# FS - Facebook Spider
# V 0.2
# Software created by Sean Colyer, contributions from Joe Kubiak
# http://seancolyer.com
# sean@seancolyer.com
#
# Copyright (C) 2008-2009 Sean Colyer
#
# This package is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License as
# published by the Free Software Foundation; either
# version 2 of the License, or (at your option) any later version.
#
# This package is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public
# License along with this package; if not, write to the Free Software
# Foundation, 51 Franklin Street, Fifth Floor, Boston, MA, 02110-1301
# USA

#for some reason creating response.read() variables seemed needed or else accessing the response.read() multiple times tended to lead to problems.
#the process is not threaded and runs through the pictures in the order they are on facebook. This is a more accurate representation of a how a human would view the photos.
#Regular expressions could probably be used to search through the page more effectively.

#FIXME -- FB no longer uses many of the abbreviated subdomains i.e. uva.facebook.com , which should simplify a bit of the code

import os, urllib,urllib2,cookielib,getpass

#Change these settings below as desired
fakeHeader = {'User-agent' : 'Mozilla/4.0 (compatible; MSIE 6.0; AOL 9.0; Windows NT 5.1)'} #Use mozilla header to avoid obvious script appearance
dirStyle = 1; #1 for "Name FacebookID#" style directory  ,  anything else for just "Name" directory , to avoid multiple people of the same name.
rootDir ='' #if not null, needs to end with /, defaults storing pictures to wherever this file is saved.
profile = 'profile'
user = 'user'
allPics = 'all'
album = 'album'

#acquire users ID based on their homepage
def getID (homeUrl):
    homeReq = urllib2.Request(homeUrl,None,fakeHeader)
    homeResponse = urllib2.urlopen(homeReq)
    homePage = homeResponse.read()
    #New Facebook Style:
    fbid = homePage[homePage.find('profile.php?id=')+15:-1]
    fbid = fbid[0:fbid.find('"><')] #I don't like this because I think its possible that someone add to their homepage that would get the first hit. Prefer to search for "ref=name"
    logoutID = homePage[homePage.find('logout.php?h=')+13:-1]
    logoutID = logoutID[0:logoutID.find('&amp')]
    print logoutID, fbid
    return logoutID, fbid

#cleanly logs out off facebook and saves cookiefile
def logout(cookieFile,logoutID):
    print logoutID
    logoutBase = 'http://www.facebook.com/logout.php?h='
    logoutRequest = urllib2.Request(logoutBase + logoutID,None,fakeHeader)
    urllib2.urlopen(logoutRequest)
    cookieJar.save(cookieFile)
    print 'You are now logged out.'

#acquires name from someones profile, either end user or searched user
def getName():
    profileUrl = shortUrl + 'profile.php?id=' + fbid
    print profileUrl
    profileReq = urllib2.Request(profileUrl,None, fakeHeader)
    profileResponse = urllib2.urlopen(profileReq)
    profilePage = profileResponse.read()
    name = profilePage[profilePage.find('<title>'):profilePage.find('/title>')]
    name = name[18:-1]
    print name
    return name

def getPhotos(dirName, view, albumid = -3, albumName = None):
    #facebook uses only slight differences between different styles of photo, so a single method can be created for different types -- 2009/03 I think even closer together now
    if view is allPics or view is user:
        picBaseUrl = shortUrl + 'profile.php?id=' + fbid +'&v=photos'
        #OLD STYLE: picBaseUrl = shortUrl + 'photo_search.php?id=' + fbid + '&view=' + view
        lcdir = '/General Pictures'
    if view is album:
        picBaseUrl = shortUrl + 'album.php?aid=' + albumid + '&id=' + fbid
        albumName = albumName.replace('/','-')
        lcdir = '/' + albumName
    if view is profile:
        picBaseUrl = shortUrl + 'album.php?profile&id='+ fbid
        lcdir = '/Profile Pictures'
    #start the search
    #print picBaseUrl
    picBaseReq = urllib2.Request(picBaseUrl,None,fakeHeader)
    picBaseResponse = urllib2.urlopen(picBaseReq)
    picBasePage = picBaseResponse.read()
    #print picBasePage

    if view is profile or view is album:
        numPics = picBasePage[picBasePage.find('"summary"><h4>')+14:-1]
        numPics = numPics[0:numPics.find('<')]
        numPics = (numPics.split())[-1]
        if not numPics.isdigit(): #special case if there are fewer than 20, I believe, as there are not multiple pages
            numPics = picBasePage[picBasePage.find('"summary"><h4>')+14:-1]
            numPics = numPics[0:numPics.find(' photos')]
        picPerPage = 20
    if view is allPics or view is user:
        numPics = picBasePage[picBasePage.find('class="caption">')+16:-1]
        numPics = numPics[0:numPics.find(' photos')]
        picPerPage = 15

    if not numPics.isdigit():
        print "No photos under the " + view + " view."
        return
    
    picBasePage = picBasePage[picBasePage.find('UIPhotoGrid_Container'):-1] #Ignore extra pics sent by FB

