[Devel] [ openser-Patches-1344272 ] Carrier ENUM support according to draft-haberler-carrier-enu

[Otmar Lendl]

On 2006/10/09 19:10, Juha Heinanen <jh at tutpro.com> wrote:
> Klaus Darilion writes:
> 
>  > Since August the interims infrastructure ENUM methods based on branching 
>  > inside the e164.arpa domain is working group item.
>  > 
>  > http://www.ietf.org/internet-drafts/draft-ietf-enum-combined-00.txt
> 
> klaus,
> 
> we have previously had policy that new features will be added when the
> drafts become working group documents or are otherwise widely
> implemented.  since this draft is working group draft, i could include
> the patch.

I'm glad to hear that.

>  > This patch adds the following items:
>  > 1. extend openser's resolver to support TXT and EBL records.
>  > 
>  > 2. extend the ENUM module to allow the use of the TXT-record based 
>  > interm solution as described in draft-ietf-enum-combined
>  > 
>  > 3. extend the ENUM module to allow usage of the EBL-record based interm 
>  > solution as described in draft-ietf-enum-branch-location-record
>  > 
>  > 4. extend the ENUM module to allow branching at the country-code level
>  > 
>  > What about adding the patch to the enum module?
> 
> what are its performance implications to a regular user enum query?  is
> there a new parameter or function name to tell that this is an infra
> enum query?

The changes to the default code-path are as follows:

Instead of enum_query_0 and enum_query_1 calling enum_query_2, now
all of those call enum_query_3 with the default parameters.

(This means one function call overhead more for the enum_query_2 case.)

There is one if() on the new parameter which triggers the 
extra code if necessary.

And then there is one extra if() in the number-reversing loop.

Summary: at worst case, user-enum gets an extra function-call overhead,
16 simple if() tests, and a few bytes more on the stack in the function.

/ol
-- 
< Otmar Lendl (lendl at nic.at) | nic.at Systems Engineer >