Bug in eml2 for inline element type defination

[Peter McCartney]

Well this seems to be an important issue. We are in the middle of a remodel
of Xylopia to use EML documents as a container for passing data between
modules using the inline element. Given all the problems with validating EML
documents, I think its easier to just check an EML document to see if it has
what I need rather that coping with a million other reasons why the file
might not be valid depending on the parser you use. Therefore, we havent
detected this problem. 

There are other equally severe problems we've noted, however! The absence of
a union in the temporal coverage field prevents us from putting a year as
opposed to a calendar date. Id like to make sure that if we are developing a
target list for 2.x bugfixes, we have a system to do it logically. Do we
need to define a new version target in bugzilla so that we can start putting
these in there rather than helterskelter through email? Ive held off adding
anything for fear that it gets attached to 2.0.0 which seems like it should
be closed.



Peter McCartney (peter.mccartney at asu.edu)
Center for Environmental-Studies
Arizona State University
 


-----Original Message-----
From: Jing Tao [mailto:tao at nceas.ucsb.edu] 
Sent: Thursday, March 06, 2003 7:44 PM
To: jones at nceas.ucsb.edu; higgins at nceas.ucsb.edu; berkley at nceas.ucsb.edu;
eml-dev at ecoinformatics.org
Subject: Bug in eml2 for inline element type defination


Hi, everyone:

Today, when we tried to upload an eml2 doucment with inline charater data to
Metacat, a error was shown that inline element can only have element
children and  can't have a character child(inline element is defined in
eml-resource.xsd). However, inline element supposes to support characters
children. Something is wrong.

We took a look on the definition and found the children of inline is defined
as "any". So it seemes schema only consider the "any" just for elements come
from any namespace and we couldn't put characters in there. This is
different to dtd "any". In dtd, "any" can be considered as elements or
characters.

Dan found an text type defined in w3c primer web site: <xsd:complexType
name="text">  <xsd:complexContent mixed="true">
  <xsd:restriction base="xsd:anyType">
   <xsd:sequence>
    <xsd:any processContents="lax" minOccurs="0" maxOccurs="unbounded"/>
   </xsd:sequence>
   <xsd:attribute ref="xml:lang"/>
  </xsd:restriction>
 </xsd:complexContent>
</xsd:complexType>

It allows an unrestricted mixture of character content and element content
from any namespace.

I wrote an very simple schema which contains the text type and a xml
instance. Then using xerces to validate the xml instance by the schema. It
works fine if we delete the line: <xsd:attribute ref="xml:lang"/> in the
complex type definition.

So, can we use this type to replace the anonymous type for inline or other
place which use "any"?

By the way, Matt, Dan and Chad. I figured out why anoymous type for inline
element didn't work in this afternoon(I didn't put the line mentioned should
be delete in eml-resource.xsd). The reseaon is I didn't put attributes
"minOccurs="0"" in xsd:any element. If put it, works fine. Else gives us the
error. It is weild because we can omit minOccurs. Is this a bug for xerces?

Thanks.

Jing


Jing Tao
National Center for Ecological
Analysis and Synthesis (NCEAS)
735 State St. Suite 204
Santa Barbara, CA 93101


_______________________________________________
eml-dev mailing list
eml-dev at ecoinformatics.org
http://www.ecoinformatics.org/mailman/listinfo/eml-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mercury.nceas.ucsb.edu/ecoinformatics/pipermail/eml-dev/attachments/20030307/20243058/attachment.htm