Next in thread → Next in month →

RE: [ubl] The trailing-space bug

From
Vint, Dan <>
Date
2004-11-30T21:46:01+00:00
ID
Thread
RE: [ubl] The trailing-space bug
Title: Re: [ubl] The trailing-space bug

My organization has taken a 
slightly different stance on these problems. 

 

First we try to keep to the 
"core" features of schema (similar to what has been done here) and not try to 
work on the edge. But even some of what we consider core has been implemented 
differently. We feel that as long as I have a working implmenetation in the Unix 
and MS world that we are ok. No need to have complete support - it is then up to 
the users of these other tools to identify these issues to their vendors or 
developers of the tools they use. 

 

We have found Xerces-J and 
MS-XML release to be pretty consistent for the features we currently use. It 
also happens that Spy is also currently in line with our expectations, but in 
past versions it has not been. Note that we have found different issues with the 
MS .Net parser that do not occur in their COM? version.

 

..dan

From: Anthony B. Coates 
[mailto:]
Sent: Tue 11/30/2004 4:35 
PM
To: 
Subject: Re: [ubl] The 
trailing-space bug

Arguably none of the tools is 100% conformant (it's like ALGOL 
69 all over 
again).  However, XML Spy commands a large market 
share, and *those* folks 
think it is 100% conformant.  On that 
basis, just as it would be a bad 
idea to *only* validate with Spy, it 
would also be a bad idea to send out 
UBL Schemas/XML-files that *don't* 
validate with Spy (sub-issue is that 
different versions of Spy may 
behave differently, but let's do what we 
can).

My experience is 
that the only way to do a Schema that works for an 
acceptable share of 
the market is to validate against as many Schema 
validators as 
possible, and only use features/constructs that work with 
all of 
them.

Java-based editors (like Topologi) generally use Xerces-J for 
validation, 
and the Schema-WG seems to feel that this is currently the 
most correct of 
the 
validators.

        
Cheers,
        
        Tony.

On Mon, 29 Nov 2004 
15:35:29 -0800 (PST), <> wrote:

> According to 
the second paragraph from the last in this section,
> XML Spy is not a 
conformant XML processor.

--
Anthony B. Coates
London Market 
Systems Limited
33 Throgmorton Street, London, EC2N 2BR, UK
http://www.londonmarketsystems.com/
mailto:
Mobile/Cell: 
+44 (79) 0543 9026
[MDDL Editor (Market Data Definition Language), http://www.mddl.org/]
[FpML Arch WG Member 
(Financial Products Markup Language), 
http://www.fpml.org/]
-----------------------------------------------------------------------
This 
Email may contain confidential information and/or copyright 
material 
and is intended for the use of the addressee only.  Any 
unauthorised use 
may be unlawful. If you receive this Email by mistake 
please advise the 
sender immediately by using the reply  facility 
in your e-mail software.  
Email is not a secure method of 
communication and London Market Systems 
Limited cannot accept 
responsibility for the accuracy or completeness of 
this message or any 
attachment(s). Please examine this email for virus 
infection, for which 
London Market Systems Limited accepts no 
responsibility. If 
verification of this email is sought then please 
request a hard copy. 
Unless otherwise stated any views or opinions 
presented are solely 
those of the author and do not represent those of 
London Market Systems 
Limited.

To unsubscribe from this mailing list (and be removed from the 
roster of the OASIS TC), go to http://www.oasis-open.org/apps/org/workgroup/ubl/members/leave_workgroup.php.
Next in thread → Next in month →