Re: [netmod] WG Last Call: draft-ietf-netmod-yang-instance-file-format-06 to -07

[Kent Watsen <[e-mail]>]

One more thing, for non-mandatory nodes that don’t have a default value specified, please ensure the “description” statement states what it means for the node to be set or not set (whichever is easier)?   For instance:

OLD:

       leaf name {
         type string;
         description
           "Name of the YANG instance data set.";
       }

NEW (assuming this makes sense):

       leaf name {
         type string;
         description
           “An arbitrary name for the YANG instance data set.  This
            value is primarily used for descriptive purposes.  However,
            when the instance data set is saved to a file, then the
            filename MUST encode the name’s value, per Section 3 
            of RFC XXXX.";
       }

BTW, should the requirement of it needing to be encoded into the
filename place constraints on the “string” type?  Should a “pattern”
statement be added?

Kent



> On Mar 17, 2020, at 10:57 PM, Kent Watsen <[e-mail]> wrote:
> 
> 
> Hi Balazs,
> 
> 
>> Have the YANG modules been validated and tested for formatting?
>> 
>> (i.e., pyang -f yang --keep-comments --yang-line-length 69 filename) 
>> 
>> BALAZS: yes. With pyang offline and yangvalidator.com
> 
> Okay.
> 
> 
>> Have the examples in the draft validated against the YANG module?
>> 
>> BALAZS: Only manually. How do you validate samples conforming to a yang data structure ?
> 
> Hmmm, seeing that the examples are still not valid, here goes:
> 
> 	Until such time as tools support validating structure-data-ext, 
> 	one can rewrite the YANG module via s/sx:structure/container/ 
> 	and perform the validation against the resulting YANG module.
> 
> 
>> Please review the Normative/Informative status of the references.
>> 
>> Not looking carefully, but RFCs 2119 and 8174 should be Normative, 
>> 
>> and I think RFCs 3688 and 6020 should be Informative, right?
>> 
>> BALAZS: OK, changed in rev 08
> 
> Did you check all the other references too?  (I’m trying to save having to do another roundtrip when I do the shepherd writeup...)
> 
> 
>> All of the “import” statements in the YANG module are missing a
>> 
>> “reference” statement.
>> 
>> BALAZS:
>> 
>> Added:
>> 
>> rfc6991 for types added.
>> 
>> Already present:
>> 
>> rfc8342 for datastores
>> 
>> ietf-netmod-yang-data-ext for ietf-yang-structure-ext
> 
> Again, all the “import” statements in the YANG module are missing a “reference” statement.
> 
> 
>> Please add a paragraph to Section 5.2 preceding the YANG module
>> 
>> indicating all the aforementioned Normative references.
>> 
>> BALAZS: OK, I did it, but
>> 
>> this is not what is required by https://tools.ietf.org/html/rfc8407#section-3.9.
>> 
>> I also see no value in this statement.
> 
> I the "reference” statements mentioned above had been added, then s3.9 applies.
> 
> 
>> The copyright in the YANG module needs to be 2020 (not 2019)
>> 
>> BALAZS: OK
>> 
>> 
>> 
>> Please ensure a blank line between paragraphs in the “description"
>> 
>> statements.
>> 
>> BALAZS: OK
>> 
>> 
>> 
>> Please add a statement to the Introduction regarding why the module
>> 
>> Isn’t compliant with NMDA.
>> 
>> BALAZS: Sorry, don’t understand. Why is this not compliant with NMDA ?
>> 
>> IMHO it is NMDA compliant, or rather it  has nothing to do with NDMA.
> 
> Either way but, per https://tools.ietf.org/html/rfc8407#section-3.5, the 
> statement should be in the Introduction section.
> 
> 
>> The tree diagram does not adhere to the syntax described in 
>> 
>> draft-ietf-netmod-yang-data-ext.  
>> 
>> BALAZS: OK I try, but what actually is the problem? Any help would be really appreciated.
> 
> I was looking at the “+—rw”, which can’t be right because yang-data is not “configuration”...
> 
> 
>> Sadly
>> 
>> pyang -p ../ietfYams ietf-yang-instance-data\@2020-03-06.yang -f tree --tree-print-yang-data --tree-print-yang-data
>> 
>> doesn’t print out anything, so I am handcrafting.
> 
> `pyang`  supports the old/RFC8040 “rc:yang-data” statement; it hasn’t been updated to support the new "sx:structure” statement.
> 
> 
>> I just updated pyang from git. Any idea why this doesn’t work for me?
>> 
>> It would be good if YangValidator would print out the tree. At some point it did. Not now. :-(
> 
> First, use the s/sx:structure/container/ trick mentioned above.
> 
> Then s/+--rw/+—/.
> 
> Then review https://tools.ietf.org/html/draft-ietf-netmod-yang-data-ext-05#section-3 and tweak accordingly until all is good.
> 
> 
>> Please update the first sentence  in Section 5.1 to also reference draft-ietf-netmod-yang-data-ext.
>> 
>> BALZS: OK
>> 
>> 
>> 
>> Please ensure that the planning-text version of the draft passes
>> 
>> IDNITS (https://www6.ietf.org/tools/idnits) at the “verbose output”
>> 
>> level and correct any issues found, or explain why they shouldn’t
>> 
>> be corrected.
>> 
>> BALAZS: OK, corrected
> 
> Thanks!  (this is almost always the cause for needing another draft update when doing the shepherd writeup)
> 
> 
> 
> NEW: looking at the new "format-version”, please add a pattern statement to constrain the string values appropriately.  Hint, it’s half a "date-and-time” type...
> 
> 
> 
> Kent  // contributor (and shepherd)
> 
> 
> 
> _______________________________________________
> netmod mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/netmod