OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

 


 

   Re: [xml-dev] The subsetting has begun

[ Lists Home | Date Index | Thread Index ]

On Fri, 21 Feb 2003 10:13:00 -0500, Elliotte Rusty Harold 
<elharo@metalab.unc.edu> wrote:


>
> This is not a decision that should be made at the parser level though. 
> Parsers do need to process documents that contain document type 
> declarations. No one should ship a parser that simply gives up when it 
> encounters a document type declaration.

I'm not sure what "processing" a DTD means in a technical sense.  Accepting 
the entity declarations, setting the default attribute values, etc.  but 
not enforcing any content model constraints?  Or just parsing DTDs and 
ignoring everthing in there.  (IIRC either would be "legal" for a non- 
validating parser).  What would you suggest J2ME parsers do?

Perhaps the reasoning for throwing an exception is that processing DTDs but 
not fully performing all the actions that a validating parser do is an 
interoperability rathole; better to handle it in a "draconian" fashion 
rather than raise expectations that cannot be realistically met in the J2ME 
target environment.  (I don't have any knowledge of what is truly realistic 
or what the target constraints are, but I'll accept on faith that they J2ME 
folks wouldn't bother subsetting if they didn't see some real technical 
barrier to full conformance.)  Again IIRC from sml-dev days, there are a 
significant number of MAY clauses in XML 1.0 that define something like 128 
permutations of features that a non-validating XML 1.0 parser could support 
or not support.  Empirically this has created interop headaches (entity 
references being the most notorious).  I can, without claiming to know much 
about the people writing the J2ME spec or the requirements they're 
operating under, feel sympathy if their intent was to just throw an 
exception when a user steps into terra incognita rather than trying to 
provide a detailed map.

Channeling my inner newbie, I'd rather get slapped down on line 2 when a 
DTD reference or subset is rejected than on line 666 when I referenced an 
entity whose declaration was "processed" but not actually defined by the 
parser.  Conformance to XML 1.0 by J2ME is cold comfort to someone with a 
bug to fix and an incomprehensible message from the XML library to work 
with.  





 

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS