build_conversation_turn2.py

def build_conversation_turn(content, message, headers):
    #...
    # Iterate over attachments
    if "attachments" in message:
        for file in message["attachments"]:
          # Isolate name of the file and remove characters before the final period
            file_name = file["name"] 
            file_extension = file["name"].split(".")[-1]
            
            # File is a supported type
            file_url = file["contentUrl"]

            # Download the file from the content URL. Include headers in case the file is protected (SharePoint)
            file_content = download_file_for_user(file_name, file_url, headers)
            
            ### Bedrock API has strict file name reqs: document file name can only contain alphanumeric characters, whitespace characters, hyphens, parentheses, and square brackets. The name can't contain more than one consecutive whitespace character.
            # Remove disallowed
            file_name = re.sub(r"[^a-zA-Z0-9\s\-\[\]\(\)]", "", file_name)
            # Only single spaces allowed
            file_name = re.sub(r"\s{2,}", " ", file_name)
            # Strip leading and trailing whitespace
            file_name = file_name.strip()