Spaces:
Running
Running
| import streamlit as st | |
| from transformers import AutoTokenizer, AutoModelForSequenceClassification | |
| import os | |
| # Initialize Hugging Face model and tokenizer | |
| MODEL_NAME = "microsoft/codebert-base" | |
| # Load the pre-trained CodeBERT model for understanding code | |
| tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME) | |
| model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME) | |
| # Helper function to analyze code | |
| def analyze_code(code): | |
| # Split the code into manageable chunks | |
| max_length = 512 | |
| lines = code.split("\n") | |
| chunks = ["\n".join(lines[i:i+max_length]) for i in range(0, len(lines), max_length)] | |
| results = [] | |
| for chunk in chunks: | |
| tokenized_code = tokenizer(chunk, return_tensors="pt", truncation=True, max_length=max_length) | |
| outputs = model(**tokenized_code) | |
| logits = outputs.logits | |
| results.append(logits.argmax(dim=1).item()) | |
| return results | |
| # Function to simulate quality and bug detection | |
| # Note: Replace with a real analysis pipeline. | |
| def check_code_quality_and_bugs(code): | |
| suggestions = [] | |
| # Example: Check for readability (placeholder logic) | |
| if len(code.split("\n")) < 5: | |
| suggestions.append("Code seems too short, ensure functionality is implemented correctly.") | |
| # Example: Check for comments (placeholder logic) | |
| if "#" not in code: | |
| suggestions.append("Consider adding comments to improve code clarity.") | |
| # Example: Check for style guide adherence (placeholder logic) | |
| if "import" in code and "os" not in code: | |
| suggestions.append("Unused imports detected; consider removing them.") | |
| # Example: Detect code smells (placeholder logic) | |
| if "try" in code and "except" not in code: | |
| suggestions.append("`try` block without `except` may lead to unhandled exceptions.") | |
| return suggestions | |
| # Streamlit app UI | |
| st.title("Code Quality and Bug Detection Tool") | |
| st.markdown("Analyze your code for syntax issues, quality, and potential bugs.") | |
| # File uploader | |
| uploaded_file = st.file_uploader("Upload a Python code file", type=["py"]) | |
| # Code snippet input | |
| code_snippet = st.text_area("Or paste your code snippet below:") | |
| if st.button("Analyze Code"): | |
| if uploaded_file is not None: | |
| code = uploaded_file.read().decode("utf-8") | |
| elif code_snippet.strip(): | |
| code = code_snippet | |
| else: | |
| st.error("Please upload a file or paste code to analyze.") | |
| st.stop() | |
| # Perform code analysis | |
| st.subheader("Analysis Results") | |
| st.write("**Code Quality and Bug Suggestions:**") | |
| suggestions = check_code_quality_and_bugs(code) | |
| if suggestions: | |
| for i, suggestion in enumerate(suggestions, 1): | |
| st.write(f"{i}. {suggestion}") | |
| else: | |
| st.write("No issues detected. Your code looks good!") | |
| # Simulated CodeBERT analysis (placeholder) | |
| st.write("**Model Analysis:**") | |
| model_results = analyze_code(code) | |
| for idx, result in enumerate(model_results, 1): | |
| st.write(f"Chunk {idx} classification result: {result}") | |
| st.markdown("---") | |
| st.markdown("*Powered by Hugging Face and Streamlit*") | |