BIND 10 #116: review: NSEC RDATA "from wire" fixes and tests

BIND 10 Development do-not-reply at isc.org
Thu May 20 05:19:23 UTC 2010


#116: review: NSEC RDATA "from wire" fixes and tests
---------------------------+------------------------------------------------
 Reporter:  jinmei         |        Owner:  jinmei                     
     Type:  defect         |       Status:  assigned                   
 Priority:  critical       |    Milestone:  03. 1st Incremental Release
Component:  DNSPacket API  |   Resolution:                             
 Keywords:                 |    Sensitive:  0                          
---------------------------+------------------------------------------------
Changes (by shane):

  * owner:  each => jinmei


Comment:

 I decided I would go ahead and have a look at this. I'm not sure if this
 code has already been merged or not, but I figured it couldn't hurt to
 review it. I'm certainly not the best person for this, but I'll have a
 look.

 Typo here:

         116            isc_throw(DNSMessageFORMERR, "NSEC RDATA from wire:
 "
         117                           "incomplete bit map filed");

 Should be "field".

 A question about the new tests. One is this:

         81          // Bitmap length is 0, which is invalid.
         82          EXPECT_THROW(rdataFactoryFromFile(RRType::NSEC(),
 RRClass::IN(),
         83
 "testdata/rdata_nsec_fromWire6"),
         84                       DNSMessageFORMERR);

 However in toText() we see this:

         assert(len >= 0 && len <= 32);

 If the length cannot be 0, then the assertion should be:

         assert(len > 0 && len <= 32);

 Further, do we consider it good practice to use assert()? Will this raise
 an exception or simply end the program?

 Hopefully this is useful. :)

-- 
Ticket URL: <https://bind10.isc.org/ticket/116#comment:3>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development


More information about the bind10-tickets mailing list