    #Loop to get individual pictures
    temp=0
    while temp< int(numPics):
        print str(temp + 1) + ' / ' + str(numPics)
        if temp > 0 and temp % picPerPage == 0:
            #once again, slight variations for different styles of pictures.
            if view is allPics or view is user:
                #picBaseUrl = shortUrl + 'photo_search.php?page=' + str(int(temp/20)+1) + '&id=' + fbid + '&view=' + view
                picBaseUrl = shortUrl + 'profile.php?id=' + fbid + '&v=photos&so=' + str(int(temp))
            if view is album:
                picBaseUrl = shortUrl + 'album.php?aid=' + str(albumid) + '&page=' + str(int(temp/picPerPage)+1) + '&id=' + fbid
                #print picBaseUrl
            if view is profile:
                picBaseUrl = shortUrl + 'album.php?page=' + str(int(temp/picPerPage)+1) + '&aid=-3&id=' + fbid
                #print picBaseUrl
            picBaseReq = urllib2.Request(picBaseUrl,None,fakeHeader)
            picBaseResponse = urllib2.urlopen(picBaseReq)
            picBasePage = picBaseResponse.read()
            #print picBaseUrl
            picBasePage = picBasePage[picBasePage.find('UIPhotoGrid_Container'):-1]
        if picBasePage.find('photo.php?pid=') is -1:
            break
        picBasePage=picBasePage[picBasePage.find('photo.php?pid=')+14:-1]
        picUrlEnd = picBasePage[0:picBasePage.find('"')] #has picid, subj and id which are now required
        picid = picUrlEnd[0:picUrlEnd.find('&op')]
        picUrlEnd = picUrlEnd.replace('&amp;','&')
        #Get and save photo
        dirName = rootDir + dirName #set active directory so it includes full path up to the user
        if not os.path.isfile(dirName + lcdir + '/' + picid[0:8] + '.jpg'): #skips over pictures which have already been saved under this name to this directory, to speed up re-running
            picUrl = shortUrl + 'photo.php?pid=' + picUrlEnd
            #print picUrl
            picRequest = urllib2.Request(picUrl,None,fakeHeader)
            picResponse = urllib2.urlopen(picRequest)
            picPage = picResponse.read()
            #print picUrl
            picPage = picPage[picPage.find('id="myphotolink"><img src="'):-1]
            picPage = picPage[picPage.find('http://photos'):-1]
            picPage = picPage[0:picPage.find('"')]
            if not os.path.isdir(dirName): #create User Directory if needed
                os.mkdir(dirName)
            if not os.path.isdir(dirName + lcdir):#Create user\category Directory if needed
                os.mkdir(dirName + lcdir)
            picout = open(dirName + lcdir + '/' + picid[0:8] + '.jpg', "wb")
            picPageFile = (urllib2.urlopen(urllib2.Request(picPage,None,fakeHeader))).read()
            picout.write(picPageFile)
            picout.close()
        temp+=1

def getPhotoAlbums(dirName):
    albumUrl = shortUrl + 'photos.php?id=' + fbid + '&ref=pb'
    albumReq = urllib2.Request(albumUrl,None,fakeHeader)
    albumResponse = urllib2.urlopen(albumReq)
    albumPage = albumResponse.read()
    if albumPage.find('No Photo Albums') is not -1:
        return
    numAlbums = albumPage[albumPage.find('"summary">')+10:-1]
    numAlbums = int(numAlbums[0:numAlbums.find('Photo')])
    temp = 0
    #print albumUrl
    while temp < numAlbums:
        if temp > 0 and temp % 5 == 0:
            albumUrl = shortUrl + 'photos.php?id=' + fbid + '&s=' + str(temp)
            albumReq = urllib2.Request(albumUrl,None,fakeHeader)
            albumResponse = urllib2.urlopen(albumReq)
            albumPage = albumResponse.read()
        albumPage = albumPage[albumPage.find('<h2><a href="album.php?aid=')+ 27 :-1]
        albumName = albumPage[albumPage.find('">')+2:albumPage.find('</a>')]
        albumid = albumPage[0:albumPage.find('&')]
        print 'Acquiring: ' + albumName + ' album.'
        getPhotos(dirName, 'album', albumid, albumName)
        temp+=1
    

