DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=6422>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=6422

With xml:space="preserve" on templates, parameters aren't first children

[EMAIL PROTECTED] changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|Normal                      |Minor
           Keywords|                            |ErrorMessage
           Priority|Other                       |Medium
            Summary|When using                  |With xml:space="preserve" on
                   |xml:space="preserve" on     |templates, parameters aren't
                   |templates , parameters dont |first children
                   |seem to pass in correct     |



------- Additional Comments From [EMAIL PROTECTED]  2002-02-13 18:59 -------
There may be a different bug for XalanC as well. In experimenting so far, it
looks like you need three things:
<xsl:template match="whatever" xml:space="preserve" > (the oddball part)
xsl:param elements inside that template
a reference to a param or variable in an expression ($param), though ANY such
  reference seems to do the trick, not just reference to a passed parameter
BIG CLUE: If you leave xml:space="preserve" out of the start tag and instead
put an LRE inside the template before the xsl:param, then a proper error is
raised. A comment in place of that LRE is ignored as it should be. A piece of
visible text (literal text for the result) raises the same "cannot convert
#STRING" error, so the issue is evidently detection of a text node between the
xsl:template start tag and the xsl:param.
The workaround, and better stylesheet coding practice IMHO, is to avoid use
of xml:space="preserve" in the stylesheet and instead have an explicit child
of xsl:template for each node (or sub-tree) to be generated in the output.

The fix will be to add detection and an error message parallel to what's
already being done for an LRE preceding xsl:param.

Reply via email to