ps5

# 6.00 Problem Set 5
# RSS Feed Filter

import feedparser
import string
import time
from project_util import translate_html
from news_gui import Popup

#-----------------------------------------------------------------------
#
# Problem Set 5

#======================
# 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 = feedparser.parse(url)
    entries = feed.entries
    ret = []
    for entry in entries:
        guid = entry.guid
        title = translate_html(entry.title)
        link = entry.link
        summary = translate_html(entry.summary)
        try:
            subject = translate_html(entry.tags[0]['term'])
        except AttributeError:
            subject = ""
        newsStory = NewsStory(guid, title, subject, summary, link)
        ret.append(newsStory)
    return ret

#======================
# Part 1
# Data structure design
#======================

# Problem 1

# TODO: NewsStory

class NewsStory(object):
    def __init__(self, guid, title, subject, summary, link):
        self.guid = guid
        self.title = title
        self.subject = subject
        self.summary = summary
        self.link = link
    def get_guid(self):
        return self.guid
    def get_title(self):
        return self.title
    def get_subject(self):
        return self.subject
    def get_summary(self):
        return self.summary
    def get_link(self):
        return self.link

#======================
# Part 2
# Triggers
#======================

class Trigger(object):
    def evaluate(self, story):
        """
        Returns True if an alert should be generated
        for the given news item, or False otherwise.
        """
        raise NotImplementedError

# Whole Word Triggers
# Problems 2-5

# TODO: WordTrigger
class WordTrigger(Trigger):
    def __init__(self, word):
        self.word = word.lower()
    def is_word_in(self, text):
        self.text = text
        self.text = self.text.lower()
        for c in string.punctuation:
            self.text = self.text.replace(c,' ')
        self.text = self.text.split()
        if self.word in self.text:
            return True
        else:
            return False

# TODO: TitleTrigger
# TODO: SubjectTrigger
# TODO: SummaryTrigger

class TitleTrigger(WordTrigger):
    def evaluate(self, story):
        return self.is_word_in(story.get_title())

class SubjectTrigger(WordTrigger):
    def evaluate(self, story):
        return self.is_word_in(story.get_subject())

class SummaryTrigger(WordTrigger):
    def evaluate(self, story):
        return self.is_word_in(story.get_summary())

# Composite Triggers
# Problems 6-8

# TODO: NotTrigger
# TODO: AndTrigger
# TODO: OrTrigger

class NotTrigger(Trigger):
    def __init__(self, other):
        self.other = other
    def evaluate(self, story):
        if not self.other.evaluate(story):
            return True
        else:
            return False

class AndTrigger(Trigger):
    def __init__(self, other1, other2):
        self.other1 = other1
        self.other2 = other2
    def evaluate(self, story):
        if self.other1.evaluate(story) and self.other2.evaluate(story):
            return True
        else:
            return False

class OrTrigger(Trigger):
    def __init__(self, other1, other2):
        self.other1 = other1
        self.other2 = other2
    def evaluate(self, story):
        if self.other1.evaluate(story) or self.other2.evaluate(story):
            return True
        else:
            return False

# Phrase Trigger
# Question 9

# TODO: PhraseTrigger

class PhraseTrigger(Trigger):
    def __init__(self, phrase):
        self.phrase = phrase
    def evaluate(self, story):
        if self.phrase in story.get_title() or \
           self.phrase in story.get_subject() or \
           self.phrase in story.get_summary():
            return True
        else:
            return False

#======================
# Part 3
# Filtering
#======================

def filter_stories(stories, triggerlist):
    """
    Takes in a list of NewsStory-s.
    Returns only those stories for whom
    a trigger in triggerlist fires.
    """
    # TODO: Problem 10
    # This is a placeholder (we're just returning all the stories, with no filtering)
    # Feel free to change this line!
    filtered_stories = []

    for i in stories:
        for f in triggerlist:
            if f.evaluate(i) is True:
                print i.get_title()
                filtered_stories.append(i)
                break

    return filtered_stories

##story = NewsStory(12,'Why do apples grow on trees?', 'Fruit', 'Because they do', 'link')
##text = 'Apples, and trees" and nature stuff!'
##word = 'do'
##test3 = TitleTrigger(word)
##test4 = SubjectTrigger(word)
##test5 = SummaryTrigger(word)
##test6 = NotTrigger(TitleTrigger(word))
##test7 = AndTrigger(TitleTrigger(word),SubjectTrigger(word))
##test8 = OrTrigger(TitleTrigger(word),SummaryTrigger(word))
##print 'text',test3.is_word_in(text)
##print 'title',test3.evaluate(story)
##print 'subject',test4.evaluate(story)
##print 'summary',test5.evaluate(story)
##print 'not=', test6.evaluate(story)
##print 'and=', test7.evaluate(story)
##print 'or=', test8.evaluate(story)

