Closed
Bug 143668
Opened 23 years ago
Closed 23 years ago
iterating over nodesets implicitly converted from RTFs fails is some cases
Categories
(Core :: XSLT, defect)
Tracking
()
RESOLVED
INVALID
People
(Reporter: julian.reschke, Assigned: peterv)
References
()
Details
Attachments
(1 file, 2 obsolete files)
|
1.55 KB,
text/xml
|
Details |
When for-each-iterating though a node-set that was
implicitly converted from a RTF, matching by qualified name doesn't seem to
work.
In my attempt to minimize the problem, I produced:
<?xml-stylesheet type="text/xsl" href="moztest.xslt" ?>
<xsl:transform xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
version="1.0"
xmlns:saxon="http://icl.com/saxon"
xmlns:test="mailto:julian.reschke@gmx.de"
xmlns:msxsl="urn:schemas-microsoft-com:xslt"
>
<xsl:output method="html" />
<xsl:template match="/">
<html>
<head>
<title>Mozilla test</title>
</head>
<body>
<xsl:variable name="test">
<xsl:call-template name="fill-var" />
</xsl:variable>
<p>
<xsl:choose>
<xsl:when test="function-available('msxsl:node-set')">
<xsl:variable name="ns" select="msxsl:node-set($test)"/>
<xsl:value-of select="count($ns/*)" />
/
<xsl:value-of select="count($ns/test:x)" />
</xsl:when>
<xsl:when test="function-available('saxon:node-set')">
<xsl:variable name="ns" select="saxon:node-set($test)"/>
<xsl:value-of select="count($ns/*)" />
/
<xsl:value-of select="count($ns/test:x)" />
</xsl:when>
<xsl:otherwise>
<xsl:variable name="ns" select="$test"/>
<xsl:value-of select="count($ns/*)" />
/
<xsl:value-of select="count($ns/test:x)" />
</xsl:otherwise>
</xsl:choose>
</p>
</body>
</html>
</xsl:template>
<xsl:template name="fill-var">
<test:x>1</test:x>
<test:x>2</test:x>
</xsl:template>
</xsl:transform>
This displays "2 / 0" rather than "2 / 2.
|
Reporter | |
Comment 1•23 years ago
|
||
|
||
Comment 2•23 years ago
|
||
test code produces "Mozilla test / / / 1 2" for me (with 1.2beta on win2k), but
I don't understand how this is supposed to work...
reporter (Julian): can you try this with a recent build of mozilla and see if it
is still a problem? if so, please comment again. if not, please resolve this bug
as WORKSFORME. thanks.
|
||
Comment 3•23 years ago
|
||
the previous testcase was referencing the stylesheet by name, I changed this
to "#".
Attachment #83199 -
Attachment is obsolete: true
|
|
||
Comment 4•23 years ago
|
||
confirmed with 1.0.1, fixed on something after the API landing.
I suspect this was fixed by the context landing, guess this was namespaces.
Status: UNCONFIRMED → RESOLVED
Closed: 23 years ago
Resolution: --- → FIXED
|
|
Reporter | |
Comment 5•23 years ago
|
||
1) The problem as initially isolated is gone.
2) However, a similar problem that likely is related still exists (sample code
attached, please compare Saxon/MSXML's output with Mozilla's: the
sections "status of this memo" and "copyright" are missing)
Status: RESOLVED → UNCONFIRMED
Resolution: FIXED → ---
|
|
Reporter | |
Comment 6•23 years ago
|
||
|
|
||
Comment 7•23 years ago
|
||
Comment on attachment 104486 [details]
test input and XSLT, compare results using Saxon/MSXML/Mozilla
that's not a testcase.
70k stylesheet and 100k source is not going to work.
Attachment #104486 -
Attachment is obsolete: true
I *think* that you might need to do something like:
$var/node()/*
to get all elements in the RTF (i.e. insert an extra "/node()"). The reason is
that our RTFs are nodesets containing a single DocumentFragment node in which
all actual children lives.
Note however that this is hackish and will not work in the future
|
|
Reporter | |
Comment 9•23 years ago
|
||
This indeed works (as a workaround).
<p>
The best solution would be to become compliant either to XSLT 1.0 (requiring a
vendor-specific node-set function that can be discovered using function-
available) or XSLT 1.1/2.0 (making RTFs full blown node-sets).
|
|
||
Comment 10•23 years ago
|
||
actually, as we decided to move to specs compliance as far as RTFs go,
this is INVALID.
implicit conversion will die soon, and we hopefully will have a node-set extension
function by that time too.
/me lifts a leg at one of peterv's trees. Whatever that means
Status: UNCONFIRMED → RESOLVED
Closed: 23 years ago → 23 years ago
Resolution: --- → INVALID
You need to log in
before you can comment on or make changes to this bug.
Description
•