Python开发实例分享bt种子爬虫程序和种子解析


Posted in Python onMay 21, 2014

看到网上也有开源的代码,这不,我拿来进行了二次重写,呵呵,上代码:

    #encoding: utf-8  
    import socket  
    from hashlib import sha1  
    from random import randint  
    from struct import unpack, pack  
    from socket import inet_aton, inet_ntoa  
    from bisect import bisect_left  
    from threading import Timer  
    from time import sleep  
    import MySQLdb  
    from datetime import *  
    import time  
    from bencode import bencode, bdecode  
    BOOTSTRAP_NODES = [  
        ("router.bittorrent.com", 6881),  
        ("dht.transmissionbt.com", 6881),  
        ("router.utorrent.com", 6881)  
    ]   
    TID_LENGTH = 4  
    KRPC_TIMEOUT = 10  
    REBORN_TIME = 5 * 60  
    K = 8  
    def entropy(bytes):  
        s = ""  
        for i in range(bytes):  
            s += chr(randint(0, 255))  
        return s  
    def random_id():  
        hash = sha1()  
        hash.update( entropy(20) )  
        return hash.digest()  
    def decode_nodes(nodes):  
        n = []  
        length = len(nodes)  
        if (length % 26) != 0:   
            return n  
        for i in range(0, length, 26):  
            nid = nodes[i:i+20]  
            ip = inet_ntoa(nodes[i+20:i+24])  
            port = unpack("!H", nodes[i+24:i+26])[0]  
            n.append( (nid, ip, port) )  
        return n  
    def encode_nodes(nodes):  
        strings = []  
        for node in nodes:  
            s = "%s%s%s" % (node.nid, inet_aton(node.ip), pack("!H", node.port))  
            strings.append(s)  
        return "".join(strings)  
    def intify(hstr):  
        return long(hstr.encode('hex'), 16)      
    def timer(t, f):  
        Timer(t, f).start()  
    class BucketFull(Exception):  
        pass  
    class KRPC(object):  
        def __init__(self):  
            self.types = {  
                "r": self.response_received,  
                "q": self.query_received  
            }  
            self.actions = {  
                "ping": self.ping_received,  
                "find_node": self.find_node_received,  
                "get_peers": self.get_peers_received,  
                "announce_peer": self.announce_peer_received,  
            }  
            self.socket = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)  
            self.socket.bind(("0.0.0.0", self.port))  
        def response_received(self, msg, address):  
            self.find_node_handler(msg)  
        def query_received(self, msg, address):  
            try:  
                self.actions[msg["q"]](msg, address)  
            except KeyError:  
                pass  
        def send_krpc(self, msg, address):  
            try:  
                self.socket.sendto(bencode(msg), address)  
            except:  
                pass  
    class Client(KRPC):  
        def __init__(self, table):  
            self.table = table  
            timer(KRPC_TIMEOUT, self.timeout)  
            timer(REBORN_TIME, self.reborn)  
            KRPC.__init__(self)  
        def find_node(self, address, nid=None):  
            nid = self.get_neighbor(nid) if nid else self.table.nid  
            tid = entropy(TID_LENGTH)  
            msg = {  
                "t": tid,  
                "y": "q",  
                "q": "find_node",  
                "a": {"id": nid, "target": random_id()}  
            }  
            self.send_krpc(msg, address)  
        def find_node_handler(self, msg):  
            try:  
                nodes = decode_nodes(msg["r"]["nodes"])  
                for node in nodes:  
                    (nid, ip, port) = node  
                    if len(nid) != 20: continue  
                    if nid == self.table.nid: continue  
                    self.find_node( (ip, port), nid )  
            except KeyError:  
                pass  
        def joinDHT(self):  
            for address in BOOTSTRAP_NODES:   
                self.find_node(address)  
        def timeout(self):  
            if len( self.table.buckets ) < 2:  
                self.joinDHT()  
            timer(KRPC_TIMEOUT, self.timeout)  
        def reborn(self):  
            self.table.nid = random_id()  
            self.table.buckets = [ KBucket(0, 2**160) ]  
            timer(REBORN_TIME, self.reborn)  
        def start(self):  
            self.joinDHT()  
            while True:  
                try:  
                    (data, address) = self.socket.recvfrom(65536)  
                    msg = bdecode(data)  
                    self.types[msg["y"]](msg, address)  
                except Exception:  
                    pass  
        def get_neighbor(self, target):  
            return target[:10]+random_id()[10:]  
    class Server(Client):  
        def __init__(self, master, table, port):  
            self.table = table  
            self.master = master  
            self.port = port  
            Client.__init__(self, table)  
        def ping_received(self, msg, address):  
            try:  
                nid = msg["a"]["id"]  
                msg = {  
                    "t": msg["t"],  
                    "y": "r",  
                    "r": {"id": self.get_neighbor(nid)}  
                }  
                self.send_krpc(msg, address)  
                self.find_node(address, nid)  
            except KeyError:  
                pass  
        def find_node_received(self, msg, address):  
            try:  
                target = msg["a"]["target"]  
                neighbors = self.table.get_neighbors(target)  
                nid = msg["a"]["id"]  
                msg = {  
                    "t": msg["t"],  
                    "y": "r",  
                    "r": {  
                        "id": self.get_neighbor(target),   
                        "nodes": encode_nodes(neighbors)  
                    }  
                }  
                self.table.append(KNode(nid, *address))  
                self.send_krpc(msg, address)  
                self.find_node(address, nid)  
            except KeyError:  
