# 6.00 Problem Set 6
# RSS Feed Filter

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

#-----------------------------------------------------------------------
#
# Problem Set 6

#======================
# Code for retrieving and parsing
# Google News feed
# 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
#======================

# Question 1

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 getGuid(self):
        return self.guid
    def getTitle(self):
        return self.title
    def getSubject(self):
        return self.subject
    def getSummary(self):
        return self.summary
    def getLink(self):
        return self.link

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

class Trigger(object):
    def evaluate(self, story):
        return True

# Whole Word Triggers
# Questions 2-5

class WordTrigger(Trigger):
    def __init__(self, word):
        self.word = word
    def isWordIn(self, text):
        for punc in string.punctuation:
            text = text.replace(punc, " ")
        if self.word.lower() in text.lower().split():
            return True
        return False

class TitleTrigger(WordTrigger):
    def evaluate(self, story):
        return self.isWordIn(story.getTitle())

class SubjectTrigger(WordTrigger):
    def evaluate(self, story):
        return self.isWordIn(story.getSubject())

class SummaryTrigger(WordTrigger):
    def evaluate(self, story):
        return self.isWordIn(story.getSummary())

# Composite Triggers
# Questions 6-8

class NotTrigger(Trigger):
    def __init__(self, t):
        self.t = t
    def evaluate(self, story):
        return not self.t.evaluate(story)
    
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)
    
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)
    
# Phrase Trigger
# Question 9

class PhraseTrigger(Trigger):
    def __init__(self, phrase):
        self.phrase = phrase
    def evaluate(self, story):
        return self.phrase in story.getTitle() \
               or self.phrase in story.getSubject() \
               or self.phrase in story.getSummary()


#======================
# 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.
    """
    result = []
    for story in stories:
        for trigger in triggerlist:
            if trigger.evaluate(story):
                result.append(story)
                break
    return result

#======================
# 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
    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)

    # 'lines' has a list of lines you need to parse
    # Build a set of triggers from it and
    # return the appropriate ones
    triggers = {}
    result = []
    for line in lines:
        words = line.split()
        if words[0] == "ADD":
            for trigger_name in words[1:]:
                result.append(triggers[trigger_name])
        else:
            trigger_name = words[0]
            trigger_type = words[1]
            if trigger_type == "TITLE":
                triggers[trigger_name] = TitleTrigger(words[2])
            if trigger_type == "SUBJECT":
                triggers[trigger_name] = SubjectTrigger(words[2])
            if trigger_type == "SUMMARY":
                triggers[trigger_name] = SummaryTrigger(words[2])
            if trigger_type == "NOT":
                triggers[trigger_name] = NotTrigger(triggers[words[2]])
            if trigger_type == "AND":
                triggers[trigger_name] = AndTrigger(triggers[words[2]], triggers[words[3]])
            if trigger_type == "OR":
                triggers[trigger_name] = OrTrigger(triggers[words[2]], triggers[words[3]])
            if trigger_type == "PHRASE":
                triggers[trigger_name] = PhraseTrigger(' '.join(words[2:]))
    return result
    
SLEEPTIME = 60 #seconds -- how often we poll
if __name__ == '__main__':
    # A sample trigger list - you'll replace
    # this with something more configurable in Question 11
    #t1 = SubjectTrigger("world")
    #t2 = SummaryTrigger("H1N1")
    #t3 = PhraseTrigger("health care")
    #t4 = OrTrigger(t2, t3)
    #triggerlist = [t1, t4]
    
    # After implementing readTriggerConfig, uncomment this line 
    triggerlist = readTriggerConfig("triggers.txt")

    guidShown = []
    p = Popup()
    
    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.getGuid() not in guidShown:
                newstories.append(story)
        
        for story in newstories:
            guidShown.append(story.getGuid())
            p.newWindow(story)

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