任意の Slack チャンネルにメールを飛ばすための Lambda スクリプト

lambda_email_to_slack.md

任意のチャンネルでメールを受信できるようにするための AWS Lambda(Python)　スクリプトです。

説明リンク: TBD

main.py

import os
import json
import email
from email.iterators import typed_subpart_iterator
from email.header import decode_header
from html.parser import HTMLParser
from base64 import urlsafe_b64decode
from urllib.parse import urlencode
from urllib.request import Request, urlopen

import boto3

BUCKET_NAME = os.environ.get('BUCKET_NAME')
SLACK_WEBHOOK = os.environ.get('SLACK_WEBHOOK')
DEFAULT_ENCODING = 'ascii'

s3 = boto3.resource('s3')
bucket = s3.Bucket(BUCKET_NAME)
        
    
def parse_mail(raw_mail):
    msg = email.message_from_bytes(raw_mail)
    header = decode_header(msg.get('Subject'))

    return {
        'body': get_body(msg),
        'from': get_header(msg['From']),
        'subject': get_header(msg['subject']),
    }

  
def post_slack(params):
    data = json.dumps(params)
    headers = {"Content-Type" : "application/json"}
    req = Request(SLACK_WEBHOOK, data=data.encode('utf-8'), headers=headers, method='POST')
    res = urlopen(req)
    return res
  
  
def get_charset(message, default=DEFAULT_ENCODING):
    if message.get_content_charset():
        return message.get_content_charset()
    if message.get_charset():
        return message.get_charset()
    return default

  
def get_header(header_text):
    headers = decode_header(header_text)
    header_sections = []
    for text, charset in headers:
        if type(text) is str:
            header_sections.append(text)
        else:
            header_sections.append(text.decode(charset or DEFAULT_ENCODING))
    return "".join(header_sections)
  
  
def get_body(message):
    if message.is_multipart():
        body = []
        # text/plain 部があったら優先して返す
        for part in typed_subpart_iterator(message, 'text', 'plain'):
            charset = get_charset(part, get_charset(message))
            body.append(part.get_payload(decode=True).decode(charset))
        text = u"\n".join(body).strip()
        if text:
            return text
        
        # text/html しかなければタグを剥がして返す
        for part in typed_subpart_iterator(message, 'text', 'html'):
            charset = get_charset(part, get_charset(message))
            body.append(part.get_payload(decode=True).decode(charset))
        return strip_tags(u"\n".join(body).strip())

    body = message.get_payload(decode=True).decode(get_charset(message))
    return strip_tags(body)
    
    
class TagStripper(HTMLParser):

    skip_tags = ['style']  # このタグで囲われた部分は無視
    box_tags = ['h1', 'h2', 'h3', 'h4', 'dir', 'p', 'tr', 'br']
    
    def __init__(self):
        super().__init__()
        self.reset()
        self.fed = []
        self.skip = False

    def handle_data(self, d):
        if self.skip:
            return
        self.fed.append(d.strip())

    def handle_starttag(self, tag, attrs):
        if tag in self.skip_tags:
            self.skip = True

    def handle_endtag(self, tag):
        if tag in self.box_tags:
            self.fed.append('\n')
        if tag in self.skip_tags:
            self.skip = False

    def stripped(self):
        return ''.join(self.fed).strip()
      

def strip_tags(html):
    s = TagStripper()
    s.feed(html)
    res = s.stripped()
    return res.replace('`', '\`')
    
    
def lambda_handler(event, context):
  
    for mail in event['Records']:
        if 'ses' not in mail:
            continue
        obj_id = mail['ses']['mail']['messageId']
        obj = bucket.Object(f'{obj_id}')
        raw_mail = obj.get()['Body'].read()
        msg = parse_mail(raw_mail)
        receipt = mail['ses']['receipt']['recipients'][0]
        channel = '#' + receipt.split('@')[0]
        res = post_slack({
            'channel': channel,
            'username': mail['ses']['mail']['commonHeaders']['from'][0],
            'icon_emoji': ':mailbox:',
            'text': f"""件名: {msg['subject']}\n```{msg['body']}```"""
        })
        print(res.__dict__)