ps5.py

# 6.0001/6.00 Problem Set 5 - RSS Feed Filter
# Name:neelima-j


import feedparser
import string
import time
import threading
from project_util import translate_html
from mtTkinter import *
from datetime import datetime
import pytz
import re
import requests


#-----------------------------------------------------------------------

#======================
# Code for retrieving and parsing
# Google and Yahoo News feeds
# Do not change this code
#======================

def process(url):
    """
    Fetches news items from the rss url and parses them.
    Returns a list of NewsStory-s.
    """
    feed_content = requests.get(url)
    feed = feedparser.parse(feed_content.text)
#   feed = feedparser.parse(url)
    entries = feed.entries
    ret = []
    for entry in entries:
        guid = entry.guid
        title = translate_html(entry.title)
        link = entry.link
        description = translate_html(entry.description)
        pubdate = translate_html(entry.published)

        try:
            pubdate = datetime.strptime(pubdate, "%a, %d %b %Y %H:%M:%S %Z")

          #  pubdate.replace(tzinfo=pytz.timezone("GMT"))
          #  pubdate = pubdate.astimezone(pytz.timezone('EST'))
            pubdate.replace(tzinfo=None)
        except ValueError:
            pubdate = datetime.strptime(pubdate, "%a, %d %b %Y %H:%M:%S %z")

        newsStory = NewsStory(guid, title, description, link, pubdate)
        ret.append(newsStory)
    return ret

#======================
# Data structure design
#======================

# Problem 1

# TODO: NewsStory
class NewsStory():
    def __init__(self,guid, title, description, link, pubdate):
        self.guid = guid
        self.title = title
        self.description = description
        self.link = link
        self.pubdate = pubdate
    
    def get_guid(self):
        return self.guid
    def get_title(self):
        return self.title
    def get_description(self):
        return self.description
    def get_link(self):
        return self.link
    def get_pubdate(self):
        return self.pubdate
    


#======================
# Triggers
#======================

class Trigger(object):
    def evaluate(self, story):
        """
        Returns True if an alert should be generated
        for the given news item, or False otherwise.
        """
        # DO NOT CHANGE THIS!
        raise NotImplementedError

# PHRASE TRIGGERS

# Problem 2
# TODO: PhraseTrigger

class PhraseTrigger(Trigger):
    def __init__(self,phrase):
        self.phrase = phrase.lower().strip()
        self.phrase = "".join((char if char.isalpha() else "") for char in self.phrase)
    def is_phrase_in(self,text):
        regexString = self.phrase+'(?!s)(?!!)'
        phraseRegEx = re.compile(regexString)
        text = text.lower()
        text = "".join((char if char.isalpha() else "") for char in text)
        text = text+' '
        repeatRegEx = re.compile(r"(.+?)\1+")
        
        if (phraseRegEx.findall(text) and repeatRegEx.match(text) == None):
            return True
        else:
            return False
            
            

# Problem 3
# TODO: TitleTrigger
class TitleTrigger(PhraseTrigger):
    def evaluate(self,story):
        if self.is_phrase_in(story.title):
            return True
        else:
            return False

        
    

# Problem 4
# TODO: DescriptionTrigger
class DescriptionTrigger(PhraseTrigger):
    def evaluate(self, story):
        if self.is_phrase_in(story.description):
            return True
        else:
               return False
        

# TIME TRIGGERS

# Problem 5
# Constructor:
#        Input: Time has to be in EST and in the format of "%d %b %Y %H:%M:%S".
#        Convert time from string to a datetime before saving it as an attribute.
class TimeTrigger(Trigger):
    def __init__(self,time):
        self.time = datetime.strptime(time,r'%d %b %Y %H:%M:%S')
        '''

        est = pytz.timezone('US/Eastern')
        self.time = est.localize(self.time)
        '''
        
# Problem 6
class BeforeTrigger(TimeTrigger):
    def evaluate(self,story):
        '''
        Fires when a story is published strictly before the trigger’s time
        '''
        story.pubdate = story.pubdate.replace(tzinfo = None)
        
        '''I have no idea why it didnt work when everythin was in EST. So i removed TZinfo from everywhere

        if story.pubdate.tzinfo == None:
            est = pytz.timezone('US/Eastern')
            story.pubdate = est.localize(story.pubdate)
        '''    
        
        if story.pubdate < self.time:
            return True
        else:
            return False
        
class AfterTrigger(TimeTrigger):
    def evaluate(self,story):
        '''
        fires when a story is published strictly after the trigger’s time
        '''
        story.pubdate = story.pubdate.replace(tzinfo = None)
        
        '''
        if story.pubdate.tzinfo == None:
            est = pytz.timezone('US/Eastern')
            story.pubdate = est.localize(story.pubdate)
            '''
        if story.pubdate > self.time:
            return True
        else:
            return False