                pass  
        def get_peers_received(self, msg, address):  
            try:  
                infohash = msg["a"]["info_hash"]  
                neighbors = self.table.get_neighbors(infohash)  
                nid = msg["a"]["id"]  
                msg = {  
                    "t": msg["t"],  
                    "y": "r",  
                    "r": {  
                        "id": self.get_neighbor(infohash),   
                        "nodes": encode_nodes(neighbors)  
                    }  
                }  
                self.table.append(KNode(nid, *address))  
                self.send_krpc(msg, address)  
                self.master.log(infohash)  
                self.find_node(address, nid)  
            except KeyError:  
                pass  
        def announce_peer_received(self, msg, address):  
            try:  
                infohash = msg["a"]["info_hash"]  
                nid = msg["a"]["id"]  
                msg = {   
                    "t": msg["t"],  
                    "y": "r",  
                    "r": {"id": self.get_neighbor(infohash)}  
                }  
                self.table.append(KNode(nid, *address))  
                self.send_krpc(msg, address)  
                self.master.log(infohash)  
                self.find_node(address, nid)  
            except KeyError:  
                pass  
    class KTable(object):  
        def __init__(self, nid):  
            self.nid = nid  
            self.buckets = [ KBucket(0, 2**160) ]  
        def append(self, node):  
            index = self.bucket_index(node.nid)  
            try:  
                bucket = self.buckets[index]  
                bucket.append(node)  
            except IndexError:  
                return  
            except BucketFull:  
                if not bucket.in_range(self.nid): return  
                self.split_bucket(index)  
                self.append(node)  
        def get_neighbors(self, target):  
            nodes = []  
            if len(self.buckets) == 0: return nodes  
            if len(target) != 20 : return nodes  
            index = self.bucket_index(target)  
            try:  
                nodes = self.buckets[index].nodes  
                min = index - 1  
                max = index + 1  
                while len(nodes) < K and ((min >= 0) or (max < len(self.buckets))):  
                    if min >= 0:  
                        nodes.extend(self.buckets[min].nodes)  
                    if max < len(self.buckets):  
                        nodes.extend(self.buckets[max].nodes)  
                    min -= 1  
                    max += 1  
                num = intify(target)  
                nodes.sort(lambda a, b, num=num: cmp(num^intify(a.nid), num^intify(b.nid)))  
                return nodes[:K]  
            except IndexError:  
                return nodes  
        def bucket_index(self, target):  
            return bisect_left(self.buckets, intify(target))  
        def split_bucket(self, index):  
            old = self.buckets[index]  
            point = old.max - (old.max - old.min)/2  
            new = KBucket(point, old.max)  
            old.max = point  
            self.buckets.insert(index + 1, new)  
            for node in old.nodes[:]:  
                if new.in_range(node.nid):  
                    new.append(node)  
                    old.remove(node)  
        def __iter__(self):  
            for bucket in self.buckets:  
                yield bucket  
    class KBucket(object):  
        __slots__ = ("min", "max", "nodes")  
        def __init__(self, min, max):  
            self.min = min  
            self.max = max  
            self.nodes = []  
        def append(self, node):  
            if node in self:  
                self.remove(node)  
                self.nodes.append(node)  
            else:  
                if len(self) < K:  
                    self.nodes.append(node)  
                else:  
                    raise BucketFull  
        def remove(self, node):  
            self.nodes.remove(node)  
        def in_range(self, target):  
            return self.min <= intify(target) < self.max  
        def __len__(self):  
            return len(self.nodes)  
        def __contains__(self, node):  
            return node in self.nodes  
        def __iter__(self):  
            for node in self.nodes:  
                yield node  
        def __lt__(self, target):  
            return self.max <= target  
    class KNode(object):  
        __slots__ = ("nid", "ip", "port")  
        def __init__(self, nid, ip, port):  
            self.nid = nid  
            self.ip = ip  
            self.port = port  
        def __eq__(self, other):  
            return self.nid == other.nid  
    #using example  
    class Master(object):  
        def __init__(self, f):  
            self.f = f  
            try:  
                self.conn=MySQLdb.connect(host='localhost',user='root',passwd='',db='bt',port=3306)  
                self.cur=self.conn.cursor()  
            except MySQLdb.Error,e:  
                print "Mysql Error %d: %s" % (e.args[0], e.args[1])  
        def log(self, infohash):  
            try:  
                sql = "insert into bt_main_new(hash,name,length,date) values(%s,%s,%s,%s)"  
                date = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())  
                re=self.cur.execute(sql,(infohash,'','',date))  
                self.conn.commit()  
                self.cur.close()  
                self.conn.close()  
                #print re  
            except MySQLdb.Error,e:  
                print "Mysql Error %d: %s" % (e.args[0], e.args[1])  
            self.f.write(infohash.encode("hex")+"\n")  
            self.f.flush()  
    try:  
        d = date.today()  
        f = open("%s.log" % d, "a")  
        m = Master(f)  
        s = Server(Master(f), KTable(random_id()), 8006)  
        s.start()       
    except KeyboardInterrupt:  
        s.socket.close()  
        f.close() 

