#!/usr/bin/python -S

"""

Retrieve various daily cartoon strips.  Currently works with Calvin
and Hobbes, Non Sequitur, Userfriendly, Badtech, Dilbert and a
Shakespearean insult.  Its rather easy to add more.

This code is loosely based on code from by Prahlad Vaidyanathan
<slime@vsnl.net>: http://www.symonds.net/~prahladv/files/cartoons.py
who in turn 'stole it from somewhere on Useless python' and customized
it: http://www.lowerstandard.com/python

Prabhu Ramachandran <prabhu@aero.iitm.ernet.in>

"""

import sys, string, os, re, time

# Let this point to your archive directory.  This is where the
# cartoons will be saved.  This directory will also contain a log
# file (cartoon.log).
ARCHIVE_DIR='/skratch/prabhu/tmp/cartoons'

# If you are behind a proxy then you can either set the enviroment
# variable 'http_proxy' in your shell or uncomment the following line
# after editing it suitably.
#os.environ['http_proxy'] = 'http://www.someproxy.com:3128'

import MailMsg, mimetypes
import urllib, imghdr


def open_file(file_name, mode):    
    """Creates the base directory if it does not exist and then
    returns the file with the requested mode."""    
    basedir = os.path.dirname(os.path.abspath(file_name))
    if not os.path.isdir(basedir):
        os.mkdir(basedir)
    return open(file_name, mode)


class GetCartoon:
    archive_dir = ARCHIVE_DIR
    #log_file = sys.stdout
    log_file = open_file(os.path.join(archive_dir, 'cartoon.log'), 'a')
    
    def __init__(self):
        if not os.path.isdir(self.archive_dir):
            os.mkdir(self.archive_dir)            

    def log(self, msg):
        print >> self.log_file, msg
        self.log_file.flush()
        
    def retrieve(self, url, filename):
        try :
            self.log('Retrieving %s -> %s' % (url, filename))
            urllib.urlretrieve(url,filename)
            self.log('Done.')
        except IOError, txt :
            self.log( 'Connect error : %s'%txt)
            sys.exit(1)
    
    def filename(self, name, ext):
        name = os.path.join(self.archive_dir, name)
        format = '%s-%s-%s-%s.%s'
        today = time.localtime(time.time())
        newfile = format % (name,today[0],today[1],today[2],ext)
        return newfile

    def cleanOldCartoon(self, name, ext):
        name = os.path.join(self.archive_dir, name)
        format = '%s-%s-%s-%s.%s'
        today = time.localtime(time.time())
        oldfile = format % (name,today[0],today[1],(today[2]-1),ext)
        if os.path.exists(oldfile) :
            os.remove(oldfile)        

    def get(self):        
        """Do cartoon specific stuff here and return the filename of
        the cartoon."""
        return None
        

class BadTech(GetCartoon):
    def get(self):
        name  = self.filename('badtech','jpg')
        if os.path.exists(name) :
            self.log('Already retrieved badtech cartoon today.')
        else :
            yday = time.localtime(time.time() - 24*3600) 
            url = 'http://www.badtech.com/a/%s/%s/%s.jpg' % \
                  (`yday[0]`[-1:],yday[1],yday[2])
            self.retrieve(url,name)
            if imghdr.what(name) != 'jpeg':
                os.remove(name)
                raise AssertionError, \
                      "Unable to get %s. Try again later."%url
            # self.cleanOldCartoon('badtech', 'jpg')
        return name    


class UserFriendly(GetCartoon):
    def get(self):
        name = self.filename('userfriendly','gif')
        if os.path.exists(name) :
            self.log('Already retrieved UserFriendly cartoon today.')
        else:
            url = 'http://www.userfriendly.org'
            try:
                page = urllib.urlopen(url)
            except IOError, txt:
                self.log('Connect error: %s'%txt)
                sys.exit(1)
            
            patn = re.compile(r'Latest Strip[^>]+?src="(http[^"]+?\.gif)"',
                              re.I)
            pic = patn.search(page.read()).groups()[0]
            self.retrieve(pic, name)            
            if imghdr.what(name) != 'gif':
                os.remove(name)
                raise AssertionError, \
                      "Unable to get %s. Try again later."%pic
            # self.cleanOldCartoon('userfriendly', 'gif')
        return name


