Source Code for Module Bio.GenBank.utils

 1  """Useful utilities for helping in parsing GenBank files. 
 2  """ 
 3   
 4   
 5 -class FeatureValueCleaner(object): 
 6      """Provide specialized capabilities for cleaning up values in features. 
 7   
 8      This class is designed to provide a mechanism to clean up and process 
 9      values in the key/value pairs of GenBank features. This is useful 
10      because in cases like: 
11   
12           /translation="MED 
13           YDPWNLRFQSKYKSRDA" 
14   
15      you'll end up with a value with \012s and spaces in it like: 
16          "MED\012 YDPWEL..." 
17   
18      which you probably don't want. 
19   
20      This cleaning needs to be done on a case by case basis since it is 
21      impossible to interpret whether you should be concatenating everything 
22      (as in translations), or combining things with spaces (as might be 
23      the case with /notes). 
24      """ 
25      keys_to_process = ["translation"] 
26   
27 -    def __init__(self, to_process = keys_to_process): 
28          """Initialize with the keys we should deal with. 
29          """ 
30          self._to_process = to_process 
31   
32 -    def clean_value(self, key_name, value): 
33          """Clean the specified value and return it. 
34   
35          If the value is not specified to be dealt with, the original value 
36          will be returned. 
37          """ 
38          if key_name in self._to_process: 
39              try: 
40                  cleaner = getattr(self, "_clean_%s" % key_name) 
41                  value = cleaner(value) 
42              except AttributeError: 
43                  raise AssertionError("No function to clean key: %s" 
44                                       % key_name) 
45          return value 
46   
47 -    def _clean_translation(self, value): 
48          """Concatenate a translation value to one long protein string. 
49          """ 
50          translation_parts = value.split() 
51          return "".join(translation_parts) 
52