dark_web_forums/Translator/translate.py

from deep_translator import GoogleTranslator


# For the arguments source and target, you can use the name of the language or its abbreviation
# chinese (simplified): zh-CN, english: en, korean: ko, persian: fa, russian: ru
def translate(input_text: str, source: str, target='english') -> str:

    if len(input_text.strip()) == 0:
        return input_text

    if source == target:
        return input_text

    batch_size = 4999
    batches = []
    whitespaces = []
    start = 0
    translated_text = ''

    while not input_text[start].strip():
        translated_text += input_text[start]
        start += 1

    # A while loop that will continue as long as start is less than the total length
    # of the HTML content (len(input_text)). This ensures that we process the entire content.
    while start < len(input_text):
        whitespace = ''
        # Set the end index for the current batch
        end = start + batch_size
        # If end is beyond the end of the content, we don't need to adjust it.
        if end < len(input_text):
            # A while loop to adjust the end index so that it doesn't split a word.
            # It continues as long as end is within the content (end < len(input_text)) and
            # the character at the end position is not a whitespace (input_text[end].strip()
            # returns True for non-whitespace characters).
            while end < len(input_text) and input_text[end].strip():
                # The inner while loop, decrements the end index by 1. This moves the end index
                # backwards until it reaches a whitespace character, ensuring that we don't split a word.
                end -= 1
            while not input_text[end].strip():
                whitespace = input_text[end] + whitespace
                end -= 1
        else:
            end = len(input_text) - 1
            while not input_text[end].strip():
                whitespace = input_text[end] + whitespace
                end -= 1

        # This line extracts a substring from the HTML content, starting from the start index
        # and ending at the end index. This substring is our batch.
        batch = input_text[start: end + 1]
        # Add our batch to the batches list
        batches.append(batch)
        whitespaces.append(whitespace)
        # Updates our start index for the next batch
        start = end + 1 + len(whitespace)

    translated_batches = GoogleTranslator(source, target).translate_batch(batches)

    for batch, whitespace in zip(translated_batches, whitespaces):
        translated_text += batch + whitespace

    return translated_text