Hi Med,
Thank you for your thorough review of the draft and the model. I just
published a version 03 which incorporates your changes and suggestions.
I'm addressing the questions you have about the model below. I have also
read the YANG authors guidance and made a few additional improvements
based on this.
An early yangdoctors review seems like a sensible approach.
Chairs, could you initiate such a review? Please specifically mention
the question whether using the base yang or yang-sx would be appropriate
for this model.
To answer the questions from the review:
> asn/asn4: Is this really needed given than both forms are covered by
> as-number
Depending on which type of Extended Community[0] is defined, the allowed
content of the Global Administrator field is different. For example, an
Extended Community of type 0 (Transitive Two-Octet AS-Specific) may not
contain a four-octet ASN. This is why the model specifies the two
options with different types. I have added constraints to the model to
make this more explicit.
Furthermore, future augmentations of the model may specify
IPv4-Address-Specific communities or other forms. For these,
"inet:as-number" cannot be used either.
> length: What is the expected use of this?
Within the model, a Local Administrator or Local Data Part may be
subdivided into multiple fields. See example 2 in the appendix for an
example of this. A parser needs to be aware of the length of each of
these sub-fields to know how many decimals or bits to match on. I have
added some text to the draft to clarify this.
I hope this addresses all of your points.
Kind regards,
Martin
[0]
https://www.iana.org/assignments/bgp-extended-communities/bgp-extended-communities.xhtml
_______________________________________________
GROW mailing list -- [email protected]
To unsubscribe send an email to [email protected]
