XOM API for Processing XML with Java

[Elliotte Rusty Harold <elharo AT metalab.unc.edu>]

At 6:52 PM +0100 2/17/03, Laurent Bihanic wrote:
> Why is that unusual? Working with the URI alone and ignoring the 
> prefix used in a document instance is just what XPath does and what 
> XOM provides with getChildElement(name, uri).

What's unusual is not working with the URI alone. XOM fully supports 
that. What's unusual is wanting to find out the prefix.

> So XOM (and XPath) provides me ways to access elements and attribute 
> ignoring the namespace prefix used but to modify the document 
> contents I have to either redeclare the namespace using an arbitrary 
> (possibly conflicting) prefix or develop some low-level code to find 
> the namespace declaration.

As I said before, almost all the time if the namespace is used at all 
in the input document, there'll be an obvious element somewhere from 
which you can read the prefix; perhaps the root, perhaps something 
else. I don't think it's that hard to get now.

I've thought this one through, and it just doesn't cross my 80/20 
line. If you want to move me, you're going to have to come up with 
some pretty convincing and explicit use-cases.

--

+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo AT metalab.unc.edu | Writer/Programmer |
+-----------------------+------------------------+-------------------+
|           Processing XML with Java (Addison-Wesley, 2002)          |
|              http://www.cafeconleche.org/books/xmljava             |
| http://www.amazon.com/exec/obidos/ISBN%3D0201771861/cafeaulaitA  |
+----------------------------------+---------------------------------+
|  Read Cafe au Lait for Java News:  http://www.cafeaulait.org/      |
|  Read Cafe con Leche for XML News: http://www.cafeconleche.org/    |
+----------------------------------+---------------------------------+
> From mmceniry AT itsc.uah.edu  Mon Feb 17 15:18:22 2003
Return-Path: <mmceniry AT itsc.uah.edu>
Delivered-To: xom-interest AT lists.ibiblio.org
Received: from home.itsc.uah.edu (home.itsc.uah.edu [146.229.234.51])
        by happyhouse.metalab.unc.edu (Postfix) with ESMTP id 4E3F42001C
        for <XOM-interest AT lists.ibiblio.org>;
        Mon, 17 Feb 2003 15:18:22 -0500 (EST)
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
        charset="US-ASCII"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [XOM-interest] Getting the prefix of a namespace knowing the  URI
Date: Mon, 17 Feb 2003 14:19:33 -0600
Message-ID: <E05C39DBFF41854D9DC891EDC9DAE58A626B85 AT home.itsc.uah.edu>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: [XOM-interest] Getting the prefix of a namespace knowing the
        URI
Thread-Index: AcLWr7+TN++NHLpkRDCdTVSSVnWkhgACNZQA
From: "McEniry, Michael" <mmceniry AT itsc.uah.edu>
To: "XOM-interest" <XOM-interest AT lists.ibiblio.org>
X-BeenThere: xom-interest AT lists.ibiblio.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: <xom-interest.lists.ibiblio.org>
List-Unsubscribe: <http://lists.ibiblio.org/mailman/listinfo/xom-interest>,
        <mailto:xom-interest-request AT lists.ibiblio.org?subject=unsubscribe>
List-Archive: <https://lists.ibiblio.org/sympa/arc/xom-interest>
List-Post: <mailto:xom-interest AT lists.ibiblio.org>
List-Help: <mailto:sympa AT lists.ibiblio.org?subject=HELP>
List-Subscribe: <http://lists.ibiblio.org/mailman/listinfo/xom-interest>,
        <mailto:xom-interest-request AT lists.ibiblio.org?subject=subscribe>
X-List-Received-Date: Mon, 17 Feb 2003 20:18:22 -0000

Elliotte Rusty Harold wrote:
> At 6:52 PM +0100 2/17/03, Laurent Bihanic wrote:
> >Why is that unusual? Working with the URI alone and ignoring the=20
> >prefix used in a document instance is just what XPath does and what=20
> >XOM provides with getChildElement(name, uri).
> =20
> What's unusual is not working with the URI alone. XOM fully supports=20
> that. What's unusual is wanting to find out the prefix.

