Replies: 3 comments 1 reply
-
None of the templates depend on the 'unnumbered' flag in the output (after my latest PR):
It only exists in the expected test output |
Beta Was this translation helpful? Give feedback.
-
I think that's exactly how it's done: https://github.com/ipspace/netlab/blob/dev/netsim/augment/links.py#L199 There's also dealing with various edge cases in https://github.com/ipspace/netlab/blob/dev/netsim/augment/links.py#L267
We could do that in interface cleanup, for example after https://github.com/ipspace/netlab/blob/dev/netsim/augment/links.py#L598
If you find it important, please go for it. |
Beta Was this translation helpful? Give feedback.
-
Implemented a long time ago |
Beta Was this translation helpful? Give feedback.
-
The way unnumbered links are created/treated has evolved, and has (imho) resulted in some complexity that could be removed.
The 'original'
unnumbered
links are based on loopback addresses: ipv4 and/or ipv6. There are also use cases for unnumbered links independent of loopback addresses.My proposal is to resolve the
unnumbered
flag into{ 'ipv4': True/False, 'ipv6': True/False }
early on in the code base, and get rid of the 'unnumbered' flag after that (also in the generated YAML)I can pick this one up if you agree
Beta Was this translation helpful? Give feedback.
All reactions