BIND 10 #2534: support minor cases for quoted character strings

BIND 10 Development do-not-reply at isc.org
Tue Feb 4 23:19:13 UTC 2014


#2534: support minor cases for quoted character strings
-------------------------------------+-------------------------------------
            Reporter:  jinmei        |                        Owner:  shane
                Type:  task          |                       Status:
            Priority:  medium        |  reviewing
           Component:  libdns++      |                    Milestone:
            Keywords:                |  bind10-1.2-release-freeze
           Sensitive:  0             |                   Resolution:
         Sub-Project:  DNS           |                 CVSS Scoring:
Estimated Difficulty:  4             |              Defect Severity:  N/A
         Total Hours:  0             |  Feature Depending on Ticket:
                                     |          Add Hours to Ticket:  0
                                     |                    Internal?:  0
-------------------------------------+-------------------------------------
Changes (by muks):

 * owner:  muks => shane


Comment:

 Replying to [comment:15 kean]:
 > It is worth noting that RFC4408 says that adjacent strings must be
 concatenated "without adding spaces" (section 3.1.3).

 You are confusing presentation format with the actual data content. There
 are 3 representations here:

 1. Actual data
 {{{
 Test-StringTest-String
 }}}

 2. Wire data format
 {{{
 {0x0b, 'T', 'e', 's', 't', '-', 'S', 't', 'r', 'i', 'n', 'g',
  0x0b, 'T', 'e', 's', 't', '-', 'S', 't', 'r', 'i', 'n', 'g'}
 }}}

 3. Presentation format
 {{{
 "Test-String" "Test-String"
 }}}

 All 3 represent the same data. The example from the section in the SPF RFC
 I pointed to also demonstrates this equivalence. "v=spf1 .... first"
 "second string..." is the same as "v=spf1 .... firstsecond string...".

 > I am happy to accept things as they are if you either: a) point out in
 the RFC where it states that {{{"}}} is a delimiter or b) state that it is
 acceptable to do so simply because that is what BIND9 does and is probably
 what RFC1035 intended, even if it doesn't explicitly state it. If b) then
 please ignore my interpretation and feel free to merge and close.

 We generally aim to be compatible with BIND 9. Our master loader
 implementation is also based on BIND 9. There are a ton of zone files
 deployed in BIND 9 installations. While 1035 doesn't explicitly state that
 the double-quote (`"`) should be considered as start of a new token, in
 this case I'd side with going the BIND 9 way with this. That's why we have
 this ticket, and I agree with it.

 I'm assigning it to Shane for his thoughts so we can get to a resolution
 soon.

-- 
Ticket URL: <http://bind10.isc.org/ticket/2534#comment:16>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development


More information about the bind10-tickets mailing list