BitTorrent文件解析：

BitTorrent文件使用bencode编码，其中包括了4种数据类型：

'd' 开头表示是dict类型，'e'表示结束

'l' （小写字母L）开头表示是list类型，'e'表示结束

'i'开头表示是integer类型，'e'表示结束，可以表示负数

以数字开头表示string类型，数字为string长度，长度与string内容以':'分割

默认所有text类型的属性为utf-8编码，但是大多数BitTorrent包含codepage 和 encoding属性，指定了text的编码格式

"announce" -- tracker服务器的地址，为string
"info" ---文件信息，为dict类型
　　"name" --单文件模式，表示文件名，多文件模式表示根目录名。
　　"length" --单文件模式表示文件长度，多文件模式不存在
　　"piece length" --文件分片大小
　　"pieces" --为一个长string, 没20个字节表示一个分片的SHA1 hash值。按照文件分片的顺序排列。
　　　　　　　　分片是按照所以文件组合在一起进行的，即一个分片可能会跨越多个文件。
　　"files" -- 多文件模式存在，为一个文件列表，每个文件为一个dict类型
　　　　　　"path" -- 文件目录列表，最后一项为文件名
　　　　　　"length" --文件长度

"peace length"  --分片大小

以下为draft bep定义的属性
"code page"
"announce-list" --tracker列表，为二维数组，即将tracker服务器分为多个组
"encoding" -- Text属性的编码类型，string 类型，如 UTF-8
"publisher" -- 发布者
"publisher url" --发布者 URL
"creater" --创建者，如btcomet,btspirit
"creation date" --创建日期，为UTC格式，需要转化为本地时区可读格式
"commnent" --注释
"nodes" -- DHT 节点列表

BitTorrent的标准参见：http://www.bittorrent.org/beps/bep_0003.html

以下是自己写的Python实现，初学Python，代码写起来还都是C/C++风格，慢慢改进吧。

 

