JDK-8011653: Upgrade to JAXP 1.5 (original) (raw)

huizhe wang huizhe.wang at oracle.com
Fri May 3 08:01:23 UTC 2013

Hi Alan, Lance,

This is an update that reflects the spec change, and also fixes for JDK-8013232 and JDK-8013484.

Please review: http://cr.openjdk.java.net/~joehw/jdk8/8011653/webrev/

Thanks, Joe

On 4/18/2013 2:43 AM, huizhe wang wrote:

On 4/15/2013 2:22 AM, Alan Bateman wrote: On 15/04/2013 08:48, Joe Wang wrote:

:

For the new properties then it specifies that a "a runtime exception" will be thrown. Can this be more specific? They can't be in XMLConstants, but they are in the specific Factories. The properties may be supported by factories that may throw different exceptions. I think it would be help if this were expanded to something like "a runtime exception that is specific to the context is thrown" and give an example so that it's clear what it saying. Absolutely! While doing so, I realized that I should have been even more specific in what throws which exception. I've added more details to the javadoc in Factories and SAXParser. Webrevs updated: http://cr.openjdk.java.net/~joehw/jdk8/8011653/webrev/ This looks much better. For now, I've stayed focused on the javadoc/spec for now as we have to get that right. The wording ""jar" plus the scheme portion" suggests it matches "jar" exactly and maybe this could be clearer because this is also case insensitive. Added 'including the keyword "jar"' in Protocols are case-insensitive. @since on the new properties 1.7. I don't know if this should have 1.8 or JAXP 1.5. I think we'll have approval to integrate JAXP 1.5 into JDK7. So it's 1.7. In JAXP javadocs, JDK versions have been used for @since. The intending of the

    and
  • looks a bit odd when the
paragraphs aren't indented. This doesn't impact the generated javadoc of course, just looks odd in the source code. It was indeed intended since the section within
    and
  • applies
 to the new property only. I've added tabs to make it easier to read. Otherwise I think the javadoc looks okay to me. Thanks, Joe -Alan

More information about the core-libs-dev mailing list