Re: ACTION (2021-08-004): Steven to create list of existing requirements

I strongly feel that namespaces should be a requirement for v1: is there 
scope for discussion on the call?


I thought we had agreed a feature freeze for v1.


Should we also be talking about visible (XML) to invisible (non-XML) parses 
(or is that what is meant by 'other serialisations')?
I don't understand the question.


Steven




Tom

_________________
Tomos Hillman
eXpertML Ltd
+44 7793 242058
On 14 Dec 2021, 11:38 +0000, Steven Pemberton <steven.pemberton@cwi.nl>, 
wrote:

I think this covers them all.
Steven

Requirements for ixml v1

* Convert context-free non XML documents to XML.
* Unicode.
* Simple.
* Don't require the user to understand the fine points of parser theory
* Therefore deal with ambiguity.
* Flexible
* User-oriented, not machine-oriented: user needs weigh heavier than
machine needs.
* No need to do what other things in the XML pipeline can do better (eg, no
transformations)
* Don't have to be able to produce any particular XML (eg no Markdown to
XHTML).

Extra possibilities

* Use other serialisations than XML

Identified additions for v2

* Textual insertions
* Element renaming
date: iso; regular.
regular: s?, day, s, month, s, year.
s: -" "+.
day: digit, digit?.
-digit: ["0"-"9"].
month: "January"; "February"; "March"; "April"; "May"; "June"; "July";
"August"; "September"; "October"; "November"; "December".
year: digit, digit, digit, digit.
iso: year, "-", numeric^month, "-", day.
numeric^month: digit, digit. {numeric^month is serialised as <month>}
* Context sensitivity
* Namespaces