from datetime import datetime  2   3 import bcodec  4   5 _READ_MAX_LEN = -1  6   7 class BTFormatError(BaseException):  8     pass  9      10 class TorrentFile(object): 11      12     __metainfo = {} 13     __file_name = '' 14      15     def read_file(self, filename): 16          17         torrent_file = open(filename, 'rb') 18         data = torrent_file.read(_READ_MAX_LEN) 19         torrent_file.close() 20          21         data = list(data) 22         metainfo = bcodec.bdecode(data) 23         if metainfo != None and type(metainfo) == type({}): 24             self.__file_name = filename 25             self.__metainfo = metainfo 26         else: 27             raise BTFormatError() 28            29     def __is_singlefile(self): 30          31         return self.__get_meta_info('length') != None 32      33     def __decode_text(self, text): 34         encoding = 'utf-8' 35         resultstr = '' 36         if self.get_encoding() != None: 37             encoding = self.get_encoding() 38         elif self.get_codepage() != None: 39             encoding = 'cp' + str(self.get_codepage()) 40         if text: 41             try: 42                 resultstr = text.decode(encoding=encoding) 43             except ValueError: 44                 return text 45         else: 46             return None 47         return resultstr 48      49     def __get_meta_top(self, key): 50         if key in self.__metainfo.keys(): 51             return self.__metainfo[key] 52         else: 53             return None 54     def __get_meta_info(self,key): 55         meta_info = self.__get_meta_top('info') 56         if meta_info != None and key in meta_info.keys(): 57                 return meta_info[key] 58         return None 59      60     def get_codepage(self): 61         return self.__get_meta_top('codepage') 62     def get_encoding(self): 63         return self.__get_meta_top('encoding') 64      65     def get_announces(self): 66         announces = self.__get_meta_top('announce-list') 67         if announces != None: 68             return announces 69          70         announces = [[]] 71         ann = self.__get_meta_top('announce') 72         if ann: 73             announces[0].append(ann) 74         return announces 75      76     def get_publisher(self): 77         return self.__decode_text(self.__get_meta_top('publisher')) 78     def get_publisher_url(self): 79         return self.__decode_text(self.__get_meta_top('publisher-url')) 80      81     def get_creater(self): 82         return self.__decode_text(self.__get_meta_top('created by')) 83     def get_creation_date(self): 84         utc_date = self.__get_meta_top('creation date') 85         if utc_date == None: 86             return utc_date 87         creationdate = datetime.utcfromtimestamp(utc_date) 88         return creationdate 89     def get_comment(self): 90         return self.__get_meta_top('comment') 91            92     def get_nodes(self): 93         return self.__get_meta_top('nodes') 94      95     def get_piece_length(self): 96         return self.__get_meta_info('piece length') 97      98     def get_piece(self, index): 99         pieces = self.__get_meta_info('pieces')100         if pieces == None:101             return None102         103         offset = index*20104         if offset+20 > len(pieces):105             return None106         return pieces[offset:offset+20]107     108     def get_files(self):109         110         files = []111         name = self.__decode_text(self.__get_meta_info('name'))112         piece_length = self.get_piece_length()113         if name == None:114             return files115         116         if self.__is_singlefile():117             file_name = name118             file_length = self.__get_meta_info('length')119             if not file_length:120                 return files121             122             pieces_num = file_length/piece_length123             last_piece_offset =  file_length % piece_length124             if last_piece_offset != 0:125                 pieces_num = int(pieces_num) + 1126                 last_piece_offset -= 1127             else:128                 last_piece_offset = piece_length - 1129 130             first_piece_offset = 0131             132             files.append({'name':[file_name], 'length':file_length, 'first-piece':(0, first_piece_offset), 'last-piece':(pieces_num-1,last_piece_offset)})133             return files134         135         folder = name136         meta_files = self.__get_meta_info('files')137         if meta_files == None:138             return files139         140         total_length = int(0)141         for one_file in self.__get_meta_info('files'):142             143             file_info = {}144             path_list = []145             path_list.append(folder)146                         147             if 'path' not in one_file.keys():148                 break149             for path in one_file['path']:150                 path_list.append(self.__decode_text(path))151             file_info['name'] = path_list152             153             if 'length' not in one_file.keys():154                 break155             156             file_info['length'] =  one_file['length']157             158             piece_index = total_length / piece_length159             first_piece_offset =  total_length % piece_length160             161             total_length += one_file['length']162             pieces_num = total_length / piece_length - piece_index163             last_piece_offset = total_length % piece_length164             165             if last_piece_offset != 0:166                 pieces_num += 1167                 last_piece_offset -= 1168             else:169                 last_piece_offset = piece_length - 1170             171             file_info['first-piece'] = (piece_index,first_piece_offset)172             file_info['last-piece'] = ((piece_index+pieces_num-1),last_piece_offset)173             files.append(file_info)174         return files175     176 if __name__ == '__main__':177     filename = r".\huapi2.torrent"178 179     torrent = TorrentFile()180 181     print "begin to read file"182     try:183         torrent.read_file(filename)184     except (IOError,BTFormatError), reason:185         print "Read bittorrent file error! Error:%s" %reason186      187     print "end to read file"188 189     print "announces: " , torrent.get_announces() 190     print "peace length:", torrent.get_piece_length()191     print "code page:" , torrent.get_codepage()192     print "encoding:" , torrent.get_encoding()193     print "publisher:" ,torrent.get_publisher()194     print "publisher url:", torrent.get_publisher_url()195     print "creater:" , torrent.get_creater()196     print "creation date:", torrent.get_creation_date()197     print "commnent:", torrent.get_comment()198     print "nodes:", torrent.get_nodes()199     torrent.get_files()200     for one_file in torrent.get_files():201         print 'name:', '\\'.join(one_file['name'])202         print 'length:', one_file['length']203         print 'first-piece:', one_file['first-piece']204         print 'last-piece:', one_file['last-piece']

