Spaces:
Sleeping
Sleeping
| import json | |
| import re | |
| def diff(str1, str2): | |
| indices = [] | |
| words1 = str1.split() | |
| words2 = str2.split() | |
| for index, (word1, word2) in enumerate(zip(words1, words2)): | |
| if word1 != word2: | |
| start_index = sum( | |
| len(word) + 1 for word in words1[:index] | |
| ) # Add 1 for the space between words | |
| indices.append((start_index, start_index + len(word1))) | |
| # If str2 has more words than str1, record the indices of the extra words in str2 | |
| for index in range(len(words1), len(words2)): | |
| start_index = sum( | |
| len(word) + 1 for word in words1 | |
| ) # Add 1 for the space between words | |
| indices.append((start_index, start_index + len(words2[index]))) | |
| return indices | |
| def format_predicted_sentence(original_sentence, corrected_word_indices): | |
| formatted_sentence = "" | |
| current_index = 0 | |
| for start_index, end_index in corrected_word_indices: | |
| # Add the part of the sentence before the corrected word | |
| formatted_sentence += original_sentence[current_index:start_index] | |
| # Add the corrected word wrapped in a span | |
| formatted_sentence += ( | |
| '<span style="color:red">' | |
| + original_sentence[start_index : end_index + 1] | |
| + "</span>" | |
| ) | |
| # Update the current index | |
| current_index = end_index + 1 | |
| # Add the remaining part of the sentence | |
| formatted_sentence += original_sentence[current_index:] | |
| return formatted_sentence | |
| def processInputAndResults(inputSentence, predictedSentence): | |
| corrected_word_indices = diff(inputSentence, predictedSentence) | |
| formatted_sentence = format_predicted_sentence( | |
| predictedSentence, corrected_word_indices | |
| ) | |
| return formatted_sentence | |
| def ensure_space_around_punctuation(sentence): | |
| punctuation_marks = ["।", ",", ".", "!", "?", ":", ";", '"', "'", "(", ")"] | |
| for mark in punctuation_marks: | |
| index = sentence.find(mark) | |
| while index != -1: | |
| # Ensure space before the punctuation mark | |
| if index > 0 and sentence[index - 1] != " ": | |
| sentence = sentence[:index] + " " + sentence[index:] | |
| index += 1 # Adjust index due to inserted space | |
| # Ensure space after the punctuation mark | |
| if index < len(sentence) - 1 and sentence[index + 1] != " ": | |
| sentence = sentence[: index + 1] + " " + sentence[index + 1 :] | |
| index += 1 # Adjust index due to inserted space | |
| # Find next occurrence of the punctuation mark | |
| index = sentence.find(mark, index + 1) | |
| return sentence | |
| def check_and_insert_space(sentence): | |
| index = sentence.find("।") # Find the index of "।" | |
| if index > 0 and sentence[index - 1] != " ": | |
| sentence = sentence[:index] + " " + sentence[index:] # Insert space before "।" | |
| return sentence | |
| def processPunctuation(sentence): | |
| # Define a regex pattern to match punctuation marks | |
| punctuation_pattern = r'([।,\.!\?:;"\'\(\)])' | |
| # Use re.sub() to replace punctuation marks with themselves surrounded by spaces | |
| # The pattern matches any punctuation mark and replaces it with " <punctuation_mark> " | |
| modified_sentence = re.sub(punctuation_pattern, r' \1 ', sentence) | |
| # Replace any occurrences of double spaces with single spaces | |
| modified_sentence = re.sub(r'\s+', ' ', modified_sentence) | |
| return modified_sentence |