RE: a case in OASIS test suite

A potential erratum to the XML 1.0 spec was pointed out a few weeks ago on
the xml-dev list, regarding already published erratum E34 [1].

After examination, it appears that the report is incorrect and that E34 does
not need to be fixed.  A typo pointed out in the rationale was very real and
has been fixed, however.

An explanation appears below, within the excerpts from xml-dev.

[1] http://www.w3.org/XML/xml-19980210-errata#E34

--
François Yergeau


> >From: "Takuki Kamiya" <kamiya@rp.open.cs.fujitsu.co.jp>
> >To: "Richard Tobin" <richard@cogsci.ed.ac.uk>
> >Cc: <xml-dev@ic.ac.uk>
> >References: <026601bf68ab$bb24ca30$866e230a@sysrap.cs.fujitsu.co.jp>
> >Subject: Re: a case in OASIS test suite
> >Date: Fri, 28 Jan 2000 22:06:14 +0900
> >
> >Richard Tobin wrote:
> >>
> >> >After revisiting the description in XML 1.0 Sec. 4.1, It
> still looks
> non-WF
> >> >to me.
> >>
> >> See erratum E34
> (http://www.w3.org/XML/xml-19980210-errata#E34) which
> >> applies to this section.
> >>
> >> The idea is that it can't be a well-formedness constraint
> if not all
> >> processors would see the error, and those that read the external
> >> subset would think it was perfectly good.
> >>
> >
> >Thanks for pointing me to the erratum.
> >
> >However, my understanding on the conclusion of E34 appears
> contrary to yours.
> >
> >E34 seems to conclude that it is now legitimately made to be a well-
> >formedness constraint by saying that:
> >
> >"the Name given in the entity reference must match that in an entity
> > declaration that does not occur within the external subset or a
> > parameter entity."
> >
> >So I believe E34 makes it evident that OASIS inv-not-sa03 is not only
> >invalid but also not-well-formed.
> >
> >= Takuki Kamiya  Phone: (045)476-4586 Fax: (045)476-4749   =
> >= FUJITSU LIMITED (COINS:7128-4217 NIFTY:HHA01731)         =
> >
> >xml-dev: A list for W3C XML Developers. To post,
mailto:xml-dev@ic.ac.uk
>Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ or CD-ROM/ISBN
981-02-3594-1

==============================================================

>To: xml-dev@ic.ac.uk
>From: richard@cogsci.ed.ac.uk (Richard Tobin)
>Subject: Re: a case in OASIS test suite
>Date: 28 Jan 2000 14:21:16 GMT
>Organization: HCRC, University of Edinburgh
>
>In article <037a01bf6990$a1a95720$80f5a8c0@246.ne.jp>,
>Takuki Kamiya <kamiya@rp.open.cs.fujitsu.co.jp> wrote:
>>E34 seems to conclude that it is now legitimately made to be a well-
>>formedness constraint by saying that:
>>
>>"the Name given in the entity reference must match that in an entity
>> declaration that does not occur within the external subset or a
>> parameter entity."
>
>Yes, I misread it.
>
>This seems to mean that a non-validating parser that reads the
>external subset even when standalone="yes" must perform a check
>when expanding entities to see where they were declared.
>
>What is the status of a document like this:
>
><?xml version="1.0" standalone="yes"?>
><!DOCTYPE foo [
><!ENTITY % pe "<!ENTITY e 'hello'>">
>%pe;
>]>
><foo>&e;</foo>
>
>&e; refers to a declaration that occurs in a parameter entity, so it
>is not well-formed.  Or did they mean "external parameter entity"?

No, we meant all parameter entities. Because of 4.4.8 non-validating parsers
are not required to include parameter entities, external or not.
Consequently, we need to say "parameter entities" to ensure that all
processors are able to determine whether the WFC is met for standalone
documents, as the rationale says.

>There seems to be a typo in the Rationale for this erratum.  The
>second sentence contrasts a processor that doesn't read the external
>subset with one that (also) doesn't.
>
>-- Richard

Received on Thursday, 9 March 2000 16:08:10 UTC