Re: [netmod] Adam Roach's No Objection on draft-ietf-netmod-rfc6087bis-18: (with COMMENT)

[Benoit Claise <[e-mail]>]

On 3/8/2018 3:31 PM, Martin Bjorklund wrote:
> Benoit Claise <[e-mail]> wrote:
>> On 3/8/2018 3:15 PM, Juergen Schoenwaelder wrote:
>>> On Thu, Mar 08, 2018 at 02:21:01PM +0100, Benoit Claise wrote:
>>>> I see two solutions from here.
>>>> 1. we mention "pyang -f yang --yang-canonical --keep-comments FILE" in
>>>> RFC6087bis, with a warning such as: "As the tool matures, a human
>>>> might need
>>>> to polish the results"
>>>> 2. we don't mention "pyang -f yang --yang-canonical --keep-comments
>>>> FILE" in
>>>> RFC6087bis, but we ask the YANG doctors to run the tests.
>>> I am not sure it is a good idea to hard code command line options of
>>> specific tools in a BCP document. We should require that things are
>>> consistently indented and stay away from the advice of the day how
>>> to achieve that.
>> RFC 6087 mentions "pyang --ietf"  and
>> draft-ietf-netmod-yang-tree-diagrams mentions "pyang -f tree
>> --tree-line-length 50"
> Maybe we can say that "for example, a tool like pyang can be used ..."
This is a good compromise I would say.

Regards, Benoit
>
>> If not in RFC6087bis, where would we document this?
>> As a rhetorical question: How many of the YANG doctors were aware of
>> and are enforcing this command?
>>
>> Background Info.
>> Typical question I've been receiving lately (in this case from IANA):
>> When requesting the final files from the RFC-Editor, the file they
>> extracted using their tool and the yang modules appear to have a lot
>> of blank space in them. Is this OK?
> The extraction tools often produce additional blank lines b/c it is
> tricky to know how many blank lines were in the original module when
> all you have is the RFC in ascii.
>
>
> /martin
> .
>