import logging logger = logging.getLogger(__name__) import json import os import re from deep_translator import GoogleTranslator from gematria import calculate_gematria import math def process_json_files(start, end, step, rounds="1", length=0, tlang="en", strip_spaces=True, strip_in_braces=True, strip_diacritics=True, translate=False): """ Processes Tripitaka JSON files and performs various text manipulations. Parameters: - start (int): Start number of the Tripitaka book. - end (int): End number of the Tripitaka book. - step (int): Step size for character selection. - rounds (str): Comma-separated list of round numbers (can include negative values). - length (int): Maximum length of the result text. - tlang (str): Target language for translation. - strip_spaces (bool): Whether to remove spaces from the text. - strip_in_braces (bool): Whether to remove text within braces. - strip_diacritics (bool): Whether to remove diacritics from the text. - translate (bool): Whether to translate the result text. Returns: - list: A list of dictionaries containing processed data or error messages. """ base_path = "texts/torah" translator = GoogleTranslator(source='auto', target=tlang) results = [] for i in range(start, end + 1): file_name = f"{base_path}/{i:02}.json" try: with open(file_name, 'r', encoding='utf-8') as file: data = json.load(file) text_blocks = data["text"] full_text = "" for block in text_blocks: full_text += ' '.join(block) clean_text = full_text if strip_in_braces: clean_text = re.sub(r"\[.*?\]", "", clean_text, flags=re.DOTALL) if strip_diacritics: clean_text = re.sub(r"[^\u05D0-\u05EA ]+", "", clean_text) if strip_spaces: clean_text = clean_text.replace(" ", "") else: # Replace multiple spaces with a single space clean_text = re.sub(r'\s+', ' ', clean_text) text_length = len(clean_text) if text_length == 0: # If after cleaning, there's no text, skip processing continue rounds_list = list(map(float, rounds.split(','))) # Allow floats result_text = "" for r in rounds_list: abs_r = abs(r) # Determine the number of full passes and the remainder. full_passes = math.floor(abs_r) remainder = abs_r - full_passes # Base chars per pass base_chars = text_length // step if base_chars == 0: if abs_r > 1: # Changed from >=1 to >1 # When step > text_length and rounds >1, pick 1 character per full pass chars_per_full_pass = 1 else: # No characters to pick chars_per_full_pass = 0 # For remainder, since base_chars=0, no remainder characters chars_for_remainder = 0 else: # Normal case chars_per_full_pass = base_chars chars_for_remainder = math.floor(base_chars * remainder) # Partial pass if r > 0: current_index = (step - 1) % text_length direction = 1 else: current_index = (text_length - step) % text_length direction = -1 pass_result = "" # Full passes, we only keep the last pass for pass_num in range(1, full_passes + 1): current_pass_chars = "" for _ in range(chars_per_full_pass): if chars_per_full_pass == 0: break current_pass_chars += clean_text[current_index] current_index = (current_index + direction * step) % text_length # Keep only the last full pass if pass_num == full_passes: pass_result = current_pass_chars # Remainder pass, for fractional rounds if remainder > 0 and chars_for_remainder > 0: current_pass_chars = "" for _ in range(chars_for_remainder): current_pass_chars += clean_text[current_index] current_index = (current_index + direction * step) % text_length pass_result += current_pass_chars # Handle cases where step > text_length and chars_per_full_pass=1 if base_chars == 0 and chars_per_full_pass == 1 and full_passes > 0: # Append the last character picked in the full passes # Since only one character is picked per full pass, pass_result is already set pass elif base_chars ==0 and chars_per_full_pass ==0 and full_passes >0: # When no characters are picked, and rounds >=1, we might want to skip pass result_text += pass_result # Only translate if needed translated_text = translator.translate(result_text) if (result_text and translate) else "" if length != 0: result_text = result_text[:length] if result_text: results.append({ "book": f"Torah {i}.", "title": data.get("title", "Unknown Title"), "result_text": result_text, "result_sum": calculate_gematria(result_text), "translated_text": translated_text, "source_language": "iw", }) except FileNotFoundError: results.append({"error": f"File {file_name} not found."}) except json.JSONDecodeError as e: results.append({"error": f"File {file_name} could not be read as JSON: {e}"}) except KeyError as e: results.append({"error": f"Expected key 'text' is missing in {file_name}: {e}"}) return results if results else None # Tests test_results = [ (process_json_files(0, 0, 21, rounds="3", length=0), "ק"), (process_json_files(0, 0, 22, rounds="1", length=0), "ת"), (process_json_files(0, 0, 22, rounds="3", length=0), "ת"), (process_json_files(0, 0, 23, rounds="3", length=0), "ג"), (process_json_files(0, 0, 11, rounds="1", length=0), "כת"), (process_json_files(0, 0, 2, rounds="1", length=0), "בדוחילנעצרת"), (process_json_files(0, 0, 23, rounds="1", length=0), None), # Expect None, when no results (process_json_files(0, 0, 23, rounds="-1", length=0), None), # Expect None, when no results (process_json_files(0, 0, 22, rounds="-1", length=0), "א"), (process_json_files(0, 0, 22, rounds="-2", length=0), "א"), (process_json_files(0, 0, 1, rounds="1,-1", length=0), "אבגדהוזחטיכלמנסעפצקרשתתשרקצפעסנמלכיטחזוהדגבא"), # Combined rounds (process_json_files(0, 0, 1, rounds="-1", length=0), "תשרקצפעסנמלכיטחזוהדגבא"), # Reversed Hebrew alphabet (process_json_files(0, 0, 1, rounds="-1.5", length=0), "תשרקצפעסנמלכיטחזוהדגבאתשרקצפעסנמל"), # Fractional rounds ] all_tests_passed = True for result, expected in test_results: result_text = result[0]['result_text'] if result else None if expected is None: # Check if no result is expected if not result: logger.warning(f"Test passed: Expected no results, got no results.") else: logger.error(f"Test failed: Expected no results, but got: {result_text}") all_tests_passed = False else: # Check if result is not empty before accessing elements if result: if result_text == expected: logger.warning(f"Test passed: Expected '{expected}', got '{result_text}'") else: logger.error(f"Test failed: Expected '{expected}', but got '{result_text}'") all_tests_passed = False else: logger.error(f"Test failed: Expected '{expected}', but got no results") all_tests_passed = False if all_tests_passed: logger.info("All round tests passed.")