import json import logging import datetime import time import requests import pytz import unittest import sqlite3 # Import sqlite3 for database handling import gradio as gr from deep_translator import GoogleTranslator from deep_translator.exceptions import NotValidLength, RequestError from utils import process_json_files, flatten_text_with_line_breaks, calculate_tanach_statistics, build_word_index # Set up logging logging.basicConfig(level=logging.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s') # Load Tanach text TANACH_DATA = process_json_files(1, 39) WORD_INDEX = build_word_index(TANACH_DATA) # --- Database Setup --- # Use a connection function to ensure each thread gets its own connection def get_db_connection(): conn = sqlite3.connect('translation_cache.db') conn.row_factory = sqlite3.Row # This line allows accessing columns by name return conn # Create the database table if it doesn't exist with get_db_connection() as conn: cursor = conn.cursor() cursor.execute(''' CREATE TABLE IF NOT EXISTS translations ( book_id INTEGER, chapter_id INTEGER, english_text TEXT, PRIMARY KEY (book_id, chapter_id) ) ''') conn.commit() # --- Utility Functions --- def get_current_word_data(client_time_str): """Gets data about the current word based on the client's time.""" try: client_time = datetime.datetime.strptime(client_time_str, "%H:%M:%S") total_seconds = int(client_time.strftime("%H")) * 3600 + \ int(client_time.strftime("%M")) * 60 + \ int(client_time.strftime("%S")) # Find the closest key in WORD_INDEX word_position = min(WORD_INDEX.keys(), key=lambda k: abs(k - total_seconds)) return WORD_INDEX[word_position], word_position except Exception as e: logging.error(f"Error processing client time: {e}") return None, None def get_formatted_verse(book_id, chapter_id, verse_id, highlight_word=True): """Returns a formatted verse with optional word highlighting.""" chapter_text = TANACH_DATA[book_id]["text"][chapter_id] flattened_chapter = flatten_text_with_line_breaks(chapter_text) return '
'.join(flattened_chapter) def translate_chapter(hebrew_chapter, book_id, chapter_id): """Translates a Hebrew chapter to English, caching the result in the database.""" # Get a new database connection for this thread with get_db_connection() as conn: cursor = conn.cursor() # Check if translation exists in the database cursor.execute( "SELECT english_text FROM translations WHERE book_id=? AND chapter_id=?", (book_id, chapter_id) ) result = cursor.fetchone() if result: return result['english_text'].split('\n') # Retrieve from database and split into lines try: translator = GoogleTranslator(source='iw', target='en') max_length = 2000 # Slightly below the limit to be safe translated_text = "" # Split the chapter into chunks smaller than the max length chunks = [hebrew_chapter[i:i + max_length] for i in range(0, len(hebrew_chapter), max_length)] for chunk in chunks: translated_text += translator.translate(chunk) # Store the translation in the database cursor.execute( "INSERT INTO translations (book_id, chapter_id, english_text) VALUES (?, ?, ?)", (book_id, chapter_id, translated_text) ) conn.commit() return translated_text.split('\n') # Return as list of lines except RequestError as e: logging.warning(f"Translation failed: Request Error - {e}") return ["Translation unavailable: Request Error"] def update_tanach_display(client_time_str, timezone): """Updates the Gradio interface with client time, verse info, and translations.""" try: # Get timezone offset using pytz tz = pytz.timezone(timezone) offset = tz.utcoffset(datetime.datetime.now()).total_seconds() / 3600 # Offset in hours # Adjust client time based on the timezone client_time_obj = datetime.datetime.strptime(client_time_str, "%H:%M:%S") client_time_obj = client_time_obj.replace(tzinfo=datetime.timezone(datetime.timedelta(hours=offset))) client_time_str = client_time_obj.strftime("%H:%M:%S") except Exception as e: logging.error(f"Error adjusting client time based on timezone: {e}") return "Error: Invalid Timezone", "", "" word_data, word_position = get_current_word_data(client_time_str) if word_data is None: logging.error(f"Word position {word_position} not found in index.") return "Error: Word not found", "", "" book_id = word_data["book_id"] chapter_id = word_data["chapter_id"] verse_id = word_data["verse_id"] logging.debug(f"Book ID: {book_id}, Chapter ID: {chapter_id}, Verse ID: {verse_id}") # Format verse information verse_info = f""" **{TANACH_DATA[book_id]['title']}** Chapter {chapter_id + 1}, Verse {verse_id} """ # Get and format Hebrew and English verses hebrew_verse = get_formatted_verse(book_id, chapter_id, verse_id, highlight_word=False) hebrew_verses = hebrew_verse.split("
") # Translate the entire chapter and get the correct verse hebrew_chapter = flatten_text_with_line_breaks(TANACH_DATA[book_id]["text"][chapter_id]) english_chapter = translate_chapter('\n'.join(hebrew_chapter), book_id, chapter_id) # Highlight the current verse in both Hebrew and English hebrew_verses[verse_id - 1] = f"{hebrew_verses[verse_id - 1]}" english_chapter[verse_id - 1] = f"{english_chapter[verse_id - 1]}" # Join the verses back with
for display hebrew_verse = "
".join(hebrew_verses) english_verse = "
".join(english_chapter) return verse_info, hebrew_verse, english_verse # --- Gradio Interface --- with gr.Blocks(css=""" .container { display: flex; flex-direction: column; align-items: center; font-family: 'Times New Roman', serif; } /* Add this highlight class styling */ .highlight { background-color: #FFFF00; /* Yellow highlight */ padding: 2px 5px; border-radius: 5px; } #verse-info { margin-bottom: 20px; text-align: center; } #verses { display: flex; flex-direction: row; justify-content: center; align-items: flex-start; gap: 50px; } #hebrew-verse { font-size: 18px; line-height: 1.5; margin-bottom: 20px; text-align: right; direction: rtl; } #english-verse { font-size: 18px; line-height: 1.5; margin-bottom: 20px; } """) as iface: with gr.Row(): timezone_input = gr.Dropdown( choices=[tz for tz in pytz.common_timezones], label="Select Your Timezone", value="UTC" # Set a default timezone ) with gr.Row(): advance_button = gr.Button("Advance to Current Time") with gr.Row(): verse_info_output = gr.Markdown(label="Verse Information", elem_id="verse-info") # Place Hebrew and English verses within a flex container with gr.Row(elem_id="verses"): hebrew_verse_output = gr.HTML(label="Hebrew Verse", elem_id="hebrew-verse") english_verse_output = gr.HTML(label="English Translation", elem_id="english-verse") # Update the display with verse information and translations when the button is clicked advance_button.click( fn=lambda tz: update_tanach_display(datetime.datetime.now(pytz.timezone(tz)).strftime("%H:%M:%S"), tz), inputs=[timezone_input], outputs=[verse_info_output, hebrew_verse_output, english_verse_output], ) class TestWordIndex(unittest.TestCase): def test_word_index_boundaries(self): # Test for 0:00:00 word_data_start, _ = get_current_word_data("00:00:00") self.assertEqual(word_data_start["book_id"], 1) self.assertEqual(word_data_start["chapter_id"], 0) self.assertEqual(word_data_start["verse_id"], 1) # Test for 23:59:59 word_data_end, _ = get_current_word_data("23:59:59") self.assertEqual(word_data_end["book_id"], 39) self.assertEqual(word_data_end["chapter_id"], 35) self.assertEqual(word_data_end["verse_id"], 23) if __name__ == '__main__': # Run tests first suite = unittest.TestLoader().loadTestsFromTestCase(TestWordIndex) unittest.TextTestRunner().run(suite) iface.launch(share=True)