On 09/19/2011 07:48 PM, Rita Greenberg wrote:
Hello.
Could somebody please explain the following code I found in a .fo file.
Here is my XML file.
<RCDATA>
<RCDATALINE>
<RC>ZE</RC>
<RCANSI>W1</RCANSI>
<TRC/>
<RCREDUCTION> 5878.14</RCREDUCTION>
<RCOVERRIDE/>
<RCMODIFIED/>
</RCDATALINE>
<RCDATALINE>
<RC>NR</RC>
<RCANSI>45</RCANSI>
<TRC/>
<RCREDUCTION> 0.00</RCREDUCTION>
<RCOVERRIDE/>
<RCMODIFIED/>
</RCDATALINE>
Here is the Following-sibling statement:
<xsl:if test="/doc/EOB/DETAIL/DETAILLINE/RCDATA/RCDATALINE/RC[.!='']/following-
sibling::*[.=''] and /doc/EOB/DETAIL/DETAILLINE/RCDATA/RCDATALINE/RC[.!
='']/following-sibling::*[.!='Y']">
My understanding of following-sibling::* is that the test is checking the
(first RC element and it it's not '' (empty) and the following siblings
(RCANSI, TRC, RCREDUCTION, RCOVERRIDE, RCMODIFIED) are = '' (empty))
AND
(first RC element and it it's not '' (empty) and the following siblings
(RCANSI, TRC, RCREDUCTION, RCOVERRIDE, RCMODIFIED) are not = 'Y'.
With the values above the condition is met and a block of code executed but I
don't understand how the first part of the test can be true since some of the
following-siblings are not = ''.
First, this is not a question for the FOP list, since this is strictly
an XSLT (or rather XPath) question.
An xsl:if test doesn't check all the nodes, but the existence of _some_
nodes that pass the test.
A parent/descendant selector doesn't pick the first node of type
<descendant>, but one that does fit in the whole expression.
The following-sibling axis, as explained on
http://www.w3.org/TR/xpath/#axes , refers to all the nodes that are
children of the same parent and follow the current node.
So, the whole test checks that:
- there are any RC nodes that have some text in them
- and they have at least one following sibling that is empty (this is
true for the TRC node)
AND, distinctly:
- there are any RC nodes that have some text in them (it doesn't have to
be the same node as in the above condition)
- and they have at least one following sibling that doesn't have Y as
their content (and this is again true for every sibling node in your
example)
Overall, I think that the whole test is wrong, since it never tries to
relate to the context node. What you probably want is to put this
condition in a template that is already on the RCDATALINE, and continue
testing from that point on with ./RC instead of going from the root with
/doc/..
Thanks,
Rita
--
Sergiu Dumitriu
http://purl.org/net/sergiu/
---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscr...@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-h...@xmlgraphics.apache.org
