Source Code for Module Bio.ExPASy.ScanProsite

  1  import urllib 
  2  from xml.sax import handler 
  3  from xml.sax.expatreader import ExpatParser 
  4   
  5   
  6 -class Record(list): 
  7      """\ 
  8  This record is a list containing the search results returned by 
  9  ScanProsite. The record also contains the data members n_match, n_seq, 
 10  capped, and warning.""" 
 11   
 12 -    def __init__(self): 
 13          self.n_match = None 
 14          self.n_seq = None 
 15          self.capped = None 
 16          self.warning = None 
 17   
 18   
 19 -def scan(seq="", mirror='http://www.expasy.org', output='xml', **keywords): 
 20      """Execute a ScanProsite search. 
 21   
 22      mirror:      The ScanProsite mirror to be used 
 23                   (default: http://www.expasy.org). 
 24      seq:         The query sequence, or UniProtKB (Swiss-Prot, 
 25                   TrEMBL) accession 
 26      output:      Format of the search results 
 27                   (default: xml) 
 28   
 29      Further search parameters can be passed as keywords; see the 
 30      documentation for programmatic access to ScanProsite at 
 31      http://www.expasy.org/tools/scanprosite/ScanPrositeREST.html 
 32      for a description of such parameters. 
 33   
 34      This function returns a handle to the search results returned by 
 35      ScanProsite. Search results in the XML format can be parsed into a 
 36      Python object, by using the Bio.ExPASy.ScanProsite.read function. 
 37      """ 
 38      parameters = {'seq': seq, 
 39                    'output': output} 
 40      for key, value in keywords.iteritems(): 
 41          if value is not None: 
 42              parameters[key] = value 
 43      command = urllib.urlencode(parameters) 
 44      url = "%s/cgi-bin/prosite/PSScan.cgi?%s" % (mirror, command) 
 45      handle = urllib.urlopen(url) 
 46      return handle 
 47   
 48   
 49 -def read(handle): 
 50      "Parse search results returned by ScanProsite into a Python object" 
 51      content_handler = ContentHandler() 
 52      saxparser = Parser() 
 53      saxparser.setContentHandler(content_handler) 
 54      saxparser.parse(handle) 
 55      record = content_handler.record 
 56      return record 
 57   
 58  # The functions below are considered private 
 59   
 60   
 61 -class Parser(ExpatParser): 
 62   
 63 -    def __init__(self): 
 64          ExpatParser.__init__(self) 
 65          self.firsttime = True 
 66   
 67 -    def feed(self, data, isFinal = 0): 
 68          # Error messages returned by the ScanProsite server are formatted as 
 69          # as plain text instead of an XML document. To catch such error 
 70          # messages, we override the feed method of the Expat parser. 
 71          # The error message is (hopefully) contained in the data that was just 
 72          # fed to the parser. 
 73          if self.firsttime: 
 74              if data[:5]!="<?xml": 
 75                  raise ValueError(data) 
 76          self.firsttime = False 
 77          return ExpatParser.feed(self, data, isFinal) 
 78   
 79   
 80 -class ContentHandler(handler.ContentHandler): 
 81      integers = ("start", "stop") 
 82      strings = ("sequence_ac", 
 83                 "sequence_id", 
 84                 "sequence_db", 
 85                 "signature_ac", 
 86                 "level", 
 87                 "level_tag") 
 88   
 89 -    def __init__(self): 
 90          self.element = [] 
 91   
 92 -    def startElement(self, name, attrs): 
 93          self.element.append(name) 
 94          self.content = "" 
 95          if self.element==["matchset"]: 
 96              self.record = Record() 
 97              self.record.n_match = int(attrs["n_match"]) 
 98              self.record.n_seq = int(attrs["n_seq"]) 
 99          elif self.element==["matchset", "match"]: 
100              match = {} 
101              self.record.append(match) 
102   
103 -    def endElement(self, name): 
104          assert name==self.element.pop() 
105          name = str(name) 
106          if self.element==["matchset", "match"]: 
107              match = self.record[-1] 
108              if name in ContentHandler.integers: 
109                  match[name] = int(self.content) 
110              elif name in ContentHandler.strings: 
111                  match[name] = self.content 
112              else: 
113                  # Unknown type, treat it as a string 
114                  match[name] = self.content 
115   
116 -    def characters(self, content): 
117          self.content += content 
118