53520 - XUL Template with lots of data only partially displayed

Status: VERIFIED INVALID

(Core :: XUL, defect, P3)

Description

[Matthew Cline]

Linux build 20000920, Linxu 2.2.14 i686, RedHat 6.1

When looking into the limits of XUL templates, I've found two test cases
where XUL seems to stop working.  Both testcases are based off of the
examples given in http://www.mozilla.org/docs/xul/xulnotes/template-nested.html

Both examples contain one piece of top level data, with a whole bunch of
sub-level data; the top level data is simply the string "alt", while each
sub-level piece of data is a number (1, 2, 3, ...) .  The first example has
150 sub-elements, but only 86 are show.  (Attachments will be uploaded
shortly).

The second example will be given in another bug.

Comment 1

[Matthew Cline]

Attachment: Testcase of XUL template

Comment 2

[Matthew Cline]

Attachment: Testace data for nested XUL template

Comment 3

[Matthew Cline]

Making summary more descriptive.  Also, it was 125 sub-elements, not 150.

The other testcase is bug 53521.

Comment 4

[Matthew Cline]

Oops!  The RDF file I gave is malformed, in that the closing </rdf:Seq>
tag is missing; when it's put back in, it displays fine.  Though it might
still be a bug that with the malformed RDF file Mozilla does almost, but not
quite, the right thing.

I'll be uploading the corrected RDF file shortly.

Comment 5

[Matthew Cline]

Attachment: Corrected RDF data for the XUL template testcase

Comment 6

[Peter Trudelle]

->waterson

Comment 7

[Chris Waterson]

Malformed RDF -> malformed results. Sorry! (It would IMO be impractical to
validate the RDF/XML document before using it, even though, in theory, we should
reject it.)

I'm assuming that with well-formed & valid RDF/XML, this works...if not re-open.

Comment 8

[Matthew Cline]

Well, it does give this error output for the non-closed datasource:

XML Error in file 'file:///usr/home/matt/temp/quux/data1.rdf', Line Number: 393,
Col Number: 6, Description: mismatched tag
Source Line:     </nc:subheadings>

Didn't notice this the first time around, since I was redirecting stderr and
stdout to /dev/null.  I guess this error message should be enough for any
RDF author.

Marking VERIFIED.