Fix pandas.io.parsers.parseExcel to be more robust / have correct handling of Excel dates about pandas

[wesm]

A user suggested this version to start with:

from datetime import datetime,time

def parseExcel(filepath, sheetname, header=False, indexCol=0):
   # ImportError is quite clear already ;-)
   import xlrd
   book = xlrd.open_workbook(filepath)
   sheet = book.sheet_by_name(sheetname)
   data = []
   for rowx in range(sheet.nrows):
       row = []
       for value,type in zip(
           sheet.row_values(i),
           sheet.row_types(i)
           ):
           if type==xlrd.XL_CELL_DATE:
               dt = xldate_as_tuple(value, book.datemode)
               if dt[0] < datetime.MINYEAR:
                   value = time(*dt[3:])
               else:
                   value = datetime(*dt)
           row.append(value)
       data.append(row)
   return simpleParser(data, header = header, indexCol = indexCol)

[wesm]

addressed in

wesm@6e38eca

thanks to Chris Withers for the help with this