Perhaps someone has mentioned this before on the list, but I think=20
what counts as "usual" depends on whether one is reading XML or writing=20
it. When reading XML, once the parsing and validation is complete, one=20
is more concerned that a namespace is tied to a particular URI than
which=20
non-universal identifier is used as a prefix. When writing XML, even=20
though the URI is still the essential part of a namespace, one must=20
specify a prefix, because that is how XML works.

A namespace prefix is similar to a CDATA section in that it is strictly=20
a syntactic device and not a semantic one. CDATA is "syntactic sugar",=20
since one can use a different way to produce the same semantic result.=20
However, a namespace prefix is more like "syntactic fiber", since there=20
really is no other way to produce the same semantic result, given=20
current XML syntax.

Perhaps one approach is to treat namespace prefix like character
encoding.=20
Make it the responsibility of the serializer. One must specify a
namespace=20
map (URI -> prefix) as a parameter just as must specify "UTF-8". If a=20
namespace map does not include a prefix for a particular URI, then the=20
serializer could use arbitrary prefixes (eg, "ns1", "ns2", etc) or could

report an exception.

-- michael mceniry

PS: In spite of the excessive verbosity and utter unreadability, I want
the=20
following XML to be considered well-formed, just for a sense of
consistency,=20
since this is how it works semantically:

<"http://www.w3.org/1999/XSL/Transform":stylesheet version=3D"1.0">
 <"http://www.w3.org/1999/XSL/Transform":template match=3D"doc">
  <"http://www.w3.org/TR/xhtml1/strict":html>
    <"http://www.w3.org/TR/xhtml1/strict":head>
      <"http://www.w3.org/TR/xhtml1/strict":title>
        <"http://www.w3.org/1999/XSL/Transform":value-of
select=3D"title"/>
      </"http://www.w3.org/TR/xhtml1/strict":title>
    </"http://www.w3.org/TR/xhtml1/strict":head>
    <"http://www.w3.org/TR/xhtml1/strict":body>
      <"http://www.w3.org/1999/XSL/Transform":apply-templates/>
    </"http://www.w3.org/TR/xhtml1/strict":body>
  </"http://www.w3.org/TR/xhtml1/strict":html>
 </"http://www.w3.org/1999/XSL/Transform":template>
</"http://www.w3.org/1999/XSL/Transform":stylesheet>
> From elharo AT metalab.unc.edu  Mon Feb 17 15:31:17 2003
Return-Path: <elharo AT metalab.unc.edu>
Delivered-To: xom-interest AT lists.ibiblio.org
Received: from mail.speakeasy.net (mail14.speakeasy.net [216.254.0.214])
        by happyhouse.metalab.unc.edu (Postfix) with ESMTP id D0C5D2001C
        for <XOM-interest AT lists.ibiblio.org>;
        Mon, 17 Feb 2003 15:31:16 -0500 (EST)
Received: (qmail 20444 invoked from network); 17 Feb 2003 20:32:32 -0000
Received: from unknown (HELO [192.168.254.4]) ([216.254.85.72])
         (envelope-sender <elharo AT metalab.unc.edu>)
         by mail14.speakeasy.net (qmail-ldap-1.03) with SMTP
         for <mmceniry AT itsc.uah.edu>; 17 Feb 2003 20:32:32 -0000
Mime-Version: 1.0
X-Sender: elharo AT mail.ibiblio.org
Message-Id: <p04330102ba76f523eba0@[192.168.254.4]>
In-Reply-To: <E05C39DBFF41854D9DC891EDC9DAE58A626B85 AT home.itsc.uah.edu>
References: <E05C39DBFF41854D9DC891EDC9DAE58A626B85 AT home.itsc.uah.edu>
Date: Mon, 17 Feb 2003 15:29:14 -0500
To: "McEniry, Michael" <mmceniry AT itsc.uah.edu>,
        "XOM-interest" <XOM-interest AT lists.ibiblio.org>
From: Elliotte Rusty Harold <elharo AT metalab.unc.edu>
Subject: RE: [XOM-interest] Getting the prefix of a namespace knowing the 
URI
Content-Type: text/plain; charset="us-ascii" ; format="flowed"
X-BeenThere: xom-interest AT lists.ibiblio.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: <xom-interest.lists.ibiblio.org>
List-Unsubscribe: <http://lists.ibiblio.org/mailman/listinfo/xom-interest>,
        <mailto:xom-interest-request AT lists.ibiblio.org?subject=unsubscribe>
List-Archive: <https://lists.ibiblio.org/sympa/arc/xom-interest>
List-Post: <mailto:xom-interest AT lists.ibiblio.org>
List-Help: <mailto:sympa AT lists.ibiblio.org?subject=HELP>
List-Subscribe: <http://lists.ibiblio.org/mailman/listinfo/xom-interest>,
        <mailto:xom-interest-request AT lists.ibiblio.org?subject=subscribe>
X-List-Received-Date: Mon, 17 Feb 2003 20:31:17 -0000

