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
python学习笔记:字典的使用示例详解
Jun 13 Python
Python检测网站链接是否已存在
Apr 07 Python
python实现xlsx文件分析详解
Jan 02 Python
运用TensorFlow进行简单实现线性回归、梯度下降示例
Mar 05 Python
Python一行代码实现快速排序的方法
Apr 30 Python
python GUI实现小球满屏乱跑效果
May 09 Python
python多进程(加入进程池)操作常见案例
Oct 21 Python
django实现web接口 python3模拟Post请求方式
Nov 19 Python
将python文件打包exe独立运行程序方法详解
Feb 12 Python
Python实现疫情通定时自动填写功能(附代码)
May 27 Python
Elasticsearch 批量操作
Apr 19 Python
Golang Web 框架Iris安装部署
Aug 14 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
深入array multisort排序原理的详解
2013/06/18 PHP
PHP之APC缓存详细介绍 apc模块安装
2014/01/13 PHP
PHP实现CSV文件的导入和导出类
2015/03/24 PHP
php数据访问之增删改查操作
2016/05/09 PHP
php 浮点数比较方法详解
2017/05/05 PHP
JavaScript中的this关键字介绍与使用实例
2013/06/21 Javascript
jquery $(&quot;#variable&quot;) 循环改变variable的值示例
2014/02/23 Javascript
JavaScript阻止浏览器返回按钮的方法
2015/03/18 Javascript
Javascript学习之谈谈JS的全局变量跟局部变量(推荐)
2016/08/28 Javascript
jQuery动画_动力节点节点Java学院整理
2017/07/04 jQuery
详解Angular如何正确的操作DOM
2018/07/06 Javascript
用vue-cli开发vue时的代理设置方法
2018/09/20 Javascript
每天学点Vue源码之vm.$mount挂载函数
2019/03/11 Javascript
axios封装,使用拦截器统一处理接口,超详细的教程(推荐)
2019/05/02 Javascript
js代码实现轮播图
2020/05/04 Javascript
Python表示矩阵的方法分析
2017/05/26 Python
Python中django学习心得
2017/12/06 Python
Python系统监控模块psutil功能与经典用法分析
2018/05/24 Python
python图像和办公文档处理总结
2019/05/28 Python
django获取from表单multiple-select的value和id的方法
2019/07/19 Python
Python pip 安装与使用(安装、更新、删除)
2019/10/06 Python
Django 用户登陆访问限制实例 @login_required
2020/05/13 Python
为什么python比较流行
2020/06/19 Python
通过一张图教会你CSS3倒影的实现
2017/09/26 HTML / CSS
德国低价购买灯具和家具网站:Style-home.de
2016/11/25 全球购物
英国家具、照明、家居用品网上商店:Wayfair.co.uk
2020/02/13 全球购物
C有"按引用传递"吗
2016/09/06 面试题
判断单链表中是否存在环
2012/07/16 面试题
涉外文秘个人求职的自我评价
2013/10/07 职场文书
优秀团员个人事迹材料
2014/01/29 职场文书
技能竞赛活动方案
2014/02/21 职场文书
2014年元旦促销活动方案
2014/02/22 职场文书
退税申请报告怎么写
2015/05/18 职场文书
准备去美国留学,那么大学申请文书应该怎么写?
2019/08/12 职场文书
MYSQL 无法识别中文的永久解决方法
2021/06/03 MySQL
SQL优化老出错,那是你没弄明白MySQL解释计划用法
2021/11/27 MySQL