Auth48 by cabo · Pull Request #515 · ietf-wg-jsonpath/draft-ietf-jsonpath-base

cabo · 2024-02-12T14:29:44Z

Files for completing AUTH48 process

glyn

Not sure of the intended use of this. Is it mainly a base for applying XML changes? Is the content equivalent to the current level from the RFC editors? It looks like many of the files are generated from the XML, presumably just to give a more granular way of viewing changes.

Should we now create PRs relative to the auth48 branch? Not sure how to do that. Clearly, I can (at least in principle) edit the XML, but how to generate the rest?

cabo · 2024-02-12T16:30:24Z

On 2024-02-12, at 15:41, Glyn Normington ***@***.***> wrote: @glyn approved this pull request.

Well, we aren’t done yet :-)

Not sure of the intended use of this. Is it mainly a base for applying XML changes?

Yes. None are in yet.

Is the content equivalent to the current level from the RFC editors? It looks like many of the files are generated from the XML, presumably just to give a more granular way of viewing changes.

The other files are for check-abnf.

Should we now create PRs relative to the auth48 branch? Not sure how to do that. Clearly, I can (at least in principle) edit the XML, but how to generate the rest?

I was planning to simply edit the rfc9535.xml based on the diff lists we now have; the other files are just helping to do consistency checks (they are created from the check-abnf script, but I committed them so we don’t overlook changes). Grüße, Carsten

(After correcting it to 2.3.1.1, it would become a meaningless self-reference.)

And clarify whose notation this sentence is about.

(We use "evaluate" sparingly, and it is not needed at all here to further qualify "produce".)

glyn · 2024-02-13T17:08:53Z

 | `..*`               | shorthand for `..[*]`                                                                                                   |
 | `'name'`            | [name selector](#name-selector): selects a named child of an object                                                     |
-| `*`                 | [wildcard selector](#name-selector): selects all children of a node                                                    |
+| `*`                 | [wildcard selector](#wildcard-selector): selects all children of a node                                            |


@cabo Good catch!

summing with -> adding to

glyn

All changes up to G10 looking good, thanks!

note that we don't have a definition for a sole "argument", but that is a standard term that has very little confusion attached to it.

glyn

Another good set of changes, thanks, up to G1.

What became of G11?

glyn

All approved. Thanks!

glyn · 2024-02-13T18:21:48Z

Another good set of changes, thanks, up to G1.

What became of G11?

I dropped G11 after discussion.

gregsdennis · 2024-02-13T23:13:34Z

@cabo do you have a link for the auth48 process and what these files are for? I thought auth48 was just final author approvals.

cabo · 2024-02-14T04:07:41Z

Hi Greg,

there is not a single document I could point to.
GPT-4 says:

The Auth48 process for IETF standards is specified in several documents, but the main ones are:

RFC 2026: The Internet Standards Process -- Revision 3. This document covers the general standards process at IETF, including the final AUTH48 review state when a draft becomes a full Internet Standard.
The IETF's official "Guidelines to Authors of Internet-Drafts", also called "Guidelines to Authors."
The IETF's official "Guidelines to Working Group Chairs," which also mentions the AUTH48 process.
In addition, the IETF Trust's "Legal Provisions Relating to IETF Documents" include guidelines for authors during the AUTH48 review period.

In summary, AUTH48 is for checking the RFC Production Center (RPC) edits between authors and RPC and making any editorial and technical changes necessary before publication. Since this is an approved document, any technical changes will be quite limited (essentially inconsequential for interoperability in this case).

gregsdennis · 2024-02-14T04:38:53Z

So why all the new files? What's the driver for these?

I don't expect any technical changes, but I'm curious about the new files.

cabo · 2024-02-14T05:01:13Z

The new files come from a script, check-abnf, that just checks whether any of the spacing fixes we made to the ABNF actually changed its contents. They are extracted from the XML via kramdown-rfc-extract-sourcecode. I was lazy and comitted all of those, even if I only care about the ABNF.

gregsdennis

Thanks for the explanation, @cabo

glyn

5c10c5a LGTM, thanks.

glyn · 2024-02-14T10:28:44Z

      <t>JSON <xref target="RFC8259"/> is a popular representation
 format for structured data values.
-JSONPath defines a string syntax for selecting and extracting JSON <xref target="RFC8259"/> values
+JSONPath defines a string syntax for selecting and extracting JSON values


Yes, that was a bit much!

glyn

C4 and C5 LGTM, thanks!

(We can't verify what was deprecated in 2014, which was 5 years after ES5 was released, replacing the ES4 proposal stalled in development.)

glyn

C6, C1, C7 are good.

