app.py

from huggingface_hub import InferenceClient
import gradio as gr

import re

# Define regex patterns for comments
java_single_line_comment_regex = r"\/\/.*"
java_multiline_comment_regex = r"\/\*(?:[^*]|\*(?!\/))*\*\/"
kotlin_single_line_comment_regex = r"\/\/.*"
kotlin_multiline_comment_regex = r"\/\*(?:[^*]|\*(?!\/))*\*\/"


def remove_comments(content,file_type):
    """
    Opens a Java or Kotlin file, removes comments, and saves the changes.

    Args:
        file_path: The path to the Java or Kotlin file.
    """


    # Determine file type based on extension
    if file_type == "java":
        pattern = java_single_line_comment_regex + "|" + java_multiline_comment_regex
    elif file_type == "kotlin":
        pattern = kotlin_single_line_comment_regex + "|" + kotlin_multiline_comment_regex
    else:
        raise ValueError(f"Unsupported file type: {file_type}")

    # Remove comments using regex
    clean_content = re.sub(pattern, "", content, )
    return clean_content
    
client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")

def format_prompt(message, history):
  prompt = "<s>"
  for user_prompt, bot_response in history:
    prompt += f"[INST] {user_prompt} [/INST]"
    prompt += f" {bot_response}</s> "
  prompt = f"[INST] {message} [/INST]"
  return prompt

def generate(
    prompt, history, temperature=0.2, max_new_tokens=16392, top_p=0.95, repetition_penalty=1.0,
):
    temperature = float(0)
    if temperature < 1e-2:
        temperature = 1e-2
    top_p = float(top_p)

    generate_kwargs = dict(
        temperature=temperature,
        max_new_tokens=max_new_tokens,
        top_p=top_p,
        repetition_penalty=repetition_penalty,
        do_sample=True,
        seed=42,
    )
    prompt = f"""Translate the given Kotlin code to Java, adhering to the following constraints:
Preserve the original names of classes, fields, and methods without renaming.
{remove_comments(
    content=prompt,file_type='kotlin'
).strip()}"""
    formatted_prompt = format_prompt(prompt, history)

    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
    output = ""

    for response in stream:
        output += response.token.text
        yield output
    return output

    
mychatbot = gr.Chatbot(
    avatar_images=["./user.png", "./botm.png"], bubble_full_width=False, show_label=False, show_copy_button=True, likeable=True,)

demo = gr.ChatInterface(fn=generate, 
                        chatbot=mychatbot,
                        title="Mixtral 8x7b Chat For Kotlin Translation",
                        retry_btn=None,
                        undo_btn=None
                       )

demo.queue().launch(show_api=False)