Spaces:
Running
Running
| import logging | |
| import json | |
| import re | |
| import math | |
| import glob | |
| from deep_translator import GoogleTranslator | |
| from gematria import calculate_gematria # Ensure this function is correctly defined | |
| import csv | |
| # Configure the logger | |
| # You can uncomment the next line to enable debugging logs | |
| # logging.basicConfig(level=logging.DEBUG, format='%(levelname)s:%(message)s') | |
| logger = logging.getLogger(__name__) | |
| def process_json_files(start=1, end=52, step=1, rounds="1,-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/tripitaka" | |
| translator = GoogleTranslator(source='sa', target=tlang) if translate else None # Correct source language for Sanskrit | |
| results = [] | |
| for i in range(start, end + 1): | |
| file_pattern = f"{base_path}/{i:02}*.json" # Adjust to Tripitaka file naming | |
| matched_files = glob.glob(file_pattern) | |
| if not matched_files: | |
| logger.warning(f"Mandala {i}: No files found for pattern {file_pattern}.") | |
| results.append({"error": f"Mandala {i}: No files found for pattern {file_pattern}."}) | |
| continue | |
| for file_name in matched_files: | |
| try: | |
| with open(file_name, 'r', encoding='utf-8') as file: | |
| data = json.load(file) | |
| full_text = "" | |
| for gatha in data.get("gathas", []): # Safe access to 'gathas' | |
| full_text += gatha + " " | |
| clean_text = full_text | |
| if strip_in_braces: | |
| # Remove content within [], {}, <> brackets | |
| clean_text = re.sub(r"\[.*?\]|\{.*?\}|\<.*?\>", "", clean_text, flags=re.DOTALL) | |
| if strip_diacritics: | |
| # Keep only Devanagari characters and spaces | |
| clean_text = re.sub(r'[^\u0900-\u097F\s]', '', clean_text) | |
| # Remove specific diacritics and punctuation | |
| clean_text = re.sub(r'[\u0951-\u0954\u0964\u0965]+', '', clean_text) | |
| # Remove Devanagari numerals | |
| clean_text = re.sub(r'[०१२३४५६७८९]+', '', clean_text) | |
| # Remove additional unwanted characters in one step | |
| clean_text = re.sub(r'[:?!\'\-]', '', clean_text) | |
| # Normalize spaces | |
| clean_text = clean_text.replace("\n\n ", " ") | |
| clean_text = clean_text.replace("\n", " ") | |
| clean_text = re.sub(r'\s+', ' ', clean_text).strip() | |
| if strip_spaces: | |
| clean_text = clean_text.replace(" ", "") | |
| text_length = len(clean_text) | |
| logger.debug(f"Mandala {i}: Clean text length = {text_length}") | |
| if text_length == 0: | |
| logger.warning(f"Mandala {i}: No text available after cleaning.") | |
| continue # Skip processing if no text is available | |
| # Parse 'rounds', allow floats | |
| try: | |
| rounds_list = list(map(float, rounds.split(','))) | |
| except ValueError as e: | |
| logger.error(f"Mandala {i}: Invalid 'rounds' format: {rounds}") | |
| results.append({"error": f"Mandala {i}: Invalid 'rounds' format: {rounds}"}) | |
| continue # Skip this Mandala due to invalid 'rounds' input | |
| 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 number of characters 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 | |
| logger.debug(f"Mandala {i}: step > text_length ({step} > {text_length}), selecting 1 character per full pass.") | |
| else: | |
| # No characters to pick | |
| chars_per_full_pass = 0 | |
| logger.debug(f"Mandala {i}: step > text_length ({step} > {text_length}) and rounds <=1, no characters selected.") | |
| # 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 | |
| logger.debug(f"Mandala {i}: Normal case, chars_per_full_pass = {chars_per_full_pass}, chars_for_remainder = {chars_for_remainder}") | |
| 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, keep only 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 | |
| logger.debug(f"Mandala {i}: Pass {pass_num}, pass_result = {pass_result}") | |
| # 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 | |
| logger.debug(f"Mandala {i}: Remainder pass_result = {pass_result}") | |
| # 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: | |
| # pass_result already contains the last character picked | |
| pass | |
| elif base_chars == 0 and chars_per_full_pass == 0 and full_passes > 0: | |
| # When no characters are picked, skip appending | |
| pass | |
| result_text += pass_result | |
| logger.debug(f"Result text for Mandala {i}: {result_text}") | |
| if length != 0: | |
| result_text = result_text[:length] | |
| logger.debug(f"Mandala {i}: Result text truncated to length {length}.") | |
| # Translate the result text if required | |
| try: | |
| translated_text = translator.translate(result_text) if translator and result_text else "" | |
| except Exception as e: | |
| logger.error(f"Translation error: {e}") | |
| translated_text = "" | |
| # Calculate the Gematria sum | |
| try: | |
| result_sum = calculate_gematria(result_text) | |
| except Exception as e: | |
| logger.error(f"Mandala {i}: Gematria calculation error: {e}") | |
| result_sum = None | |
| if result_text: | |
| result = { | |
| 'book': f"Tripitaka {i}.", | |
| 'title': f'{data.get("title", "Unknown Title")} {data.get("book_name", "Unknown Book")} {data.get("chapter", "")}', | |
| 'result_text': result_text, | |
| 'result_sum': result_sum, | |
| 'translated_text': translated_text, | |
| 'source_language': "sa", | |
| } | |
| results.append(result) | |
| except (FileNotFoundError, json.JSONDecodeError, KeyError) as e: | |
| logger.error(f"Mandala {i}: Error processing {file_name}: {e}") | |
| results.append({"error": f"Mandala {i}: Error processing {file_name}: {e}"}) | |
| except Exception as e: | |
| logger.error(f"Mandala {i}: Unexpected error processing {file_name}: {e}") | |
| results.append({"error": f"Mandala {i}: Unexpected error processing {file_name}: {e}"}) | |
| return results if results else None | |
| # Tests | |
| test_results = [ | |
| (process_json_files(1,1,386,"1,-1", translate=True),"अअमुोच्टेकोसचण") | |
| ] | |
| # Function to run tests | |
| def run_tests(test_results): | |
| all_tests_passed = True | |
| for idx, (result, expected) in enumerate(test_results): | |
| if expected is None: # Check if no result is expected | |
| if not result: | |
| logger.warning(f"Test {idx}: Passed (Expected no results, got no results.)") | |
| else: | |
| # Extract result_text from the first result | |
| result_text = result[0].get('result_text', None) | |
| logger.error(f"Test {idx}: Failed (Expected no results, but got: {result_text})") | |
| all_tests_passed = False | |
| else: | |
| # Check if result is not empty before accessing elements | |
| if result: | |
| result_text = result[0].get("result_text") | |
| if result_text == expected: | |
| logger.warning(f"Test {idx}: Passed (Expected: '{expected}', Got: '{result_text}')") | |
| else: | |
| logger.error(f"Test {idx}: Failed (Expected: '{expected}', but got: '{result_text}')") | |
| all_tests_passed = False | |
| else: | |
| logger.error(f"Test {idx}: Failed (Expected: '{expected}', but got no results)") | |
| all_tests_passed = False | |
| if all_tests_passed: | |
| logger.info("All round tests passed.") | |
| else: | |
| logger.info("Some tests failed. Please check the error messages above.") | |
| # Run the tests | |
| run_tests(test_results) | |