Python try / except issue with SMTPLib

I wrote a simple SMTP client using Python SMTPLib. Just try adding some error handling - especially in this case when the target server to connect is not available (for example, wrong IP address!)

Currently the trace looks like this:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "smtplib_client.py", line 91, in runClient
    try:
  File "/usr/local/lib/python2.7/smtplib.py", line 251, in __init__
    (code, msg) = self.connect(host, port)
  File "/usr/local/lib/python2.7/smtplib.py", line 311, in connect
    self.sock = self._get_socket(host, port, self.timeout)
  File "/usr/local/lib/python2.7/smtplib.py", line 286, in _get_socket
    return socket.create_connection((host, port), timeout)
  File "/usr/local/lib/python2.7/socket.py", line 571, in create_connection
    raise err
socket.error: [Errno 111] Connection refused

      

        
        
        
      

    

It is so clear that this is "create_connection" in socket.py going bang. This has its own try / except block:

for res in getaddrinfo(host, port, 0, SOCK_STREAM):
    af, socktype, proto, canonname, sa = res
    sock = None
    try:
      sock = socket(af, socktype, proto)
      if timeout is not _GLOBAL_DEFAULT_TIMEOUT:
          sock.settimeout(timeout)
      if source_address:
          sock.bind(source_address)
      sock.connect(sa)
      return sock

    except error as _:
      err = _
      if sock is not None:
          sock.close()

if err is not None:
  raise err
else:
  raise error("getaddrinfo returns an empty list")

      

        
        
        
      

    

My runClient () function looks like this:

def runClient(configName = 'default', testFile = './test.xml'):
  cliCfg = getClientConfig(configName)
  print cliCfg.find('logfile').text
  # startLogger(cliCfg.find('logfile').text)
  clientCert = cliCfg.find('cert').text
  clientKey = cliCfg.find('key').text
  serverHost = cliCfg.find('serverhost').text
  serverPort = int(cliCfg.find('serverport').text)

  myMail = MailMessageHandler()
  msgSrc = myMail.readMessageSource(testFile)
  allMsgs = myMail.processMessages(msgSrc)
  inx = 1
  for msg in allMsgs:
    validMsg = True
    requiredKeys = ('ehlo', 'sndr', 'rcpt', 'body')
    for msgItems in requiredKeys:
      if len(msg[msgItems]) == 0:
        validMsg = False
    if validMsg:
      try:
        server = smtplib.SMTP(serverHost, serverPort)
        server.ehlo(msg['ehlo'])
        thisSender = msg['sndr']
        thisRecipient = msg['rcpt']
        thisMessage = MIMEText(msg['body'])
        thisMessage['To'] = email.utils.formataddr(('', thisRecipient))
        thisMessage['From'] = email.utils.formataddr(('', thisSender))
        thisMessage['Subject'] = msg['subject']
        thisMessage['Message-id'] = email.utils.make_msgid()
        now = datetime.now()
        day = now.strftime('%a')
        date = now.strftime('%d %b %Y %X')
        thisMessage['Date'] = day + ', ' + date + ' -0000'
        if msg['tls'].lower() == 'true':
          server.starttls('certs/client/client.key', 'certs/client/client.crt')
          logging.info ("Message: " + thisMessage['Message-id'] + " to be sent over TLS")
        server.sendmail(thisSender, thisRecipient.split(","), thisMessage.as_string())
        logging.info ("Message: " + thisMessage['Message-id'] + " sent successfully to " + serverHost + ":" + cliCfg.find('serverport').text)
        logging.info ("Message: " + thisMessage['Message-id'] + " had sender: " + thisMessage['From'])
        logging.info ("Message: " + thisMessage['Message-id'] + " had recipient(s): " + thisMessage['To'])
      except socket.error as e:
        print "Could not connect to server - is it down? ({0}): {1}".format(e.strerrror)
      except:
        print "Unknown error:", sys.exc_info()[0]
      finally:
          server.quit()
    else:
      print "Improperly formatted source mail - please check"

      

        
        
        
      

    

What I don't get is the trace shows the call raise err

. So it is clear that err

it is not None and so it must be set as part of except error as _:

.... So the error is handled first, but as part of the handler a copy ( err

) is created - which is then hoisted outside the try / except block, so it unprocessed. This unhandled error must then "go through a backup" of the call stack ( get_socket

does not have a try / except block as well as connect

or __init__

), so outside of the try / except block for the original error

in create_connection

, the copy of the error err

must necessarily "cascade" back into the try / except block in my runClient function?