Minutes ixml Call 2021-01-28

New Actions
ACTION: Steven to specify what happens when a name isn't an XML name
ACTION: Steven to research where to put S for attributes.
ACTION: Steven create W3C Community Group
ACTION: All create W3C account


ixml Meeting Thursday 28 January 2021


Present: M, S
Regrets: T


Agenda
Previous Actions
Website
Namespace
Specification
Implementations
Test Suite
Discussions:
   Does ixml need a mailing list for discussion?
   Checking the grammar - what is an ixml parser obligated to do?
   How would you use ixml to define a language with reserved words?
   Two possible additions to ixml? Epsilon and (?)
   Any-character symbol
AOB
Next call




Previous Actions


ACTION: Steven to specify what happens when a name isn't an XML name
[Continues]
ACTION: Steven to research where to put S for attributes.
[Continues]
ACTION: Steven to add John to org
[Done]
ACTION: Steven update the readme.md
[Done]
ACTION: Steven to create the Namespace document
[Done]


Website
S: Namespace document added at /NS


Namespace
S: Now in place. See above


Specification
S: Updated with correction. Updated all the IDs. 


Implementations
   No discussion
   
Test Suite
M: I have a proof of concept running.
    https://github.com/cmsmcq/ixml-tests
    https://github.com/cmsmcq/gingersnap
    Generates test cases and catalog files.
    There are some positive tests, and the negative test generator is 
running as we speak.
    
Discussions:
* Does ixml need a mailing list for discussion?
   S: W3C Community group?
   M: They do provide mailing list, archive and some visibility. We should 
look into it.
   S: (Looks up the conditions) We only need 5 people with W3C accounts to 
OK it.
   M: Let's do it!
   [ACTION: Steven create W3C Community Group]
   [ACTION: All create W3C account]


* What is an ixml parser obligated to do?
   M: I don't remember what I asked. I think it needs discussion, but isn't 
urgent.


   M: It should be possible to generate a RNG grammar for any ixml
   grammar except for character data. I can match the element and
   attribute structure, in some cases I can generate patterns for the
   contents of attributes, but not all, but I can't require in RNG
   either a colon or equal sign appear in the text at this point. There
   are no constraints on text nodes in mixed content.
   S: But for the xml form of an ixml grammar we don't care about those
   things anyway. All we care about is there is a sequence of <rule>,
   and each <rule> has an @name and a sequence of <alt>s.


* How would you use ixml to define a language with reserved words?


* Any-character symbol
   S: [] would be unsatisfiable, but then ~[] would always match one 
character regardless. Hmmm ["z"-"a"] is already the empty set according to 
the spec!
   M: [] would normally be an error. The current grammar is nice because it 
gives you syntax errors, but I like the empty set. An ixml parser can issue 
a warning if needs be.


* Two possible additions to ixml? Epsilon and (?)
   M: The other was 'unsatisfiable', the empty set solution solves my 
problem.


AOB
[None]


Next call
Thu 25 Feb 2021
15:00Z