# Offlickr
# Hugo Haas -- mailto:hugo@larve.net -- http://larve.net/people/hugo/
# Homepage: http://larve.net/people/hugo/2005/12/offlickr/
# License: GPLv2

import sys
import libxml2
import urllib
import getopt
import time
import os.path

from flickrapi import FlickrAPI

version = '0.2 - 2006-01-03'

class Offlickr:

    # Gotten from Flickr
    __flickrAPIKey = '1391fcd0a9780b247cd6a101272acf71'
    __flickrSecret = 'fd221d0336de3b6d'

    def __init__(self, uid, browser = "lynx"):
        """Instantiates an Offlickr object"""
        # Get authentication token
        self.fapi = FlickrAPI(self.__flickrAPIKey, self.__flickrSecret)
        self.token = self.fapi.getToken(browser=browser)
        self.flickrUserId = uid

    def __testFailure(self, rsp):
        """Returns whether the previous call was successful"""
        return rsp['stat'] == "fail"

    def getPhotoList(self, dateLo, dateHi):
        """Returns a list of photo given a time frame"""
        n = 0
        photos = [ ]

        while True:
            n = n + 1
            rsp = self.fapi.photos_search(api_key=self.__flickrAPIKey, auth_token=self.token,
                                          user_id = self.flickrUserId,
                                          per_page = "500", # Max allowed by Flickr
                                          page = str(n),
                                          min_upload_date = dateLo, max_upload_date = dateHi)
            if self.__testFailure(rsp):
                return None
            if rsp.photos[0]['total'] == '0':
                print "No photos to back up"
                sys.exit(0)
            photos += rsp.photos[0].photo
            if len(photos) >= int(rsp.photos[0]['total']):
                break

        return photos

    def getPhotoMetadata(self, pid):
        """Returns a string containing the photo metadata (in XML)"""
        rsp = self.fapi.photos_getInfo(api_key=self.__flickrAPIKey, auth_token=self.token,
                                       photo_id=pid)
        if self.__testFailure(rsp):
            return None
        doc = libxml2.parseDoc(rsp.xml)
        metadata = str(doc.xpathEval( "/rsp/photo")[0])
        doc.freeDoc()
        return  [ metadata, rsp.photo[0]['originalformat'] ]

    def getPhotoSizes(self, pid):
        """Returns a string with is a list of available sizes for a photo"""
        rsp = self.fapi.photos_getSizes(api_key=self.__flickrAPIKey, auth_token=self.token,
                                        photo_id=pid)
        if self.__testFailure(rsp):
            return None
        return rsp

    def getOriginalPhoto(self, pid):
        """Returns a URL which is the original photo, if it exists"""
        source = None
        rsp = self.getPhotoSizes(pid)
        if rsp == None:
            return None
        for s in rsp.sizes[0].size:
            if s['label'] == 'Original':
                source = s['source']
        return source

    def __downloadReportHook(self, count, blockSize, totalSize):
        if self.__verbose == False:
            return
        p = 100 * count * blockSize / totalSize
        if (p > 100):
            p = 100
        print "\r %3d %%" % p,
        sys.stdout.flush()

    def downloadURL(self, url, filename, verbose = False):
        """Saves a photo in a file"""
        self.__verbose = verbose
        urllib.urlretrieve(url, filename, reporthook=self.__downloadReportHook)

def usage():
    print "Usage: Offlickr.py -d <flickr Id>"
    print "Backs up Flickr photos and metadata"
    print "Options:"
    print "\t-f <date>\tbeginning of the date range"
    print "\t\t\t(default: since you started using Flickr)"
    print "\t-t <date>\tend of the date range"
    print "\t\t\t(default: until now)"
    print "\t-d <dir>\tdirectory for saving files (default: ./dst)"
    print "\t-p\t\tback up photos in addition to metadata"
    print "\t-b <browser>\tbrowser to use for authentication (default: opera)"
    print "\t-h\t\tthis help message"
    print "\nDates are specified in seconds since the Epoch (00:00:00 UTC, January 1, 1970)."
    print "\nVersion " + version

def main():
    # Default options
    flickrUserId = None
    dateLo = '1'
    maxTime = '9999999999'
    dateHi = maxTime
    getPhotos = False
    target = 'dst'
    browser = 'opera'

    # Parse command line
    try:
        opts, args = getopt.getopt(sys.argv[1:], "hpb:f:t:d:i:", ["help"])
    except getopt.GetoptError:
        usage()
        sys.exit(2)
    for o, a in opts:
        if o in ('-h', '--help'):
            usage()
            sys.exit(0)
        if o == '-i':
            flickrUserId = a
        if o == '-p':
            getPhotos = True
        if o == '-f':
            dateLo = a
        if o == '-t':
            dateHi = a
        if o == '-d':
            target = a
        if o == '-b':
            browser = a

    # Check that we have a user id specified
    if flickrUserId == None:
        print "You need to specify a Flickr Id"
        sys.exit(1)

    # Check that the target directory exists
    if not os.path.isdir(target):
        print target + " is not a directory; please fix that."
        sys.exit(1)

    if dateHi == maxTime:
        t = time.time()
        print "For incremental backups, the current time is %.0f" % t
        print "You can rerun the program with '-f %.0f'" % t

    offlickr = Offlickr(flickrUserId, browser)
    photos = offlickr.getPhotoList(dateLo, dateHi)
    if photos == None:
        print "There seems to be a problem somewhere!"
        sys.exit(1)
    
    total = len(photos)
    print "Backing up" , total , "photos"

    i = 0
    for p in photos:
        i = i + 1
        pid = str(int(p['id'])) # Making sure we don't have weird things here
        print str(i) + "/" + str(total) + ": " + pid + ": " + p['title']
        # Get Metadata
        metadata = offlickr.getPhotoMetadata(pid)
        if metadata == None:
            print "Failed!"
            continue
        fn = target + '/' + pid + '.xml'
        f = open(fn, 'w')
        f.write(metadata[0])
        f.close()
        print 'Saved as ' + fn
        # Do we want the picture too?
        if getPhotos == False:
            continue
        f = pid + '.' + metadata[1]
        source = offlickr.getOriginalPhoto(pid)
        if source == None:
            print "Oopsie, no photo found"
        print 'Retrieving ' + source + ' as ' + f
        offlickr.downloadURL(source, target + '/' + f, verbose = True);
        print "\r... done!"

if __name__ == "__main__":
    main()