# COMPOSITE TRIGGERS

# Problem 7


class NotTrigger(Trigger):
    def __init__(self,T):
        self.T = T
    def evaluate(self,story):
        return not self.T.evaluate(story)
# Problem 8
class AndTrigger(Trigger):
    def __init__(self,t1,t2):
        self.t1 = t1
        self.t2 = t2
    def evaluate(self,story):
        return  self.t1.evaluate(story) and self.t2.evaluate(story)

# Problem 9
class OrTrigger(Trigger):
    def __init__(self,t1,t2):
        self.t1 = t1
        self.t2 = t2
    def evaluate(self,story):
        return  self.t1.evaluate(story) or self.t2.evaluate(story)



#======================
# Filtering
#======================

# Problem 10
def filter_stories(stories, tlist):
    """
    Takes in a list of NewsStory instances.
    Returns: a list of only the stories for which a trigger in triggerlist fires.
    """
    i=0
    ret_stories = []
    for story in stories:
        for trigger in tlist:
            if trigger.evaluate(story):
                ret_stories.append(story)
                break
    return ret_stories



#======================
# User-Specified Triggers
#======================
# Problem 11
def read_trigger_config(filename):
    """
    filename: the name of a trigger configuration file

    Returns: a list of trigger objects specified by the trigger configuration
        file.
    """

    trigger_file = open(filename, 'r')
    lines = []
    triggerlist = []
    ret = []

    for line in trigger_file:
        line = line.rstrip()
        if not (len(line) == 0 or line.startswith('//')):
            lines.append(line)

    # TODO: Problem 11
    # line is the list of lines that you need to parse and for which you need
    # to build triggers
    i = 0
    for line in lines:
        t,tname,*arg = line.split(',')
        if t != 'ADD':
            tname = tname.capitalize()+'Trigger'
            tobject = globals()[tname]
            if len(arg)==1:
                triggerlist.append(tobject(arg[0]))
            else:
                if tname == 'AND' or 'OR':
                    triggerlist.append(tobject(triggerlist[int(arg[0][1])-1],triggerlist[int(arg[1][1])-1]))
                else:    
                    triggerlist.append(tobject(arg[0],arg[1]))
        elif t == 'ADD':
            ret.append(triggerlist[int(tname[1])-1])
            ret.append(triggerlist[int(arg[0][1])-1])
    return ret

    



SLEEPTIME = 120 #seconds -- how often we poll

def main_thread(master):
    # A sample trigger list - you might need to change the phrases to correspond
    # to what is currently in the news
    try:
        '''
        
        t1 = TitleTrigger("The")
        t2 = DescriptionTrigger("Covid")
        t3 = DescriptionTrigger("India")
        t4 = AndTrigger(t2, t3)
        triggerlist = [t1, t4]
        '''
        
        # Problem 11
        triggerlist = read_trigger_config('triggers.txt')
        
        # HELPER CODE - you don't need to understand this!
        # Draws the popup window that displays the filtered stories
        # Retrieves and filters the stories from the RSS feeds
        frame = Frame(master)
        frame.pack(side=BOTTOM)
        scrollbar = Scrollbar(master)
        scrollbar.pack(side=RIGHT,fill=Y)

        t = "Google Top News"
        title = StringVar()
        title.set(t)
        ttl = Label(master, textvariable=title, font=("Helvetica", 18))
        ttl.pack(side=TOP)
        cont = Text(master, font=("Helvetica",14), yscrollcommand=scrollbar.set)
        cont.pack(side=BOTTOM)
        cont.tag_config("title", justify='center')
        button = Button(frame, text="Exit", command=root.destroy)
        button.pack(side=BOTTOM)
        guidShown = []
        def get_cont(newstory):
            if newstory.get_guid() not in guidShown:
                cont.insert(END, newstory.get_title()+"\n", "title")
                cont.insert(END, "\n---------------------------------------------------------------\n", "title")
                cont.insert(END, newstory.get_description())
                cont.insert(END, "\n*********************************************************************\n", "title")
                guidShown.append(newstory.get_guid())

        while True:

            print("Polling . . .", end=' ')
            # Get stories from Google's Top Stories RSS news feed
            stories = process("http://news.google.com/news?output=rss")
            # Get stories from Yahoo's Top Stories RSS news feed
#            stories.extend(process("http://news.yahoo.com/rss/topstories"))

            stories = filter_stories(stories, triggerlist)

            list(map(get_cont, stories))
            scrollbar.config(command=cont.yview)


            print("Sleeping...")
            time.sleep(SLEEPTIME)

    except Exception as e:
        print(e)


if __name__ == '__main__':
    root = Tk()
    root.title("Some RSS parser")
    t = threading.Thread(target=main_thread, args=(root,))
    t.start()
    root.mainloop()