metanorma adds an <aside> element to rfc xml for NOTE:

[douglm]

Trying to build (locally on my mac) the spec here:

https://github.com/CalConnect/PUBLIC_DRAFTS/tree/master/ical-tasks

Failed with the messages:

sources/draft-ietf-calext-ical-tasks.rfc.xml(344): Error: Element dd has extra content: aside, at /rfc/middle/section[10]/section[4]/dl/dd[1]/aside
sources/draft-ietf-calext-ical-tasks.rfc.xml(9): Error: Invalid document before running preptool.
Unable to complete processing sources/draft-ietf-calext-ical-tasks.rfc.xml
sources/draft-ietf-calext-ical-tasks.rfc.xml(344): Error: Element dd has extra content: aside, at /rfc/middle/section[10]/section[4]/dl/dd[1]/aside
sources/draft-ietf-calext-ical-tasks.rfc.xml(9): Error: Invalid document before running preptool.
Unable to complete processing sources/draft-ietf-calext-ical-tasks.rfc.xml

There is an <aside> element at line 344 - it seems every NOTE: generates an <aside> element

Tried a brew upgrade metanorma

and it told me

 metanorma/metanorma/metanorma 1.4.17.20211007_1 already installed

Edited the xml and removed all the <aside> elements. xml2rfc accepted that.

[ronaldtse]

Thank you @douglm for raising this!

@opoudjis could you help review the aside situation?

@CAMOBAP could you please help check the issue with Homebrew?

[douglm]

I've made a few updates to the draft and reran and the problem seems to have disappeared. I did remove one of the NOTEs

On 2/19/23 23:35, Ronald Tse wrote:

Thank you @douglm https://github.com/douglm for raising this!

@opoudjis https://github.com/opoudjis could you help review the |aside| situation?

@CAMOBAP https://github.com/CAMOBAP could you please help check the issue with Homebrew?

— Reply to this email directly, view it on GitHub https://github.com/metanorma/metanorma-ietf/issues/182#issuecomment-1436318785, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAW2MBWVKUEOJ6WPYI3V5ATWYLYCPANCNFSM6AAAAAAVASDDHI. You are receiving this because you were mentioned.Message ID: @.***>

[CAMOBAP]

@ronaldtse @douglm metanorma version looks really old, I investigating right now why brew upgrade doesn't work

As workaround I can suggest try brew uninstall metanorma && brew install metanorma/metanorma/metanorma

Update

https://github.com/metanorma/homebrew-metanorma/issues/84#issuecomment-1436968179

It looks like auto-update is broken for our formula, it hard to say why this may happens. there are possible workarounds for this:

• brew tap --force-auto-update metanorma/metanorma && brew upgrade metanorma or
• brew update && brew upgrade metanorma or
• brew uninstall metanorma && brew install metanorma/metanorma/metanorma

[opoudjis]

This is a XML RFC v3 schema restriction that I cannot say I welcome.

• Metanorma allows notes to be embedded in definition lists. In fact, Metanorma prefers to embed notes inside of other blocks (e.g. tables)
• I see in that document a NOTE after a table, and a NOTE after a list. These will both be embedded into their preceding block by default in Metanorma.
• I have indeed used aside in XML RFC v3 for notes: aside is "is a container for content that is semantically less important or tangential to the content that surrounds it." https://datatracker.ietf.org/doc/html/rfc7991#section-2.6 On the other hand, XML RFC v3 note is restricted to a single note after the abstract, giving information to reviewers: https://datatracker.ietf.org/doc/html/rfc7991#section-2.33
• XML RFC v3 as of 3.5.13 only permits aside inside of dd and section (which means that what was reported as an error in your version of xml2rfc, /rfc/middle/section[10]/section[4]/dl/dd[1]/aside, is now correct. (It was not in the original XML2RFC in RFC 7991.)
• However, this clearly means that aside is severely restricted in where it can occur in XML2RFC.
• To address this, I am going to:
  • transform aside back to a t paragraph with "NOTE:" unless it is at the top level of a clause: I will not allow aside inside of different type of block, such as a table
  • not fold notes into preceding blocks, as is the default for Metanorma, even for definition lists; avoiding that kind of scenario is preferable.

[opoudjis]

In fact, I already have code moving asides to after blocks they are embedded with; it was only applied to t figure table. I'm going to generalise that to everything but section.