class Dilbert(GetCartoon):
    def get(self):
        name  = self.filename('dilbert', 'gif')
        if os.path.exists(name) :
            self.log('Already retrieved Dilbert cartoon today.')
        else:
            url = 'http://www.dilbert.com'
            try:
                page = urllib.urlopen(url)
            except IOError, txt:
                self.log('Connect error: %s'%txt)
                sys.exit(1)
            patn = re.compile(r'src="/([^"]+?dilbert\d+\.gif)"', re.I)
            comic = patn.search(page.read()).groups()[0]
            pic = urllib.basejoin(url, comic)
            self.retrieve(pic, name)
            if imghdr.what(name) != 'gif':
                os.remove(name)
                raise AssertionError, \
                      "Unable to get %s. Try again later."%pic
        return name

        
class CalvinAndHobbes(GetCartoon):
    def get(self):
        name  = self.filename('ch', 'gif')
        if os.path.exists(name) :
            self.log('Already retrieved Calvin and Hobbes cartoon today.')
        else:
            url = 'http://www.ucomics.com/calvinandhobbes/viewch.htm'
            try:
                page = urllib.urlopen(url)
            except IOError, txt:
                self.log('Connect error: %s'%txt)
                sys.exit(1)
            patn = re.compile(r'src="([^"]+?ch\d+\.gif)"', re.I)
            pic = patn.search(page.read()).groups()[0]
            self.retrieve(pic, name)
            if imghdr.what(name) != 'gif':
                os.remove(name)
                raise AssertionError, \
                      "Unable to get %s. Try again later."%pic            
        return name    


class NonSequitur(GetCartoon):
    def get(self):
        name  = self.filename('nq', 'gif')
        if os.path.exists(name) :
            self.log('Already retrieved Non Sequitur cartoon today.')
        else:
            url = 'http://www.ucomics.com/nonsequitur/viewnq.htm'
            try:
                page = urllib.urlopen(url)
            except IOError, txt:
                self.log('Connect error: %s'%txt)
                sys.exit(1)
            patn = re.compile(r'src="([^"]+?nq\d+\.gif)"', re.I)
            pic = patn.search(page.read()).groups()[0]
            self.retrieve(pic, name)
            if imghdr.what(name) != 'gif':
                os.remove(name)
                raise AssertionError, \
                      "Unable to get %s. Try again later."%pic            
        return name    


class ShakespeareanInsult(GetCartoon):
    def get(self):
        name = self.filename('shakespeare','html')
        if os.path.exists(name):
            self.log('Already retrieved shakespearean insult today.')
        else :
            url = 'http://www.pangloss.com/seidel/Shaker/index.html'
            page = urllib.urlopen(url)
            patn = re.compile(r'"the Bard">\s+(<p>.*?)<form', re.S|re.I)
            quote = patn.search(page.read()).groups()[0]
            f = open(name, 'w')
            f.write(quote)
            f.close()
            # self.clearOldCartoon('shakespeare','html')
        return name


def mailEmAll(from_addr='prabhu@aero.iitm.ernet.in',
              to_addrs=['prabhu@localhost']):
    """Mails the cartoons to all the folks in the to_addrs
    argument."""    

    #toons = [CalvinAndHobbes(), Dilbert(), BadTech(),
    #         UserFriendly(), ShakespeareanInsult()]
    toons = [CalvinAndHobbes(), Dilbert(), 
             UserFriendly(), NonSequitur()]
    
    m = MailMsg.MailMsg(from_addr, 'Daily Cartoon Strips!')
    #m.addText('Enjoy!\n')
    
    for toon in toons:
        file_name = toon.get()
        if mimetypes.guess_type(file_name)[0] == 'text/html':
            txt = open(file_name).read()
            m.addHtml(txt)
        else:
            m.addAttachment(file_name)
            
    m.mail(from_addr, to_addrs)


if __name__ == '__main__':
    # Change the from and to addresses to suit your needs.
    frm = 'prabhu@aero.iitm.ernet.in'
    to = ['prabhu@localhost', 'krishna@aero.iitm.ernet.in']
    mailEmAll(from_addr=frm, to_addrs=to)