本爬虫程序,会自动爬取得网络上分享的bt种子,写入文件盒数据库,爬取的只是个种子的hash码,还需要到网络上下载种子进行分析

下载种子,相信大家都知道国外有几个免费分享种子的网站,大家可以根据hash码去下载,分析,下面呈上我写的一个分析种子的程序:

#! /usr/bin/python  
# -*- coding: utf-8 -*-  
import MySQLdb  
from datetime import *  
import time  
import re  
from time import sleep  
import bencode  
import urllib2  
import base64  
try:  
    conn=MySQLdb.connect(host='localhost',user='root',passwd='',db='bt',port=3306)  
    cur=conn.cursor()  
    sql = "select * from bt_main where name = '' order by id desc"  
    count = cur.execute(sql)  
    rows = cur.fetchall()  
    for row in rows:  
        if row[2].strip() != '':  
            continue  
        id = row[0]  
        hash = row[1]  
        url = "http://haofuli.duapp.com/go/info.php?hash=%s" % hash  
        file = urllib2.urlopen(url).read()  
        if "error!" == file:  
            try:  
                sql = "update bt_main set isTrue = 0 where id = %s "  
                re = cur.execute(sql,(id))  
                conn.commit()  
            except MySQLdb.Error,e:  
                print "Mysql Error %d: %s" % (e.args[0], e.args[1])  
        else:  
            #decode  
            try:  
                fileEncode = bencode.bdecode(file)  
            except Exception,e:pass  
            if 'name.utf-8' in fileEncode['info']:  
                filename=fileEncode['info']['name.utf-8']  
            else:  
                filename = fileEncode['info']['name']  
            ##length  
            if "length" in fileEncode['info']:  
                length = fileEncode['info']['length']  
            else:  
                length = 0  
            try:  
                sql = "update bt_main set name = %s , length = %s , isTrue = 1 where id = %s"  
                re = cur.execute(sql,(base64.b64encode(filename),length,id))  
                conn.commit()  
            except MySQLdb.Error,e:  
                print "Mysql Error %d: %s" % (e.args[0], e.args[1])  
except MySQLdb.Error,e:  
    print "Mysql Error %d: %s" % (e.args[0], e.args[1])

