colly.py

   1 import sys
   2 import csv
   3
   4 class Colly(object):
   5     _setA = None
   6     _setB = None
   7
   8     ''' Csv parsing and formatting
   9     '''
  10     def __init__(self, a, b, aM=[], bM=[]):
  11         self.setA = self._prep(a, *aM)
  12         self.setB = self._prep(b, *bM)
  13         if self.setA == None or self.setB == None:
  14             print "Operation failed"
  15             raise
  16
  17     def _prep(self, csv_file, col=0, start=0, end=sys.maxint):
  18         ''' Prepare values in selected column from CSV file (f)
  19         '''
  20         with open(csv_file) as f:
  21             fn = []
  22             if col > 0:
  23                 fn += [0]*(col-len(fn)) #: Pad/ ignore columns (< col)
  24             fn.append('name') #: Mark selected column
  25             raw = list(csv.DictReader(f, fieldnames=fn))
  26             return set(map(lambda e: e['name'], raw[start:end]))
  27
  28     ''' Make properties use validation & write-once only
  29     '''
  30     @property
  31     def setA(self): return self._setA
  32
  33     @property
  34     def setB(self): return self._setB
  35
  36     @setA.setter
  37     def setA(self, v):
  38         if self._setA == None:
  39             self._setA = v
  40
  41     @setB.setter
  42     def setB(self, v):
  43         if self._setB == None:
  44             self._setB = v
  45
  46     @setA.getter
  47     def setA(self): return set(self._setA)
  48
  49     @setB.getter
  50     def setB(self): return set(self._setB)
  51
  52     ''' Comparse datasets (/ wrappers for set built-in)
  53     '''
  54     # Comparison sets
  55     @property
  56     def all(self): return self.setB.union(self.setB)
  57
  58     @property
  59     def clean(self): return self.setB.intersection(self.setA)
  60
  61     @property
  62     def added(self): return self.setB.difference(self.setA)
  63
  64     @property
  65     def erased(self): return self.setA.difference(self.setB)
  66
  67     # Calculations
  68     def p_erased(self):
  69         c, e = len(self.clean), len(self.erased)
  70         if c > e:
  71             return 100 * float(e) / float(c)
  72         else:
  73             return 100