# 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):
    '''
    Get to store guid, title, subject, summary, link.
    '''
    #initialize the object with the imformation that parsed

    def __init__(self, guid, title, subject, summary, link):
        self.guid = guid
        self.title = title
        self.subject = subject
        self.summary = summary
        self.link = link
    #provide functions to access the information we stored 
    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
import string 
class WordTrigger(Trigger):
    #store the word that wants to be triggered 
    def __init__(self,word):
        self.word = word

    #test if the word in text   
    def is_word_in(self,text):

        textlist = text.split(' ')
        #check all punctuation in the textlist
        #textlist should be the inner loop, because textlist.remove will allert each time inner loop runs
        for punc in string.punctuation:
            for i in textlist:
                if punc in i:
                    textlist.extend(i.split(punc))
                    textlist.remove(i)
        return self.word in textlist

class TitleTrigger(WordTrigger):
    def __init__(self,word):
        WordTrigger.__init__(self,word.lower())
    def evaluate (self,story):
        title = story.title.lower()
        return WordTrigger.is_word_in(self,title)
    

# TODO: SubjectTrigger
class SubjectTrigger(WordTrigger):
    def __init__(self, word):
        WordTrigger.__init__(self,word.lower())

    def evaluate (self, story):
        subject = story.subject.lower()
        return WordTrigger.is_word_in(self,subject)
        

    
# TODO: SummaryTrigger
class SummaryTrigger(WordTrigger):
    def __init__(self,word):
        WordTrigger.__init__(self,word.lower())
    def evaluate (self,story):
        summary = story.summary.lower()
##        return WordTrigger.is_word_in(self,summary)
        return self.is_word_in(summary)

# Composite Triggers
# Problems 6-8

# TODO: NotTrigger
class NotTrigger(WordTrigger):
    def __init__(self,trigger):
        self.trigger = trigger
	
    def evaluate (self, story):

        return not self.trigger.evaluate(story)



    
# TODO: AndTrigger
class AndTrigger(WordTrigger):
    #take in two triggers
    def __init__(self,trigger1,trigger2):
        self.trigger1 = trigger1
        self.trigger2 = trigger2
    #return True only if both triggers individually return True
    def evaluate (self, story):
        return self.trigger1.evaluate(story) and self.trigger2.evaluate(story)
    

# TODO: OrTrigger
class OrTrigger(WordTrigger):
    #take in two triggers
    def __init__(self, trigger1, trigger2):
        self.trigger1 = trigger1
        self.trigger2 = trigger2
    #return True if either trigger returns True
    def evaluate(self,story):
        return self.trigger1.evaluate(story) or self.trigger2.evaluate(story)


# Phrase Trigger
# Question 9

# TODO: PhraseTrigger
class PhraseTrigger(WordTrigger):
    def __init__(self, phrase):
        self.word = phrase

    def is_phrase_in(self,text):
        if self.word in text:
            return True
        else:
            return False
    def evaluate (self, story):
        return self.is_phrase_in(story.subject) or self.is_phrase_in(story.title) or self.is_phrase_in(story.summary)

#======================
# 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!

    #stories wanted stored here
    stories_wanted = []
    for story in stories:
        for trigger in triggerlist:
            if trigger.evaluate(story):
                stories_wanted.append(story)
                break
    return stories_wanted

#======================
# Part 4
# User-Specified Triggers
#======================
##problem not fixed:
###1 len(line) in my code but len(line)-1 in examples
###unhandled problem in main_thread

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
    triggerfile = open(filename, "r")
    all = [ line.rstrip() for line in triggerfile.readlines() ]
    lines = []
    for line in all:
        if len(line) == 0 or line[0] == '#':
            continue
        lines.append(line)

    # 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

    #return a triggerlist when finished
    triggerlist = []
    triggerdic ={}
    #parse each line in a list of lists
    for line in lines:
        #line is a string 
        line = line.split()
        #line converted into a list of string
        #lines become a list of lists

    #find ADD
    for line in lines:
        if 'ADD' in line:
            #parse to find triggers
            for i in range (1,len(line)-1):
                triggerlist.append(triggerdic[line[i]])
        #title, subject, summary triggers built with one argument
        elif 'SUBJECT' in line:
            #SubjectTrigger object definition
            subject_trigger = SubjectTrigger(line[2])
            #added into the dict{trigger's name:trigger}
            triggerdic[line[0]] = subject_trigger
        elif 'TITLE' in line:
            title_trigger = TitleTrigger(line[2])
            triggerdic[line[0]] = title_trigger
        elif 'SUMMARY' in line:
            summary_trigger = SummaryTrigger(line[2])
            triggerdic[line[0]] = summary_trigger
        elif 'PHRASE' in line:
            #add triggerwords together
            word = ''
            for string in range (2,len(line)-1):
                word = word +' '+ line[string]
            phrase_trigger = PhraseTrigger(word)
            triggerdic[line[0]] = phrase_trigger
        #not, and or triggers take triggers as arguments
        elif 'AND' in line:
            andtrigger = AndTrigger(triggerdic[line[2]],triggerdic[line[3]])
            triggerdic[line[0]] = andtrigger
        elif 'OR' in line:
            ortrigger = OrTrigger(triggerdic[line[2]],triggerdic[line[3]])
            triggerdic[line[0]] = ortrigger
        elif ' NOT' in line:
            not_trigger = NotTrigger(triggerdic[line[2]])
            triggerdic[line[0]] = not_trigger

    return triggerlist
        
    
import thread

def main_thread(p):
    # A sample trigger list - you'll replace
    # this with something more configurable in Problem 11
##    t1 = SubjectTrigger("Obama")
##    t2 = SummaryTrigger("MIT")
##    t3 = PhraseTrigger("Supreme Court")
##    t4 = OrTrigger(t2, t3)
##    triggerlist = [t1, t4]
    
    # TODO: Problem 11
    # After implementing readTriggerConfig, uncomment this line 
    triggerlist = readTriggerConfig("triggers.txt")
    print triggerlist
    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()