BIND 10 #363: "H" and "I" are probably harmful for PyArg_ParseTuple()

[BIND 10 Development]

#363: "H" and "I" are probably harmful for PyArg_ParseTuple()
----------------------------+-----------------------------------------------
  Reporter:  jinmei         |            Owner:  jinmei               
      Type:  defect         |           Status:  new                  
  Priority:  minor          |        Milestone:  y2 12 month milestone
 Component:  DNSPacket API  |         Keywords:                       
 Sensitive:  0              |   Estimatedhours:  0                    
     Hours:  0              |         Billable:  1                    
Totalhours:  0              |         Internal:  0                    
----------------------------+-----------------------------------------------
 Some of our python binding code uses "H" (unsigned short) and "I"
 (unsigned int) with !PyArg_ParseTuple().

 I suspect most (if not all) of this usage is questionable because
 overflows are ignored.  For example, the following test current passes:
 {{{
         self.assertEqual(RRType("A"), RRType(0x100000001))
 }}}
 (assuming sizeof(int) is 4)

 Admittedly this is an artificial case but it still makes it awkward even
 if we don't call it a bug.

 I think we should use a signed version of larger size of integer ("h",
 "i", etc) and perform more strict range checks, i.e., reject value < 0 and
 value > possible_max.

-- 
Ticket URL: <http://bind10.isc.org/ticket/363>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development