Download Jiandan OOXX gallery. http://jandan.net/ooxx

jdooxx.py

#!/usr/bin/env python

# Example:
# curl -s http://jandan.net/ooxx | ./jdooxx.py -o 5 -r 2.0 -u | wget -nv -i -
# This will download all pictures with oo >= 5 in the last page of Jiandan OOXX.

import HTMLParser
import sys
import getopt

class OOXXImage(object):

    def __init__(self):
        self.oo = 0
        self.xx = 0
        self.url = ""

    def test(self, ooMin, xxMax, ratio):
        if self.xx == 0:
            self.xx = 1
        return self.oo >= ooMin and self.xx <= xxMax and float(self.oo) / float(self.xx) >= ratio

class JDOOXXParser(HTMLParser.HTMLParser):

    def __init__(self):
        HTMLParser.HTMLParser.__init__(self)
        self.withinPostLi = False
        self.fetchData = None
        self.currentImage = OOXXImage()
        self.imageList = []

    def handle_starttag(self, tag, attrs):
        attrMap = dict(attrs)
        if self.withinPostLi == True:
            if tag == "img" and "class" not in attrMap:
                self.currentImage.url = attrMap["src"]
            elif tag == "span" and "id" in attrMap and attrMap["id"].startswith("cos_support-"):
                self.fetchData = "oo"
            elif tag == "span" and "id" in attrMap and attrMap["id"].startswith("cos_unsupport-"):
                self.fetchData = "xx"

        if tag == "li" and "id" in attrMap and attrMap["id"].startswith("comment-"):
            self.withinPostLi = True

    def handle_endtag(self, tag):
        if tag == "li" and self.withinPostLi == True:
            self.withinPostLi = False
            self.imageList.append(self.currentImage)
            self.currentImage = OOXXImage()

    def handle_data(self, data):
        if self.fetchData == "oo":
            self.currentImage.oo = int(data)
            self.fetchData = None
        elif self.fetchData == "xx":
            self.currentImage.xx = int(data)
            self.fetchData = None

def usage():
    print "A parser for Jiandan (http://jandan.net/ooxx) MM Gallery."
    print "Read from standard input and print image informations."
    print "  -h, --help         Print this infomation."
    print "  -o, --oomin=INT    Set the minimum OO value; images with a lower OO value will be excluded."
    print "  -x, --xxmax=INT    Set the maximum XX value; images with a greater XX value will be excluded."
    print "  -r, --ratio=FLOAT  Set the OO/XX ratio; images with lower ratio will be excluded"
    print "  -u, --urlonly      Only print the URLs; both OO and XX value will be omitted."

if __name__ == "__main__":
    try:
        opts, args = getopt.getopt(sys.argv[1:], "ho:x:r:u", ["help", "oomin=", "xxmax=", "ratio=", "urlonly"])
    except getopt.GetoptError as err:
        print str(err)
        usage()
        sys.exit(2)

    ooMin = 0
    xxMax = 1000
    ratio = -1
    urlOnly = False

    for o, a in opts:
        if o in ("-h", "--help"):
            usage()
            sys.exit()
        elif o in ("-o", "--oomin"):
            ooMin = int(a)
        elif o in ("-x", "--xxmax"):
            xxMax = int(a)
        elif o in ("-r", "--ratio"):
            ratio = float(a)
        elif o in ("-u", "--urlonly"):
            urlOnly = True

    parser = JDOOXXParser()
    parser.feed(sys.stdin.read().decode("utf-8"))

    for img in parser.imageList:
        if img.test(ooMin, xxMax, ratio):
            if urlOnly:
                print img.url
            else:
                print "oo = %3d, xx = %3d, %s" % (img.oo, img.xx, img.url)