 

'''  2 Created on 2012-9-30  3   4 @author: ddt  5 '''  6 def bdecode(data):  7     data = list(data)  8     return _read_chunk(data)  9      10 def _read_chunk(data): 11      12     chunk = None 13      14     if len(data) == 0: 15         return chunk 16      17     leading_chr = data[0] 18     #print leading_chr,                  19     if leading_chr.isdigit(): 20         chunk = _read_string(data) 21         #print chunk 22     elif leading_chr == 'd': 23         chunk = _read_dict(data) 24         #print chunk is None 25     elif leading_chr == 'i': 26         chunk = _read_integer(data) 27         #print chunk 28     elif leading_chr == 'l': 29         chunk = _read_list(data) 30  31     #print chunk 32     return chunk 33                             34 def _read_dict(data): 35      36     if  len(data) == 0 or data.pop(0) != 'd':  37         return None 38      39     chunk = {}  40     while len(data) > 0 and data[0] != 'e': 41          42         key = _read_chunk(data) 43         value = _read_chunk(data) 44          45         if key != None and value != None and type(key) == type(''): 46             if key in chunk.keys(): 47                 value_type = type(chunk[key]) 48                 if value_type == type([]): 49                     chunk[key] += value 50                 elif value_type == type({}): 51                     chunk[key].update(value) 52                 else: 53                     chunk[key] = value 54             else: 55                 chunk[key] = value 56                 #print key 57         else: 58             return None 59          60     if len(data) == 0 or data.pop(0) != 'e': 61         return None 62      63     return chunk 64  65 def _read_list(data): 66  67     if  len(data) == 0 or data.pop(0) != 'l':  68         return None 69      70     chunk = [] 71     while len(data) > 0 and data[0] != 'e': 72         value = _read_chunk(data) 73         if value != None: 74             chunk.append(value) 75         else: 76             return None 77          78     if len(data) == 0 or data.pop(0) != 'e':  79         return None 80  81     return chunk 82  83 def _read_string(data): 84      85     str_len = '' 86     while len(data) > 0 and data[0].isdigit(): 87         str_len +=  data.pop(0) 88      89     if len(data) == 0 or data.pop(0) != ':': 90         return None 91      92     str_len = int(str_len) 93     if str_len > len(data): 94         return None 95      96     value = data[0:str_len] 97     del data[0:str_len] 98     return ''.join(value) 99 100 def _read_integer(data):101    102     integer = ''103     if len(data) < len('i2e') or data.pop(0) != 'i': 104         return None105     106     sign = data.pop(0)107     if sign != '-' and not sign.isdigit():108         return None109     integer += sign110     111     while len(data) > 0 and data[0].isdigit():112         integer += data.pop(0)113     114     if len(data) == 0 or data.pop(0) != 'e':115         return None116     return  int(integer)117 118 def bencode(data):119     result = None120     data_type = type(data)121     if data_type == type({}):122         result = _write_dict(data)123     elif data_type == type([]):124         result = _write_list(data)125     elif data_type == type(''):126         result = _write_string(data)127     elif data_type == type(int(0)):128         result = _write_integer(data)129     return result130 131 def _write_dict(data):132     result = 'd'133     for key, value in data.items():134         key_encode = bencode(key)135         value_encode = bencode(value)136         if key_encode != None and value_encode != None:137             result += key_encode138             result += value_encode139         else:140             return None141     result += 'e'142     return result143 144 def _write_list(data):145     result = 'l'146     for value in data:147         value_encode = bencode(value)148         if value_encode != None:149             result += value_encode150         else:151             return None152     result += 'e'153     return result154 155 def _write_string(data):156     result = '%d' %len(data)157     result += ':'158     result += data159     return result160 161 def _write_integer(data):162     result = 'i'163     result += '%d' %data164     result += 'e'165     return result