pierred 00/11/30 15:33:02
Modified: jasper/src/share/org/apache/jasper/compiler
ParserController.java
Log:
A JSP page uses the JSP 1.2 XML syntax only if the
jsp:root element is present.
-----
As reported on tomcat-dev:
Danno Ferrin:
>
> I am sorry to start this thread here, but I believe that jasper-4.0's
> behavior is in error. The behavior I think the spec calls for in
> determining if a page is a JSP Document (xml jsp) or an XMl document
> with JSP markup is the presence or absence of a jsp:root element.
> [...]
Hans Bergsten:
I agree. JSP 1.1 says that everything that's not a JSP element is
treated as template text, and that any type of markup (text) language
can be used as template text. This means that a pure XML document,
with an <?xml version="1.0?> element at the top, is a valid JSP
page using the JSP 1.1 syntax.
The best (only?) way to identify a page that uses the JSP 1.2 XML syntax
is what you suggest: look for a jsp:root element. This should be
clarified in the JSP 1.2 spec.
Submitted by: Danno Ferrin <[EMAIL PROTECTED]>
Revision Changes Path
1.8 +10 -15
jakarta-tomcat-4.0/jasper/src/share/org/apache/jasper/compiler/ParserController.java
Index: ParserController.java
===================================================================
RCS file:
/home/cvs/jakarta-tomcat-4.0/jasper/src/share/org/apache/jasper/compiler/ParserController.java,v
retrieving revision 1.7
retrieving revision 1.8
diff -u -r1.7 -r1.8
--- ParserController.java 2000/10/28 03:20:22 1.7
+++ ParserController.java 2000/11/30 23:33:02 1.8
@@ -111,7 +111,6 @@
* Static information used in the process of figuring out
* the kind of document we're dealing with.
*/
- private static final String XML_PROLOG_TAG = "<?xml";
private static final String JSP_ROOT_TAG = "<jsp:root";
/*
@@ -241,20 +240,16 @@
jspReader.setSingleFile(true);
Mark startMark = jspReader.mark();
- // check the prolog (ideally it's been included if it is an xml doc)
- Mark mark = jspReader.skipUntil(XML_PROLOG_TAG);
- if (mark != null) {
- isXml = true;
- } else {
- // must have the jsp:root tag
- jspReader.reset(startMark);
- mark = jspReader.skipUntil(JSP_ROOT_TAG);
- if (mark != null) {
- isXml = true;
- } else {
- isXml = false;
- }
- }
+ // Check for the jsp:root tag
+ // No check for xml prolog, since nothing prevents a page
+ // to output XML and still use JSP syntax.
+ jspReader.reset(startMark);
+ Mark mark = jspReader.skipUntil(JSP_ROOT_TAG);
+ if (mark != null) {
+ isXml = true;
+ } else {
+ isXml = false;
+ }
// Figure out the encoding of the page
// FIXME: We assume xml parser will take care of