#======================
# Part 4
# User-Specified Triggers
#======================

def readTriggerConfig(filename):
    """
    Returns a list of trigger objects
    that correspond to the rules set
    in the file filename
    """
    # Here's some code that we give you
    # to read in the file and eliminate
    # blank lines and comments
    triggerlist = []
    triggerkeys = {}
    triggered = []
    triggerfile = open(filename, "r")
    all = [ line.rstrip() for line in triggerfile.readlines() ]
##    print 'all=', all
    lines = []
    for line in all:
        if len(line) == 0 or line[0] == '#':
            continue
        lines.append(line)
##    print 'lines=', lines

    for i in range(len(lines)):
        triggerlist.append(lines[i].split())
##    print 'sep=', triggerlist

    for i in range(len(triggerlist)):
        if triggerlist[i][1] == 'TITLE':
            print 'title=', triggerlist[i][2]
            triggerkeys[triggerlist[i][0]] = TitleTrigger(triggerlist[i][2])
        elif triggerlist[i][1] == 'SUBJECT':
            triggerkeys[triggerlist[i][0]] = SubjectTrigger(triggerlist[i][2])
        elif triggerlist[i][1] == 'SUMMARY':
            print 'summary=', triggerlist[i][2]
            triggerkeys[triggerlist[i][0]] = SummaryTrigger(triggerlist[i][2])
        elif triggerlist[i][1] == 'NOT':
            triggerkeys[triggerlist[i][0]] = NotTrigger(triggerlist[i][2])
        elif triggerlist[i][1] == 'PHRASE':
            stitch = ''
            for w in triggerlist[i][2:]:
                if w == triggerlist[i][2]:
                    stitch = stitch + w
                else:
                    stitch = stitch + ' ' + w
            print 'stitch=', stitch
            triggerkeys[triggerlist[i][0]] = PhraseTrigger(stitch)
        elif triggerlist[i][1] == 'AND':
            triggerkeys[triggerlist[i][0]] = AndTrigger(triggerkeys[triggerlist[i][2]],triggerkeys[triggerlist[i][3]])
        elif triggerlist[i][1] == 'OR':
            triggerkeys[triggerlist[i][0]] = OrTrigger(triggerkeys[triggerlist[i][2]],triggerkeys[triggerlist[i][3]])
        elif triggerlist[i][0] == 'ADD':
            for t in triggerlist[i]:
##                print t
                if t in triggerkeys:
##                    print 'is a key'
                    triggered.append(triggerkeys[t])
##    print 'trigkey=', triggerkeys
##    print 'triggered=',  triggered


    return triggered
    # TODO: Problem 11
    # 'lines' has a list of lines you need to parse
    # Build a set of triggers from it and
    # return the appropriate ones

##t1 = TitleTrigger("Obama")
##t2 = SummaryTrigger("MIT")
##t3 = PhraseTrigger("Bombing")
##t4 = OrTrigger(t2, t3)
##triggerlist = [t1, t4]
##print 'trig=', triggerlist

import thread

def main_thread(p):
    # A sample trigger list - you'll replace
    # this with something more configurable in Problem 11
##    t1 = TitleTrigger("Obama")
##    t2 = SummaryTrigger("MIT")
##    t3 = PhraseTrigger("Bombing")
##    t4 = OrTrigger(t2, t3)
##    triggerlist = [t1, t4]
##    print 'triggerlist=', triggerlist

    # TODO: Problem 11
    # After implementing readTriggerConfig, uncomment this line
    triggerlist = readTriggerConfig("triggers.txt")

    guidShown = []

    while True:
        print "Polling..."

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

        # Only select stories we're interested in
        stories = filter_stories(stories, triggerlist)

        # Don't print a story if we have already printed it before
        newstories = []
        for story in stories:
            if story.get_guid() not in guidShown:
                newstories.append(story)

        for story in newstories:
            guidShown.append(story.get_guid())
            p.newWindow(story)

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

SLEEPTIME = 60 #seconds -- how often we poll
if __name__ == '__main__':
    p = Popup()
    thread.start_new_thread(main_thread, (p,))
    p.start()