name'): markdown += f"**License:** {repo['license']['name']}\n\n" markdown += f"[View Repository]({repo['html_url']}) | [Clone URL]({repo['clone_url']})\n\n" markdown += "---\n\n" return markdown except Exception as e: return f"Error searching for repositories: {str(e)}" def perform_stack_search(query, tag, sort_by): """Perform Stack Overflow search with UI parameters""" try: if not query: return "Please enter a search query" # Add tag to query if specified if tag and tag != "any": query_with_tag = f"{query} [tag:{tag}]" else: query_with_tag = query # Map sort_by to Stack Exchange API parameters sort_param = "votes" if sort_by == "newest": sort_param = "creation" elif sort_by == "activity": sort_param = "activity" results = search_stackoverflow(query_with_tag, sort=sort_param) if not results: return "No questions found. Try different search terms." # Format results as markdown markdown = "## Stack Overflow Search Results\n\n" for i, question in enumerate(results, 1): markdown += f"### {i}. [{question['title']}]({question['link']})\n\n" # Score and answer stats markdown += f"**Score:** {question['score']} | **Answers:** {question['answer_count']}" if question.get('is_answered'): markdown += " ✓ (Accepted answer available)" markdown += "\n\n" # Tags if question.get('tags'): markdown += "**Tags:** " for tag in question['tags']: markdown += f"`{tag}` " markdown += "\n\n" # Asked info markdown += f"**Asked:** {question['creation_date']} | **Views:** {question.get('view_count', 'N/A')}\n\n" markdown += f"[View Question]({question['link']})\n\n" markdown += "---\n\n" return markdown except Exception as e: return f"Error searching Stack Overflow: {str(e)}" # Gradio interface with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo: current_session_id = gr.State(None) pdf_state = gr.State({"page_images": [], "total_pages": 0, "total_words": 0}) gr.HTML("""
Tech-Vision
Analyze technical documents with Groq's LLM API.
""") with gr.Row(elem_classes="container"): with gr.Column(scale=1, min_width=300): pdf_file = gr.File(label="Upload PDF Document", file_types=[".pdf"], type="binary") upload_button = gr.Button("Process PDF", variant="primary") pdf_status = gr.Markdown("No PDF uploaded yet") model_dropdown = gr.Dropdown( choices=["llama3-70b-8192", "llama3-8b-8192", "mixtral-8x7b-32768", "gemma-7b-it"], value="llama3-70b-8192", label="Select Groq Model" ) # Tech Tools Section gr.Markdown("### Developer Tools", elem_classes="tool-title") with gr.Box(elem_classes="tool-container"): with gr.Tabs(): with gr.TabItem("GitHub Search"): repo_query = gr.Textbox(label="Search Query", placeholder="Enter keywords to search for repositories") with gr.Row(): language = gr.Dropdown( choices=["any", "JavaScript", "Python", "Java", "C++", "TypeScript", "Go", "Rust", "PHP", "C#"], value="any", label="Language" ) min_stars = gr.Dropdown( choices=["0", "10", "50", "100", "1000", "10000"], value="0", label="Min Stars" ) sort_by = gr.Dropdown( choices=["stars", "forks", "updated"], value="stars", label="Sort By" ) repo_search_btn = gr.Button("Search Repositories") with gr.TabItem("Stack Overflow"): stack_query = gr.Textbox(label="Search Query", placeholder="Enter your technical question") with gr.Row(): tag = gr.Dropdown( choices=["any", "python", "javascript", "java", "c++", "react", "node.js", "android", "ios", "sql"], value="any", label="Tag" ) so_sort_by = gr.Dropdown( choices=["votes", "newest", "activity"], value="votes", label="Sort By" ) so_search_btn = gr.Button("Search Stack Overflow") with gr.TabItem("Code Explainer"): code_input = gr.Textbox( label="Code to Explain", placeholder="Paste your code here...", lines=10 ) explain_btn = gr.Button("Explain Code") with gr.Column(scale=2, min_width=600): with gr.Tabs(): with gr.TabItem("PDF Viewer"): with gr.Column(elem_classes="pdf-viewer-container"): page_slider = gr.Slider(minimum=1, maximum=1, step=1, label="Page Number", value=1) pdf_image = gr.Image(label="PDF Page", type="pil", elem_classes="pdf-viewer-image") stats_display = gr.Markdown("No PDF uploaded yet", elem_classes="stats-box") with gr.TabItem("GitHub Results"): repo_results = gr.Markdown("Search for repositories to see results here") with gr.TabItem("Stack Overflow Results"): stack_results = gr.Markdown("Search for questions to see results here") with gr.TabItem("Code Explanation"): code_explanation = gr.Markdown("Paste your code and click 'Explain Code' to see an explanation here") with gr.Row(elem_classes="container"): with gr.Column(scale=2, min_width=600): chatbot = gr.Chatbot(height=500, bubble_full_width=False, show_copy_button=True, elem_classes="chat-container") with gr.Row(): msg = gr.Textbox(show_label=False, placeholder="Ask about your document, type /github to search repos, or /stack to search Stack Overflow...", scale=5) send_btn = gr.Button("Send", scale=1) clear_btn = gr.Button("Clear Conversation") # Event Handlers upload_button.click( process_pdf, inputs=[pdf_file], outputs=[current_session_id, pdf_status, pdf_state] ).then( update_pdf_viewer, inputs=[pdf_state], outputs=[page_slider, pdf_image, stats_display] ) msg.submit( generate_response, inputs=[msg, current_session_id, model_dropdown, chatbot], outputs=[chatbot] ).then(lambda: "", None, [msg]) send_btn.click( generate_response, inputs=[msg, current_session_id, model_dropdown, chatbot], outputs=[chatbot] ).then(lambda: "", None, [msg]) clear_btn.click( lambda: ([], None, "No PDF uploaded yet", {"page_images": [], "total_pages": 0, "total_words": 0}, 0, None, "No PDF uploaded yet"), None, [chatbot, current_session_id, pdf_status, pdf_state, page_slider, pdf_image, stats_display] ) page_slider.change( update_image, inputs=[page_slider, pdf_state], outputs=[pdf_image] ) # Tech tool handlers repo_search_btn.click( perform_repo_search, inputs=[repo_query, language, sort_by, min_stars], outputs=[repo_results] ) so_search_btn.click( perform_stack_search, inputs=[stack_query, tag, so_sort_by], outputs=[stack_results] ) explain_btn.click( explain_code, inputs=[code_input], outputs=[code_explanation] ) # Add footer with attribution gr.HTML("""
Created by Calvin Allen Crawford
""") # Launch the app if __name__ == "__main__": demo.launch()