BIND 10 #2009: const uint8_t {aka const unsigned char}' inside { } is ill-formed in C++11

[BIND 10 Development]

#2009: const uint8_t {aka const unsigned char}' inside { } is ill-formed in C++11
-------------------------------------+-------------------------------------
                   Reporter:  jreed  |                 Owner:  muks
                       Type:         |                Status:  reviewing
  defect                             |             Milestone:
                   Priority:         |  Sprint-20120612
  medium                             |            Resolution:
                  Component:         |             Sensitive:  0
  Unclassified                       |           Sub-Project:  DNS
                   Keywords:         |  Estimated Difficulty:  4
            Defect Severity:  N/A    |           Total Hours:  0
Feature Depending on Ticket:         |
        Add Hours to Ticket:  0      |
                  Internal?:  0      |
-------------------------------------+-------------------------------------

Comment (by muks):

 Replying to [comment:7 jinmei]:
 > The change (basically) looks okay, but I'm afraid it's not so obvious
 why we
 > need to do that, so please add some comments about why.  I don't
 > think I need to check the comment - please add whatever you think
 > makes sense, and just merge it.

 I've added this comment now:
 {{{
 +    // Splitting msg_len below performs something called a 'narrowing
 +    // conversion' (conversion of uint16_t to uint8_t). C++0x (and GCC
 +    // 4.7) requires explicit casting when a narrowing conversion is
 +    // performed. For reference, see 8.5.4/6 of n3225.
 }}}

 > One note: why was 20a21a87777944fea10cf615b0b54d6e96df00ae included in
 > the branch?  It doesn't seem to be related to the topic of this ticket.

 It isn't related. This is why it's prefixed with [master]. It was 'found'
 during work on this ticket, so I just checked it into this branch instead
 of directly in master.

 > Another note: we should kill this stuff anyway and migrate to the
 > socket session module.  And, for that reason, I don't necessarily
 > think we need a changelog entry for it.  But if you think it's worth
 > mentioning, please do so.

 I'll skip the ChangeLog entry for this.

 > Final note: I've found this:
 > http://stackoverflow.com/questions/4434140/narrowing-conversions-in-c0x-
 is-it-just-me-or-does-this-sound-like-a-breakin

 This was helpful as a reference to write the comment above.

-- 
Ticket URL: <http://bind10.isc.org/ticket/2009#comment:9>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development