279878 - Unknown extension function shouldn't stop processing unless called

Status: RESOLVED FIXED

(Core :: XSLT, defect, P3)

Description

[Julian Reschke]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0

1) This is the default content type for files served by Apache 2 with extension
"xslt". So if this is an incorrect type, an issue for Apache2/httpd should be opened
2) IE accepts this content type

Reproducible: Always

Steps to Reproduce:

Actual Results:  
The message displayed is

"Error loading stylesheet: (null)"

Expected Results:  
Either the XSLT should be processed, or, at a minimum, Mozilla should display a
meaningful error message.

Comment 1

[Peter Van der Beken [:peterv]]

You should always test with a nightly, this bug is already fixed. However, we
signal an unknown function but I think we shouldn't.

Sicking: I see an error for myns:parseXml, even though there's an xsl:if with
function-available just before the call (<xsl:if test="... and
function-available('myns:parseXml')">). I don't think that should error and it
looks like fall-out from your fcp fix.

Comment 2

[Peter Van der Beken [:peterv]]

In particular I think the fcp() call in
http://lxr.mozilla.org/seamonkey/source/extensions/transformiix/source/xpath/ExprParser.cpp#578
is wrong (see 14.2). We probably need a new error code though, we need to
differentiate between functions with or without a prefix, the latter need the
fcp check but the former don't.

Comment 3

[Peter Van der Beken [:peterv]]

Attachment: v1 (diff -w)

Here's one way to fix it. Have to think it through wrt to XForms etc.

Comment 4

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Yeah, you're right. I was staring myself blind on the fcp section of the spec.
We should take this part into account too:

http://www.w3.org/TR/xslt#section-Extension-Functions

Comment 5

[Peter Van der Beken [:peterv]]

Attachment: v1.1

Comment 6

[Peter Van der Beken [:peterv]]

Attachment: v1.1 (diff -w)

Comment 7

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Comment on attachment 172459 [details] [diff] [review]
v1.1 (diff -w)

>Index: extensions/transformiix/source/xslt/txStylesheetCompiler.h
>+    /**
>+     * Should the expression be parsed in forwards compatible parsing mode.
>+     */

"Should the stylesheet..."

r=me

Comment 8

[Johnny Stenback (:jst)]

Comment on attachment 172459 [details] [diff] [review]
v1.1 (diff -w)

sr=jst

Comment 9

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

don't forget to remove the fcp function in the xforms context when you land this.