fix-outlook-quote.py

#!/usr/bin/python3

from email.parser import Parser
import re
import sys

known_quote_header_marks = (
    "-----Original Message-----",
    "_____________________________________________",
    "-----Original Appointment-----",
)

headerre = re.compile(r"^(\w+) ?: (.*)$")
from_words = ("From", "Von", "De")
sent_words = ("Sent", "Gesendet", "Envoyé")

def detect_quote_header(lines):
    #print(lines)
    if len(lines) < 4:
        return None

    m = headerre.match(lines[0])
    if m is not None and m.group(1) in from_words:
        idx = 0
    elif lines[0].strip() not in known_quote_header_marks:
        return None
    else:
        idx = 1

    if len(lines[idx].strip()) == 0:
        idx += 1
        
    m = headerre.match(lines[idx])
    if m is None or m.group(1) not in from_words:
        return None

    sender = m.group(2)

    if len(lines[idx].strip()) == 0:
        idx += 1
        
    m = headerre.match(lines[idx+1])
    if m is None or m.group(1) not in sent_words:
        return None

    date = m.group(2)

    end_idx = None
    for i in range(idx+1, len(lines)):
        if len(lines[i].strip()) == 0:
            end_idx = i
            break

    if end_idx is None:
        return None

    for i in range(end_idx, len(lines)):
        if len(lines[i].strip()) != 0:
            end_idx = i
            break

    return (sender.strip(), date.strip(), end_idx)

def get_message_text(message):
    charset = 'ascii'
    for key, val in message.get_params():
        if key == 'charset':
            charset = val
    return (charset, message.get_payload(decode=True).decode(val))

MAX_LEVEL = 5

def requote(lines, level):
    message_lines = []
    text = ""
    
    for i in range(0, len(lines)):
        info = detect_quote_header(lines[i:i+10])
        if info is None:
            message_lines.append(lines[i])
        else:
            sender, date, end_idx = info

            text += ">" * level + " On {} {} wrote:\n".format(date, sender)
            if level <= MAX_LEVEL:
                text += requote(lines[i+end_idx:], level + 1)
            else:
                for line in lines[i+end_idx:]:
                    text += ">" * (level + 1) + " " + line + "\n"
            break

    for line in message_lines:
        text += ">" * level + " " + line + "\n"
    return text
    
def retrieve_text_from_multipart(message):
    charset = None
    text = None
    for part in message.get_payload():
        #print(part.get_content_type(), part.items())
        if part.get_content_type() == 'text/plain' and 'Content-Disposition' not in part:
            charset, text = get_message_text(part)
        elif part.get_content_type() == 'multipart/alternative':
            charset, text = retrieve_text_from_multipart(part)
    return (charset, text)
        
def main():
    parser = Parser()
    message = parser.parse(sys.stdin)
    text = None
    if message.is_multipart():
        charset, text = retrieve_text_from_multipart(message)
    else:
        charset, text = get_message_text(message)

    if text is None:
        print("> *** No usable payload ***")
        return

    lines = text.split('\n')

    text = requote(lines, 1)
    sys.stdout.buffer.write(text.encode('utf-8'))

if __name__ == '__main__':
    main()