ps5.py

#Unfinished, will complete in far off future.

import feedparser
import string
import time
import threading
from project_util import translate_html
from mtTkinter import *
from datetime import datetime
import pytz
# Code for retrieving and parsing Google and Yahoo News feeds

def process(url):
    """
    Fetches news items from the rss url and parses them.
    Returns a list of NewsStory-s.
    """
    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(object):
    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
class PhraseTrigger(Trigger):
    def __init__(self, phrase):
        self.phrase = phrase

    def is_phrase_in(self, text):
        new_text_list = []

        while "  " in text:  #If there's duplicate blanks, it gets rid of them
            text = text.replace("  ", " ")
        text = list(text)

        for characters in text:           #Gets rid of punctuation
            if characters not in string.punctuation:
                new_text_list.append(characters)
        new_text_string = ''.join(new_text_list)

        if self.phrase in new_text_string.lower():
            return True
        else:
            return False

class TitleTrigger(PhraseTrigger):
    def evaluate(self, story):
        return self.is_phrase_in(story.get_title())



class DescriptionTrigger(PhraseTrigger):
    def evaluate(self, story):
        return self.is_phrase_in(story.get_description())


class TimeTrigger(Trigger):  #Takes in date and time in format "DD MMM YYYY HH:MM:SS" E.g 03 Oct 2006 17:00:00
    def __init__(self, time_str):
        format_data = "%d %b %Y %H:%M:%S"   #This is how you format the data for the datetime module
        time = datetime.strptime(time_str, format_data).replace(tzinfo = pytz.timezone("EST"))  #Strptime converts the time input into proper format. The .replace will change it to the timezone while also using the pytz module.
        self.time = time

class BeforeTrigger(TimeTrigger): #Returns true if publication date was before the timestamp of the trigger
    def __init__(self, time_str):
        TimeTrigger.__init__(self, time_str)

    def evaluate(self, time_str):
        pub_date = self.pubdate
        if self.time > datetime.strptime(time_str, "%d %b %Y %H:%M:%S").replace(tzinfo = pytz.timezone("EST")):
            return True

class AfterTrigger(TimeTrigger):
    def __init__(self, time_str):
        TimeTrigger.__init__(self, time_str)

    def evaluate(self, time_str):
        pub_date = self.pubdate
        if self.time < datetime.strptime(time_str, "%d %b %Y %H:%M:%S").replace(tzinfo = pytz.timezone("EST")):
            return True


class NotTrigger(Trigger):
    def __init__(self, trigger):
        self.trigger = trigger

    def evaluate(self, story):
        return not self.trigger.evaluate(story)

class AndTrigger(Trigger):
    def __init__(self, trigger1, trigger2):
        self.trigger1 = trigger1
        self.trigger2 = trigger2

    def evaluate(self, story):
        if self.trigger1.evaluate(story) == True and self.trigger2.evaluate(story) == True:    #Should this be self.story or just story?
            return True

class OrTrigger(Trigger):
    def __init__(self, trigger1, trigger2):
        self.trigger1 = trigger1
        self.trigger2 = trigger2

    def evaluate(self, story):
        if self.trigger1.evaluate(story) == True or self.trigger2.evaluate(story) == True:    #Should this be self.story or just story?
            return True

def filter_stories(stories, triggerlist): #Returns only stories that have a trigger fired
    triggered_stories = []

    for story in stories:
        for trigger in triggerlist:
            if trigger.evaluate(story):
                triggered_stories.append(story)

    return triggered_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.
#     """
#     # We give you the code to read in the file and eliminate blank lines and
#     # comments. You don't need to know how it works for now!
#     trigger_file = open(filename, 'r')
#     lines = []
#     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
#
#     print(lines) # for now, print it so you see what it contains!
#
#
#
# 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("election")
#         t2 = DescriptionTrigger("Trump")
#         t3 = DescriptionTrigger("Clinton")
#         t4 = AndTrigger(t2, t3)
#         triggerlist = [t1, t4]
#
#         # Problem 11
#         # TODO: After implementing read_trigger_config, uncomment this line
#         # 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 & Yahoo 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()