xml-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Eric Hodges <harmo...@swbell.net>
Subject Re: success stories with xml.apache.org
Date Tue, 29 Feb 2000 23:46:30 GMT
Why did we switch:  because OpenXML said they merged into Apache and OpenXQL
disappeared.  Also we want to support the XPath standard.

Our experience with Xalan and Xerces:
Negative:
    Some of the source code won't build on JDK 1.2.2.
    Some simple XPaths caused exceptions to be thrown.
    Resetting the parser causes the document it last generated to become
corrupt.
    Still having a hard time figuring out which method to call, in what
order, and what arguments they expect.
    Very little documentation on the API.
Positive:
    Some good responses from the mailing list and people working on these
products.
    Supports the XPath standard.

In general, I'm getting the gut feeling that these things work when plugged
into Apache, but the APIs haven't been exercised by many other applications.
A lot of the information about how to use these APIs must be encoded in
Apache.  It would be nice if some helper classes could provide common usage
patterns with a simplified API.

For instance, to execute an XPath, Xalan requires this code:

        XPathSupportDefault xpath_support = new XPathSupportDefault();
        XPath xpath = new XPath(xpath_support,null);
        XPathProcessorImpl parser = new XPathProcessorImpl(xpath_support);
        xpath.installFunction("document", new
org.apache.xalan.xslt.FuncDocument());
        parser.initXPath(xpath, queryString,null);
        XObject xobject = xpath.execute(xpath_support,document,null);

while XPathDOM (another OpenSource XPath implementation) requires this code:

        Evaluator ev = new Evaluator(doc);
        Object obj = ev.eval(queryString);

In the Xalan code, it's hard to find out what an XPathSupportDefault is, or
why XPath requires one to call installFunction() before using it.  Without
help from the mailing list, I'd have never figured some of this stuff out.

I don't want to make anyone angry by complaining, just want to give an
outsider's perspective.  I'm glad this code is out there and I hope it
becomes more robust and easier to use.  Hats off to all those folks that
contribute to it.



----- Original Message -----
From: Brian Behlendorf <brian@apache.org>
To: <general@xml.apache.org>
Sent: Tuesday, February 29, 2000 6:55 AM
Subject: success stories with xml.apache.org


>
> So in ~8 hours I'll be giving a short keynote at XTech in San Jose on
> Apache, and focus however much I can on XML and the xml.apache.org
> project.  I follow general@xml fairly closely but I've been unsubscribed
> from the main development lists; so if there are developments or
> noteworthy things you'd like to let others know about, let me know.  I'm
> also very interested in hearing about people who were formerly using
> someone else's XML parser or stylesheet engine or (anything else) and
> switched to something created at xml.apache.org, and why.  Or any other
> random data points you think might be noteworthy.  If anyone would like to
> contribute some, email me or the list.  Thanks!
>
> Brian
>
>
>


Mime
View raw message