The word "strict" is used nowhere else but table 11. It serves no useful purpose here. (It might have been meant to distinguish strict inequalities (<) from non-strict (<=) ones, but extracting that intention is too much of a transfer for the reader.)

glyn

C8 is just what I would have done.

The text in 2.4.8 (value()) apparently wasn't updated when we stopped talking about "Nothing" for nodelists (these are now simply "empty").

glyn

C9 was a corker. Thanks for fixing.

cabo · 2024-02-15T05:42:59Z

I'm now planning to keep this as a draft pull request until the RFC is published.
Then I'll have a look if the published version can be back-ported to the .md

cabo added 2 commits February 12, 2024 15:07

Auth48: Note c&p error; date

0c5fbdb

Add 2024-02-07 RPC version of XML and extracted files; initial check

0ac20fe

cabo requested a review from glyn February 12, 2024 14:29

glyn approved these changes Feb 12, 2024

View reviewed changes

cabo added 11 commits February 12, 2024 17:48

First round of abnf spacing changes

b9e27e9

Put "; skip surrogate code points" on its own line

fe62a78

Fix some weird blank-space inconsistencies in ABNF

e49cfa1

Use and reference "I-Regexp" term from RFC 9485

756b304

Search and "at least one" substring

b6060cb

Add DIGIT to name-first at the end

3ea201e

(G27) Better Section Name "Function Extensions Subregistry"

ac6f4dc

(G26) Use § 2.4.3 in both places for "well-typed" cross-reference

1894d53

(G25) Remove incorrect cross-reference in 2.3.1.1

6c104f2

(After correcting it to 2.3.1.1, it would become a meaningless self-reference.)

(G21): Appendix B: Use JSONPath terminology for JSONPath

d4ffb90

And clarify whose notation this sentence is about.

(G18): Delete unnecessary text

cbfdb2a

(We use "evaluate" sparingly, and it is not needed at all here to further qualify "produce".)

glyn reviewed Feb 13, 2024

View reviewed changes

cabo added 2 commits February 13, 2024 18:16

(G10): Make sentence easier to understand

91099d9

summing with -> adding to

(G6): There is only one root identifier example

ec1d5d2

glyn reviewed Feb 13, 2024

View reviewed changes

cabo added 3 commits February 13, 2024 18:24

(G5): emphasize loneness by saying "containing a single node"

a2c0f6e

(G2): Remove remnant from when the definition was for "argument"

fdd18c2

note that we don't have a definition for a sole "argument", but that is a standard term that has very little confusion attached to it.

(G1): Remove unfortunate double use of the term "structure"

30b1579

glyn reviewed Feb 13, 2024

View reviewed changes

glyn approved these changes Feb 13, 2024

View reviewed changes

(C3): Avoid <referencegroup XML that is still at risk.

52d12f6

gregsdennis approved these changes Feb 14, 2024

View reviewed changes

RPC changes communicated at 2024-02-13 23:22Z

5c10c5a

glyn approved these changes Feb 14, 2024

View reviewed changes

cabo added 2 commits February 14, 2024 11:22

(C4) Complete fix for placement of "at most".

c8dfc85

(C5): Remove overzealous duplicate redundant superfluous citation

8076529

glyn reviewed Feb 14, 2024

View reviewed changes

glyn approved these changes Feb 14, 2024

View reviewed changes

cabo added 3 commits February 14, 2024 18:45

(C6) simplify editorial memento for ES4

440d44a

(We can't verify what was deprecated in 2014, which was 5 years after ES5 was released, replacing the ES4 proposal stalled in development.)

(C1) 3 more word-joiners to avoid confusing line breaks

33b6c01

(C7) use past tense for IANA registration that has happened

5b6ac37

glyn approved these changes Feb 14, 2024

View reviewed changes

(C9) We now distinguish empty nodelists from Nothing

9ad9249

The text in 2.4.8 (value()) apparently wasn't updated when we stopped talking about "Nothing" for nodelists (these are now simply "empty").

glyn approved these changes Feb 15, 2024

View reviewed changes

Uh oh!

Conversation

cabo commented Feb 12, 2024

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

cabo commented Feb 12, 2024 via email

Uh oh!

glyn Feb 13, 2024

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

glyn commented Feb 13, 2024

Uh oh!

gregsdennis commented Feb 13, 2024

Uh oh!

cabo commented Feb 14, 2024

Uh oh!

gregsdennis commented Feb 14, 2024

Uh oh!

cabo commented Feb 14, 2024

Uh oh!

gregsdennis left a comment

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

glyn Feb 14, 2024

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

glyn left a comment

Choose a reason for hiding this comment

Uh oh!

cabo commented Feb 15, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants