This really simple script written in Python that send by email job offers published on a RSS feed. (I left the RSS feed that work well with the name of fields I put) It runs every 5 minutes. It downloads a RSS feed with feedparser, put the content in a simple object named offer, track new offers and save them in a pickle file, send an email with…

db.py

from offer import offer
import psycopg2

class dbOffer:

	def __init__(self,sourcename,host="localhost",port="5432",database="rsstracker",user="rsstracker",password="rsstracker"):
		psycopg2.extensions.register_type(psycopg2.extensions.UNICODE)
		psycopg2.extensions.register_type(psycopg2.extensions.UNICODEARRAY)
		self.conn = psycopg2.connect(host=host,port=port,database=database,user=user,password=password)
		self.sourcename = sourcename

	def prepareOffer(self,offer):
		return (offer.link,offer.description,offer.date,offer.sent_message_id,offer.reply_message_id,offer.deleted,self.sourcename,offer.title)	

	def convertToOffer(self,sqlArray):
		o = offer(sqlArray[1],sqlArray[2],sqlArray[3],sqlArray[4])
		o.sent_message_id = sqlArray[5]
		o.reply_message_id = sqlArray[6]
		o.deleted = sqlArray[7]
		return o

	def insertOfferList(self,offerList):
		sqlList = []
		for offer in offerList:
			sqlList.append(self.prepareOffer(offer))
		cur = self.conn.cursor()
		cur.executemany("INSERT INTO offers (link,description,date,sent_message_id,reply_message_id,deleted,source,title) VALUES (%s,%s,%s,%s,%s,%s,%s,%s)",sqlList)
		self.conn.commit()

	def updateOfferList(self,offerList):
		sqlList = []
		for offer in offerList:
			sqlList.append(self.prepareOffer(offer))
		cur = self.conn.cursor()
		cur.executemany("UPDATE offers SET link=%s, description=%s, date=%s, sent_message_id=%s, reply_message_id=%s, deleted=%s WHERE source=%s AND title=%s",sqlList)
		self.conn.commit()

	def getValidOfferList(self):
		cur = self.conn.cursor()
		cur.execute("SELECT * FROM offers WHERE source=%s AND deleted IS NOT TRUE",(self.sourcename,))
		sqlArray = cur.fetchall()
		offerList = []
		for array in sqlArray:
			offerList.append(self.convertToOffer(array))
		return offerList
		
		
		

init.sql

CREATE DATABASE rsstracker
  WITH ENCODING='UTF8'
       CONNECTION LIMIT=-1;
CREATE ROLE rsstracker LOGIN ENCRYPTED PASSWORD 'md530f6cd783e03a6ab3e9e3a601400cbfa'
   VALID UNTIL 'infinity';
CREATE ROLE rsstracker_group
   VALID UNTIL 'infinity';
ALTER DEFAULT PRIVILEGES 
    GRANT INSERT, SELECT, UPDATE, DELETE, TRUNCATE, REFERENCES, TRIGGER ON TABLES
    TO rsstracker_group;
ALTER DEFAULT PRIVILEGES 
    GRANT SELECT, UPDATE, USAGE ON SEQUENCES
    TO rsstracker_group;
ALTER DEFAULT PRIVILEGES 
    GRANT EXECUTE ON FUNCTIONS
    TO rsstracker_group;
CREATE TABLE offers
(
	source varchar(10),
	title varchar(100),
	link varchar(200),
	description text,
	date varchar(50),
	sent_message_id varchar(50),
	reply_message_id varchar(50),
	deleted boolean,
	primary key (source,title)
)

job offer rss mail.py

import feedparser
import time,datetime
from offer import offer
from db import dbOffer
from listOffers import listOffers
from mailOffer import mailOffer
#import pprint

if __name__ == '__main__':
    # Url for the RSS feed
    rss_url = "https://careers.societegenerale.com/groupe/fr/rss-offre.html?0&zone=83&lang=fr&JOBTYPE_VIE=34"

    # Setting up objects
    dbo = dbOffer('socgen')
    lo = listOffers(dbo)
    mo = mailOffer(email_from  = '',
                   email_to    = '',
                   server_smtp = '',
                   login       = '',
                   password    = '',
                   subject_keyword = 'SG')

    while(True):
        # Getting information from the web
        print datetime.datetime.now()
        feed = feedparser.parse(rss_url)
        
        if len(feed['items']) > 0: # Avoid resending ALL the offer when feed is empty.
            # Traking changes
            for item in feed['items']:
                #pprint.pprint(item)
                o = offer(item['title'],item['link'],item['description'],item['updated'])
                lo.trackOffer(o)
    
            # Simple output
            print 'list',len(lo.listOffers)
            print 'new',len(lo.newOffers)
            print 'del',len(lo.delOffers)
    
            # Sending new offers by email
            for item in lo.newOffers:
                mo.sendOffer(item)
                time.sleep(2) # Wait in order not to flood STMP server and/or the recipient mail server
                pass
            # Sending del offers by email
            for item in lo.delOffers:
                mo.delOffer(item)
                time.sleep(2)
                pass