At 2:19 PM -0600 2/17/03, McEniry, Michael wrote:
> A namespace prefix is similar to a CDATA section in that it is strictly
> a syntactic device and not a semantic one. CDATA is "syntactic sugar",
> since one can use a different way to produce the same semantic result.
> However, a namespace prefix is more like "syntactic fiber", since there
> really is no other way to produce the same semantic result, given
> current XML syntax.

Which is why XOM lets you specify the prefix. In fact, it requires 
you to do so. It does not assign them randomly.

XOM also lets you find out which prefix any given element or attribute has.

The question is only whether there should be a convenience method 
that walks up the tree to find the nearest prefix mapped to a given 
URI. (Here's another monkey wrench for that idea: there may be more 
than one such prefix in scope, even on a single element so the method 
should return a String array, not a single String.) You can do this 
now. XOM gives you everything you need. There's non information 
missing. We're just arguing about whether I should put this into the 
core or not.

> Perhaps one approach is to treat namespace prefix like character
> encoding.
> Make it the responsibility of the serializer. One must specify a
> namespace
> map (URI -> prefix) as a parameter just as must specify "UTF-8". If a
> namespace map does not include a prefix for a particular URI, then the
> serializer could use arbitrary prefixes (eg, "ns1", "ns2", etc) or could
>
> report an exception.

No. The prefixes are part of the element names and are reported and 
stored as such. This is not just an artifact of serialization so it 
can't really be changed by the Serializer.

Yes, this whole shebang with prefixes vs. URIs is ugly, but it's ugly 
because XML is ugly here. XOM can't and won't pretend XML is simpler 
than it is.


> PS: In spite of the excessive verbosity and utter unreadability, I want
> the
> following XML to be considered well-formed, just for a sense of
> consistency,
> since this is how it works semantically:

But it's not well-formed. Sorry about that, but I didn't make up 
these rules. I just follow them. :-(

>  <"http://www.w3.org/1999/XSL/Transform":stylesheet version="1.0">
>   <"http://www.w3.org/1999/XSL/Transform":template match="doc">
>    <"http://www.w3.org/TR/xhtml1/strict":html>
>      <"http://www.w3.org/TR/xhtml1/strict":head>
>        <"http://www.w3.org/TR/xhtml1/strict":title>
>          <"http://www.w3.org/1999/XSL/Transform":value-of
> select="title"/>
>        </"http://www.w3.org/TR/xhtml1/strict":title>
>      </"http://www.w3.org/TR/xhtml1/strict":head>
>      <"http://www.w3.org/TR/xhtml1/strict":body>
>        <"http://www.w3.org/1999/XSL/Transform":apply-templates/>
>      </"http://www.w3.org/TR/xhtml1/strict":body>
>    </"http://www.w3.org/TR/xhtml1/strict":html>
>   </"http://www.w3.org/1999/XSL/Transform":template>
>  </"http://www.w3.org/1999/XSL/Transform":stylesheet>

