app.py

import streamlit as st
import pandas as pd
import requests
from transformers import MarianMTModel, MarianTokenizer
import io

def fetch_languages(url):
    response = requests.get(url)
    if response.status_code == 200:
        # Convert bytes to a string using decode, then create a file-like object with io.StringIO
        csv_content = response.content.decode('utf-8')
        df = pd.read_csv(io.StringIO(csv_content), delimiter="|", skiprows=2, header=None).dropna(axis=1, how='all')
        df.columns = ['ISO 639-1', 'ISO 639-2', 'Language Name', 'Native Name']
        df['ISO 639-1'] = df['ISO 639-1'].str.strip()
        language_options = [(row['ISO 639-1'], f"{row['ISO 639-1']} - {row['Language Name']}") for index, row in df.iterrows()]
        return language_options
    else:
        return []

# Make sure to replace the URL with the correct one if it has changed
url = "https://huggingface.co/Lenylvt/LanguageISO/resolve/main/iso.md"
language_options = fetch_languages(url)

# Streamlit UI components
st.title("📜 Translator")
st.write("We use model from [Language Technology Research Group at the University of Helsinki](https://huggingface.co/Helsinki-NLP). For API use please visit [this space](https://huggingface.co/spaces/Lenylvt/Translator-API). 🔴 All Language are not Available")

source_language = st.selectbox("1️⃣ Select Source Language", options=language_options, format_func=lambda x: x[1])
target_language = st.selectbox("2️⃣ Select Target Language", options=language_options, format_func=lambda x: x[1])
text = st.text_area("✒️ Enter text to translate...", height=150)

def translate_text(text, source_language_code, target_language_code):
    model_name = f"Helsinki-NLP/opus-mt-{source_language_code}-{target_language_code}"
    if source_language_code == target_language_code:
        return "🔴 Translation between the same languages is not supported."
    try:
        tokenizer = MarianTokenizer.from_pretrained(model_name)
        model = MarianMTModel.from_pretrained(model_name)
        translated = model.generate(**tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512))
        translated_text = tokenizer.decode(translated[0], skip_special_tokens=True)
        return translated_text
    except Exception as e:
        return f"Failed to load model for {source_language_code} to {target_language_code}: {str(e)}"

if st.button("📁 Translate"):
    source_language_code, _ = source_language
    target_language_code, _ = target_language
    translation = translate_text(text, source_language_code, target_language_code)
    st.text_area("⬇️ Translated Text", value=translation, height=150, key="translation_output")