            
            # Reseting the traking and saving 
            lo.saveList()
            lo.resetTracking()
    
        # Waiting
        print
        time.sleep(300) # every 5 minutes = 300 seconds

listOffers.py

import cPickle as pickle
from db import dbOffer

class listOffers:
    ''' listOffers tracks if there are new offers and deleted ones. It also keep track of the list in a pickle file '''
    def __init__(self,db):
        self.db = db
        self.listOffers = None
        self.newOffers = []
        self.delOffers = []
        self.loadList()
        self.resetTracking()
    def loadList(self):
        self.listOffers = self.db.getValidOfferList()
    def saveList(self):
        self.db.insertOfferList(self.newOffers)
        for o in self.delOffers:
            o.deleted = True
        self.db.updateOfferList(self.delOffers)
    def resetTracking(self):
        self.listOffers.extend(self.newOffers)
        for off in self.delOffers:
            self.listOffers.remove(off)
        self.newOffers = []
        self.delOffers = [item for item in self.listOffers]
    def trackOffer(self,offer):
        try:
            self.delOffers.remove(offer)
        except ValueError:
            self.newOffers.append(offer)

mailOffer.py

import smtplib
import pprint
from bs4 import BeautifulSoup
from email.utils import make_msgid
from email.mime.text import MIMEText
from email.mime.multipart import MIMEMultipart
from email import charset
from offer import offer


class mailOffer:
    ''' mailOffer set up everything to be able to send offers by email '''
    def __init__(self, email_from, email_to, server_smtp, login, password, subject_keyword):
        self.email_from = email_from
        self.email_to = email_to
        self.server_smtp = server_smtp
        self.login = login
        self.password = password
        self.subject_keyword = subject_keyword
        charset.add_charset('utf-8', charset.SHORTEST, charset.QP)
    def sendOffer(self,o):
        # Create message container - the correct MIME type is multipart/alternative.
        msg = MIMEMultipart('alternative')
        msg['Subject'] = "[VIE "+self.subject_keyword+" "+o.date+"] "+o.title
        msg['From'] = self.email_from
        msg['To'] = self.email_to
        if o.sent_message_id is None:
            o.sent_message_id = make_msgid();
        msg['Message-ID'] = o.sent_message_id;
        print msg['Message-ID']
        
        # Create the body of the message (a plain-text and an HTML version).
        text = o.link + "\n\n" + BeautifulSoup(o.description).get_text()
        html = """\
        <html>
          <head></head>
          <body>
            <p>"""+o.link+"""</p><br>
            """+o.description+"""
          </body>
        </html>
        """

        # Record the MIME types of both parts - text/plain and text/html.
        part1 = MIMEText(text.encode('utf-8'), 'plain', _charset='utf-8')
        part2 = MIMEText(html.encode('utf-8'), 'html', _charset='utf-8')

        # Attach parts into message container.
        # According to RFC 2046, the last part of a multipart message, in this case
        # the HTML message, is best and preferred.
        msg.attach(part1)
        msg.attach(part2)

        # Send the message via local SMTP server.
        s = smtplib.SMTP(self.server_smtp)
        s.starttls()
	s.login(self.login,self.password)
        # sendmail function takes 3 arguments: sender's address, recipient's address
        # and message to send - here it is sent as one string.
        s.sendmail(self.email_from, self.email_to, msg.as_string())
        print 'Sent email', o.title
        s.quit()
    def delOffer(self,o):
        # Create message container - the correct MIME type is multipart/alternative.
        msg = MIMEMultipart('alternative')
        msg['Subject'] = "RE: [VIE "+self.subject_keyword+" "+o.date+"] "+o.title
        msg['From'] = self.email_from
        msg['To'] = self.email_to
        if o.reply_message_id is None:
            o.reply_message_id = make_msgid();
        if o.sent_message_id is not None:
            msg['References'] = o.sent_message_id;
            msg['In-Reply-To'] = o.sent_message_id;
        msg['Message-ID'] = o.reply_message_id;
        # Create the body of the message (a plain-text and an HTML version).
        text = "DELETED"
        html = """\
        <html>
          <head></head>
          <body>
            <h1>DELETED</h1>
          </body>
        </html>
        """

        # Record the MIME types of both parts - text/plain and text/html.
        part1 = MIMEText(text.encode('utf-8'), 'plain', _charset='utf-8')
        part2 = MIMEText(html.encode('utf-8'), 'html', _charset='utf-8')

        # Attach parts into message container.
        # According to RFC 2046, the last part of a multipart message, in this case
        # the HTML message, is best and preferred.
        msg.attach(part1)
        msg.attach(part2)

        # Send the message via local SMTP server.
        s = smtplib.SMTP(self.server_smtp)
	s.starttls()
        s.login(self.login,self.password)
        # sendmail function takes 3 arguments: sender's address, recipient's address
        # and message to send - here it is sent as one string.
        s.sendmail(self.email_from, self.email_to, msg.as_string())
        print 'Sent email DEL', o.title
        s.quit()

offer.py

class offer:
    ''' Container for offers' information '''
    def __init__(self,title,link,description,date):
        self.title = title
        self.link = link
        self.description = description
        self.date = date
        self.sent_message_id = None
        self.reply_message_id = None
	self.deleted = False
    def __eq__(self, other):
        return self.title == other.title
    def __str__(self):
        return str(title)