Spaces:
Running
Running
File size: 9,101 Bytes
cf81250 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 |
import gradio as gr
import torch
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import logging
import json
import re
from typing import List, Dict
from datetime import datetime
# Setup logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
logger = logging.getLogger(__name__)
# Model configuration
MODEL_NAME = "wizcodes12/snaxfix-model"
SUPPORTED_LANGUAGES = [
"python", "javascript", "java", "c", "cpp", "csharp", "rust",
"php", "html", "css", "sql"
]
MAX_LENGTH = 512
# Example code snippets with errors for testing
EXAMPLE_SNIPPETS = {
"python": {
"broken": 'def add(a b):\n return a + b',
"description": "Missing comma in function parameters"
},
"javascript": {
"broken": 'function greet() {\n console.log("Hello"\n}',
"description": "Missing closing parenthesis and brace"
},
"java": {
"broken": 'public class Hello {\n public static void main(String[] args) {\n System.out.println("Hello World")\n }\n}',
"description": "Missing semicolon"
},
"c": {
"broken": '#include <stdio.h>\n\nint main() {\n printf("Hello World")\n return 0;\n}',
"description": "Missing semicolon"
},
"cpp": {
"broken": '#include <iostream>\n\nint main() {\n std::cout << "Hello World" << std::endl\n return 0;\n}',
"description": "Missing semicolon"
},
"csharp": {
"broken": 'class Program {\n static void Main(string[] args) {\n Console.WriteLine("Hello World")\n }\n}',
"description": "Missing semicolon"
},
"rust": {
"broken": 'fn main() {\n println!("Hello World")\n}',
"description": "Missing semicolon"
},
"php": {
"broken": '<?php\n echo "Hello World"\n?>',
"description": "Missing semicolon"
},
"html": {
"broken": '<div>\n <p>Hello World</div>\n</p>',
"description": "Incorrect tag nesting"
},
"css": {
"broken": 'body {\n background-color: #ffffff\n}',
"description": "Missing semicolon"
},
"sql": {
"broken": 'SELECT name age FROM users WHERE age > 18',
"description": "Missing comma in SELECT clause"
}
}
class SyntaxFixerApp:
def __init__(self):
logger.info("Initializing SyntaxFixerApp...")
self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
logger.info(f"Using device: {self.device}")
# Load model and tokenizer
try:
self.tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
self.model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME)
self.model.to(self.device)
self.model.eval()
logger.info("Model and tokenizer loaded successfully")
except Exception as e:
logger.error(f"Failed to load model: {e}")
raise
# Initialize history
self.history = []
def fix_syntax(self, broken_code: str, language: str) -> str:
"""Fix syntax errors in the provided code."""
if not broken_code.strip():
return "Error: Please enter code to fix."
if language not in SUPPORTED_LANGUAGES:
return f"Error: Language '{language}' is not supported. Choose from: {', '.join(SUPPORTED_LANGUAGES)}"
try:
# Prepare input
input_text = f"<{language.upper()}> Fix the syntax errors in this {language} code: {broken_code}"
inputs = self.tokenizer(
input_text,
max_length=MAX_LENGTH,
truncation=True,
padding=True,
return_tensors="pt"
).to(self.device)
# Generate fixed code
with torch.no_grad():
outputs = self.model.generate(
**inputs,
max_length=MAX_LENGTH,
num_beams=4,
early_stopping=True,
do_sample=False,
temperature=0.7,
pad_token_id=self.tokenizer.pad_token_id,
use_cache=True
)
fixed_code = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
# Save to history
self.history.append({
"timestamp": datetime.now().isoformat(),
"language": language,
"broken_code": broken_code,
"fixed_code": fixed_code
})
return fixed_code
except Exception as e:
logger.error(f"Error fixing code: {e}")
return f"Error: Failed to fix code - {str(e)}"
def load_example(self, language: str) -> str:
"""Load example broken code for the selected language."""
return EXAMPLE_SNIPPETS.get(language, {}).get("broken", "No example available for this language.")
def get_history(self) -> str:
"""Return formatted history of fixes."""
if not self.history:
return "No history available."
history_text = "=== Fix History ===\n"
for entry in self.history:
history_text += f"Timestamp: {entry['timestamp']}\n"
history_text += f"Language: {entry['language']}\n"
history_text += f"Broken Code:\n{entry['broken_code']}\n"
history_text += f"Fixed Code:\n{entry['fixed_code']}\n"
history_text += "-" * 50 + "\n"
return history_text
def clear_history(self) -> str:
"""Clear the history of fixes."""
self.history = []
return "History cleared."
def create_gradio_interface():
"""Create and return the Gradio interface."""
app = SyntaxFixerApp()
with gr.Blocks(theme=gr.themes.Soft()) as demo:
gr.Markdown("# SnaxFix: Advanced Syntax Error Fixer")
gr.Markdown("Fix syntax errors in code across multiple programming languages using the `snaxfix-model` by wizcodes12.")
with gr.Row():
with gr.Column(scale=2):
language_dropdown = gr.Dropdown(
choices=SUPPORTED_LANGUAGES,
label="Select Programming Language",
value="python"
)
code_input = gr.Code(
label="Enter Code with Syntax Errors",
lines=10,
language="python"
)
with gr.Row():
fix_button = gr.Button("Fix Syntax", variant="primary")
example_button = gr.Button("Load Example", variant="secondary")
clear_button = gr.Button("Clear Input", variant="secondary")
with gr.Column(scale=2):
code_output = gr.Code(
label="Fixed Code",
lines=10,
language="python"
)
with gr.Accordion("History of Fixes", open=False):
history_output = gr.Textbox(label="Fix History", lines=10)
clear_history_button = gr.Button("Clear History")
with gr.Accordion("About & License", open=False):
gr.Markdown("""
**About SnaxFix**
SnaxFix is an AI-powered tool for fixing syntax errors in multiple programming languages, built with `google/flan-t5-base` and fine-tuned by wizcodes12.
**MIT License**
This project is licensed under the MIT License. See the [LICENSE](https://github.com/wizcodes12/snaxfix-model/blob/main/LICENSE) file for details.
""")
# Event handlers
def update_code_language(language):
return gr.update(language=language)
fix_button.click(
fn=app.fix_syntax,
inputs=[code_input, language_dropdown],
outputs=code_output
)
example_button.click(
fn=app.load_example,
inputs=language_dropdown,
outputs=code_input
)
clear_button.click(
fn=lambda: "",
inputs=None,
outputs=code_input
)
language_dropdown.change(
fn=update_code_language,
4
outputs=code_input
)
clear_history_button.click(
fn=app.clear_history,
inputs=None,
outputs=history_output
)
fix_button.click(
fn=app.get_history,
inputs=None,
outputs=history_output,
_js="() => {return []}" # Ensure history updates after fixing
)
return demo
if __name__ == "__main__":
logger.info("Starting Gradio application...")
demo = create_gradio_interface()
try:
demo.launch()
except Exception as e:
logger.error(f"Failed to launch Gradio app: {e}")
raise |