上面的只是简单的分析,对于多文件的,还没有处理。我最近在解析种子的时候,总是出现莫名的填充文件的问题,可能是版本过低吧,最近仍旧在解决。

Python 相关文章推荐
pyv8学习python和javascript变量进行交互
Dec 04 Python
用python删除java文件头上版权信息的方法
Jul 31 Python
Python isinstance函数介绍
Apr 14 Python
使用Python的Django框架实现事务交易管理的教程
Apr 20 Python
python使用smtplib模块通过gmail实现邮件发送的方法
May 08 Python
python生成多个只含0,1元素的随机数组或列表的实例
Nov 12 Python
基于Python实现迪杰斯特拉和弗洛伊德算法
May 27 Python
Python 文件操作之读取文件(read),文件指针与写入文件(write),文件打开方式示例
Sep 29 Python
详解python中docx库的安装过程
Nov 08 Python
python爬虫基础知识点整理
Jun 02 Python
解决python运行效率不高的问题
Jul 20 Python
python利用线程实现多任务
Sep 18 Python
从零学Python之引用和类属性的初步理解
May 15 #Python
python中xrange和range的区别
May 13 #Python
Python中os和shutil模块实用方法集锦
May 13 #Python
Python中的jquery PyQuery库使用小结
May 13 #Python
Python getopt模块处理命令行选项实例
May 13 #Python
Python random模块(获取随机数)常用方法和使用例子
May 13 #Python
Python自动化测试工具Splinter简介和使用实例
May 13 #Python
You might like
用php和MySql来与ODBC数据连接
2006/10/09 PHP
Zend Framework实现将session存储在memcache中的方法
2016/03/22 PHP
模仿百度三维地图的js数据分享
2011/05/12 Javascript
js捕获鼠标右键菜单中的粘帖事件实现代码
2013/04/01 Javascript
浅谈Javascript中深复制
2014/12/01 Javascript
JavaScript生成随机字符串的方法
2015/03/19 Javascript
jquery+CSS实现的水平布局多级网页菜单效果
2015/08/24 Javascript
浅谈jQuery this和$(this)的区别及获取$(this)子元素对象的方法
2016/11/29 Javascript
JS Select下拉框(支持输入模糊查询)
2017/02/04 Javascript
javascript滚轮事件基础实例讲解(37)
2017/02/14 Javascript
Canvas实现放射线动画效果
2017/02/15 Javascript
Vue动态组件实例解析
2017/08/20 Javascript
Vue.js移动端左滑删除组件的实现代码
2017/09/08 Javascript
常用的9个JavaScript图表库详解
2017/12/19 Javascript
jQuery滚动条美化插件nicescroll简单用法示例
2018/04/18 jQuery
详解datagrid使用方法(重要)
2020/11/06 Javascript
vue解决跨域问题(推荐)
2020/11/10 Javascript
js删除指定位置超链接中含有百度与360的标题
2021/01/06 Javascript
详解微信小程序轨迹回放实现及遇到的坑
2021/02/02 Javascript
[02:35]DOTA2超级联赛专访XB 难忘一年九冠称王
2013/06/20 DOTA
tensorflow 加载部分变量的实例讲解
2018/07/27 Python
在PyCharm中实现关闭一个死循环程序的方法
2018/11/29 Python
python 动态迁移solr数据过程解析
2019/09/04 Python
纯CSS3编写的的精美动画进度条(无flash/无图像/无脚本/附源码)
2013/01/07 HTML / CSS
3种方式实现瀑布流布局小结
2019/09/05 HTML / CSS
大韩航空官方网站:Korean Air
2017/10/25 全球购物
倩碧澳大利亚官网:Clinique澳大利亚
2019/07/22 全球购物
英国豪华装饰照明品牌的在线零售商:Inspyer Lighting
2019/12/10 全球购物
STP协议的主要用途是什么?为什么要用STP
2012/12/20 面试题
简单叙述一下MYSQL的优化
2016/05/09 面试题
2014年关于两会精神的心得体会
2014/03/17 职场文书
实习推荐信
2014/05/10 职场文书
财务检查整改报告
2014/11/06 职场文书
毕业生学校组织意见
2015/06/04 职场文书
导游词之日月潭
2019/11/05 职场文书
python3.7.2 tkinter entry框限定输入数字的操作
2021/05/22 Python