W3C home > Mailing lists > Public > xml-dist-app@w3.org > April 2000

Re: LOPT: serialization algorithm suggestions

From: Constantine Plotnikov <cap@mail.novosoft.ru>
Date: Fri, 14 Apr 2000 18:13:59 +0700
Message-ID: <38F6FD77.BA93FDA1@mail.novosoft.ru>
To: xml-dist-app@w3.org
CC: xml-dist-app@w3.org
Hi!

Bill la Forge wrote:
> 
> Sorry for the nit pick, but could you please put a '#' in front of
> the id in the href field? The original SOAP spec introduced this error
> and, while they have corrected the error, it seems that once something
> has been done so publicly, it assumes a life of its own!
> 
It is not very important for me. What is very important for me 
is single pass and simple (de)serialization.

"id" is expected to of type ID. And "href" is expected to be of 
type "IDREF". I have an impression that if href is declared as 
IDREF in schema, it should have the same value as id somewhere. 
(http://www.w3.org/TR/xmlschema-2/#IDREF) Because of it "#name" 
syntax was not used. 

AFAIU it is not requirement for LOPT to be completly compatible 
with SOAP. What I'm suggesting is much more radical revision
of protocol then href issue which will be usually single line 
patch.

BTW what is xlink is saying about it? Does it still use #id syntax?
AFAIR xhtml 1.1 will use xlink for cross referencing and will not 
use #name syntax.

There are bugs in example, it should be:

<List id="i0">
  <Type id="i1" >
    <value>
      <Element id="i2">
        <parent>
          <Type href="i1"/>
        </parent>
      <Element id="i2">
    </value>
  </Type>
  <Type href="i1"/>
</List>

(This is the list taht contains one object with the id "i1" twice)

Constantine

 
> ----- Original Message -----
> From: Constantine Plotnikov <cap@mail.novosoft.ru>
> To: MOREAU Jean-Jacques <moreau@crf.canon.fr>
> Cc: Dan Connolly <connolly@w3.org>; Henrik Frystyk Nielsen <frystyk@microsoft.com>; David Burdett <david.burdett@commerceone.com>;
> 'Eric Prud'hommeaux' <eric@w3.org>; Ken MacLeod <ken@bitsko.slc.ut.us>; <xml-dist-app@w3.org>
> Sent: Friday, April 14, 2000 5:17 AM
> Subject: Fwd: LOPT: serialization algorithm suggestions
> 
> > Hi!
> >
> > 1. Could you please reconsider serialization algorithm?
> > As for as I understand from your site the SOAP is starting
> > point for LOPT development.
> >
> > We had some problems with implementing soap protocol in java.
> > The algorithm require two passes for serialization and
> > deserialization.
> >
> > The basic idea I suggest is the same as in java and XMI 1.1
> > serialization algorithm.
> >
> > When object is serialized, it is assgined id and it is written
> > as:
> > <Type id="id0" >
> >   // contents
> > </Type>
> >
> > Later (in the body of the the element or or ), when reference is
> > encountered, empty element with href is used.
> >
> > <List t="i0">
> >   <Type id="i1" >
> >     <value>
> >       <Element id="i2">
> >         <parent>
> >           <Type href="i1"/>
> >         </parent>
> >       <Element id="i2">
> >     </value>
> >   </Type>
> >   <Type id="id0"/>
> > </List>
> >
> > It allow single pass serialization/desirailaization and references
> > to parent. I do not suggest to use exactly this representation
> > for protocol. For example XMI 1.1 like optimization for representation
> > of values may be used. I just want to make (de)serialization simple and
> > single pass.
> >
> > 2. Could you please include standard collection types in LOPT namespace.
> > I think there should be:
> > Map
> > List
> > Bag
> >
> > There may be other like Array.
> >
> > Constantine
> >
> >
Received on Friday, 14 April 2000 07:13:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:56 GMT