--

+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo AT metalab.unc.edu | Writer/Programmer |
+-----------------------+------------------------+-------------------+
|           Processing XML with Java (Addison-Wesley, 2002)          |
|              http://www.cafeconleche.org/books/xmljava             |
| http://www.amazon.com/exec/obidos/ISBN%3D0201771861/cafeaulaitA  |
+----------------------------------+---------------------------------+
|  Read Cafe au Lait for Java News:  http://www.cafeaulait.org/      |
|  Read Cafe con Leche for XML News: http://www.cafeconleche.org/    |
+----------------------------------+---------------------------------+
> From rogers AT cadenhead.org  Tue Feb 18 17:23:31 2003
Return-Path: <rogers AT cadenhead.org>
Delivered-To: xom-interest AT lists.ibiblio.org
Received: from imf22bis.bellsouth.net (mail022.mail.bellsouth.net
        [205.152.58.62])
        by happyhouse.metalab.unc.edu (Postfix) with ESMTP id C02652001F
        for <xom-interest AT lists.ibiblio.org>;
        Tue, 18 Feb 2003 17:23:30 -0500 (EST)
Received: from ZEUS ([66.20.159.8]) by imf22bis.bellsouth.net
         (InterMail vM.5.01.04.25 201-253-122-122-125-20020815) with SMTP
         id <20030218222641.MNAW8977.imf22bis.bellsouth.net@ZEUS>
         for <xom-interest AT lists.ibiblio.org>;
         Tue, 18 Feb 2003 17:26:41 -0500
From: Rogers Cadenhead <rogers AT cadenhead.org>
To: XOM-Interest <xom-interest AT lists.ibiblio.org>
X-Mailer: PocoMail 2.63 (1077) - Licensed Version
Date: Tue, 18 Feb 2003 17:24:41 -0500
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Message-Id: <20030218222641.MNAW8977.imf22bis.bellsouth.net@ZEUS>
Subject: [XOM-interest] Linux Magazine article on XOM
X-BeenThere: xom-interest AT lists.ibiblio.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: <xom-interest.lists.ibiblio.org>
List-Unsubscribe: <http://lists.ibiblio.org/mailman/listinfo/xom-interest>,
        <mailto:xom-interest-request AT lists.ibiblio.org?subject=unsubscribe>
List-Archive: <https://lists.ibiblio.org/sympa/arc/xom-interest>
List-Post: <mailto:xom-interest AT lists.ibiblio.org>
List-Help: <mailto:sympa AT lists.ibiblio.org?subject=HELP>
List-Subscribe: <http://lists.ibiblio.org/mailman/listinfo/xom-interest>,
        <mailto:xom-interest-request AT lists.ibiblio.org?subject=subscribe>
X-List-Received-Date: Tue, 18 Feb 2003 22:23:31 -0000

I have an article in the March 2003 issue of Linux Magazine that=

covers XOM. I just got it in the mail this afternoon, so it will=

probably be in stores within a week or two. (It won't be online=
for 
several months.)

I have a few questions from working with version 1.0d8 to write=
the 
piece, but I'll hold off until I can see if they still apply to 
1.0d10.

Overall, I found XOM to be an amazingly well-organized, intuitive=
API 
that's easy to learn and to use. I like how it reinforces good 
practices and provides insight about XML -- such as the lack of 
whitespace when XML is produced without a serializer and the 
identical treatment of text whether it consists of character 
entities, CDATA sections, or regular characters.

I can't compare it to JDOM, but it's appreciably more pleasant to=

work with than the Simple API for XML Processing.
--
Rogers Cadenhead, rogers AT cadenhead.org on 02/18/2003
Weblog: http://www.cadenhead.org/workbench