Source Code for Module usufy.wrappers.platforms

  1  # -*- coding: cp1252 -*- 
  2  # 
  3  ################################################################################## 
  4  # 
  5  #       This file is part of usufy.py. 
  6  # 
  7  #       Usufy is free software: you can redistribute it and/or modify 
  8  #       it under the terms of the GNU General Public License as published by 
  9  #       the Free Software Foundation, either version 3 of the License, or 
 10  #       (at your option) any later version. 
 11  # 
 12  #       This program is distributed in the hope that it will be useful, 
 13  #       but WITHOUT ANY WARRANTY; without even the implied warranty of 
 14  #       MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the 
 15  #       GNU General Public License for more details. 
 16  # 
 17  #       You should have received a copy of the GNU General Public License 
 18  #       along with this program.  If not, see <http://www.gnu.org/licenses/>. 
 19  # 
 20  ################################################################################## 
 21   
 22  import urllib2 
 23  import os 
 24   
 25 -class Platform(): 
 26          """  
 27                  <Platform> abstract class. 
 28          """ 
 29 -        def __init__(self): 
 30                  """  
 31                          Constructor without parameters... 
 32                  """ 
 33                  #self.platformName = "Abstract Class" 
 34                  # These tags will be the one used to label this platform 
 35                  #self.tags = [] 
 36                  # CONSTANT OF TEXT TO REPLACE 
 37                  #self.NICK_WILDCARD = "HERE_GOES_THE_NICK" 
 38                  # Usually it will contain a phrase like  \"<HERE_GOES_THE_NICK>\" that will be the place where the nick will be included 
 39                  #self.url = "http://test.com/" + self.NICK_WILDCARD 
 40                  # Text to find when the user was NOT found 
 41                  #self.notFoundText = [] 
 42                  #self.score= 0.0 
 43                  pass 
 44   
 45 -        def _genURL(self, nick): 
 46                  """      
 47                          Private method that returns an URL for a given nick.  
 48                           
 49                          Return values: 
 50                                  string containing a URL 
 51                  """ 
 52                  return self.url.replace(self.NICK_WILDCARD, nick) 
 53   
 54 -        def _getResourceFromUser(self, url): 
 55                  """  
 56                          Este método privado de la clase padre puede ser sobreescrito por cada clase hija si la verificación 
 57                          a realizar es más compleja que la verificación estándar. 
 58   
 59                          Valores retornados: 
 60                                  html    Si el usuario en cuestión existe en esta red social. 
 61                                  None    Si el usuario en cuestión no existe en esta red social. 
 62                  """ 
 63                  try: 
 64                          recurso = urllib2.urlopen(url) 
 65                  except : 
 66                          # no se ha conseguido retornar una URL de perfil, por lo que se devuelve None 
 67                          # print ">\tERROR: Something happened when trying to recover the resource..." 
 68                          return None 
 69                  html = recurso.read() 
 70           
 71                  for t in self.notFoundText: 
 72                          # si se encuentra el contenido en cuestión que confirma que el usuario no existe 
 73                          if t in html: 
 74                                  # Returning that it does not exist 
 75                                  return None 
 76                  return html 
 77           
 78 -        def getUserPage(self, nick, outputF=None, avoidProcessing=True): 
 79                  """  
 80                          Este método público es el que se encarga de recuperar la información de la página del usuario a buscar y de procesar si procede la descrga. 
 81                           
 82                          List of parameters used by this method: 
 83                                  nick:           nick to search 
 84                                  outputF:        will contain a valid path to the outputFolder 
 85                                  avoidProcessing:will define whether a further process is performed 
 86           
 87                          Return values: 
 88                                  url     URL del usuario en cuestión una vez que se haya confirmado su validez. 
 89                                  None    En el caso de que no se haya podido obtener una URL válida. 
 90                  """ 
 91                  # generando la URL para el nick dado 
 92                  url = self._genURL(nick) 
 93                  # en función de la respuesta, se hace la comprobación de si el perfil existe o no 
 94                  html = self._getResourceFromUser(url)  
 95                  if html != None: 
 96                          if not avoidProcessing: 
 97                                  # Storing file if the user has NOT said to avoid the process...          
 98                                           
 99                                  outputPath = os.path.join(outputF, nick) 
100                                  if not os.path.exists(outputPath): 
101                                          os.makedirs(outputPath) 
102                                  self._processProfile(os.path.join(outputPath, nick + "_" + str(self).lower()), html) 
103                          return url 
104                  else: 
105                  #       raise Exception, "UserNotFoundException: the user was not found in " + self.socialNetworkName 
106                          return None 
107   
108 -        def _processProfile(self, oP, html): 
109                  """      
110                          Method to process an URL depending on the functioning of the site. By default, it stores the html downloaded on a file. 
111                          This method might be overwritten by each and every class to perform different actions such as indexing the contents with tools like pysolr, for example. 
112   
113                          Return values: 
114                                  None 
115                  """ 
116                  with open (oP, "w") as oF: 
117                          oF.write(html) 
118                  return True 
119   
120 -        def __str__(self): 
121                  """  
122                          Función para obtener el texto que se representará a la hora de imprimir el objeto. 
123                           
124                          Return values: 
125                                  self.platformName 
126                  """ 
127                  return self.platformName 
128