def search () : #searches using facebook's s.php, for the current time it only takes the top hit -- POOR search method.
    searchUrl = homeUrl[0:-8] + 's.php?q=' + searchTerm.replace(' ','+')
    #searchValues = urllib.urlencode({'q' : searchTerm})#,{'init': 'q'}) #BROKEN --  i think urlencode doesn't work properly (sid not returned). Used hack on previous line
    searchRequest = urllib2.Request(searchUrl,None,fakeHeader)
    searchResponse = urllib2.urlopen(searchRequest)
    searchPage = searchResponse.read()
    sid = searchPage[searchPage.find('&sid=')+5:-1]
    sid = sid[0:sid.find('")')]
    searchRequest = urllib2.Request(searchUrl + '&sid=' + sid, None, fakeHeader)
    searchResponse = urllib2.urlopen(searchRequest)
    searchPage = searchResponse.read()
    #print searchPage
    searchPage = searchPage[searchPage.find('friend='):-1]
    searchUrl = searchPage[searchPage.find('http%3A%2F%2F')+13:-1] #trims search page to everything after the first search
    searchid = searchUrl[searchUrl.find('%3Fid%3D')+8:searchUrl.find('%26')] #finds the FB id #
    searchUrl = 'http://' + searchUrl[0:searchUrl.find('%2Fprofile.php')] + '/' #finds the url with the home network of the person
    return searchUrl , searchid

if __name__== "__main__":
    cookieFile = '.cookies.lwp'  #creates/uses if created, .cookies.lwp file to store cookie information
    cookieJar = cookielib.LWPCookieJar()
    if os.path.isfile(cookieFile):
       cookieJar.load(cookieFile)
    opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookieJar))
    urllib2.install_opener(opener)  #installs cookieJar into urllib2 for automated use in accessing FB
    
    email = raw_input('What is your Facebook Email Address?')
    password = getpass.getpass('What is your Facebook password?')
    loginValues ={'email' : email, 'pass' : password }
    encodedLogin =  urllib.urlencode(loginValues)
    getRequest = urllib2.Request('https://login.facebook.com/login.php',None, fakeHeader)   #Need 2 requests to conform with POST requirements of Facebook
    getResponse = urllib2.urlopen(getRequest)
    loginRequest = urllib2.Request('https://login.facebook.com/login.php',encodedLogin,fakeHeader)
    loginResponse = urllib2.urlopen(loginRequest)
    homePage = loginResponse.read()
    searchTerm = -99
    homeUrl = 'http://www.facebook.com/home.php'
    #homeUrl = homePage[homePage.find('\'')+1:homePage.rfind('\'')]   #http://__USERS-NETWORK__.facebook.com/home.php?
    #print homeUrl
    logoutID,fbid = getID(homeUrl)
    print 'ID: ' + fbid
    
    while searchTerm is not 1:
        searchTerm = raw_input('Who would you like to search for? ( 0 for yourself , 1 to exit, type name of others )')
        #wishes to exit
        if searchTerm.isdigit() and int(searchTerm) is 1:
            break
        #searches for oneself
        if searchTerm.isdigit() and int(searchTerm) is 0:
            shortUrl = homeUrl[0:-8] #Need to re-establish fbid? or keep copy of old?
        #name entered to be searched
        else:
            shortUrl, fbid = search()
            #print shortUrl
            #print fbid

        name = getName()

        if dirStyle is 1:
            dirName = name + ' ' + fbid
        else:
            dirName = name
        
        print 'Acquiring Profile pictures of ' + name + '.'
        getPhotos(dirName, profile)
        print 'Acquiring pictures of ' + name + ' owned by others.'
        getPhotos(dirName, allPics)
        print 'Acquiring pictures of ' + name + ' owned by them.'
        getPhotos(dirName, user)
        print 'Acquiring Photo albums owned by ' + name
        getPhotoAlbums(dirName)
    logout(cookieFile,logoutID)