94223 - Dynamically created frame does not load or run JS code

Status: RESOLVED DUPLICATE of bug 92912

(Core :: DOM: Core & HTML, defect)

Description

[William J. Edney]

Mozilla .93 fails when trying to set the location of a frame that is part of a
dynamically generated frameset to a URL that contains JS code.

Reproduction:

I've attached a testcase that contains 3 files:

TestFrames.html -> contains the top level frameset. There are two frames, test1
and test2. We set only the 'src=' on test2.

ShouldExec.html -> the file which contains the code that we'll try to run.
WriteFrames.html -> the file which gets loaded from the top level frameset into
test2. It contains code that will dynamically build a frameset, 10 buffers that
will be named 'buffer_0' - 'buffer_9'. It will write document.write() this html
into parent.test1 (the 'test1' frame as specified in the top level frameset). It
will then try to access 'buffer_0's location to try to set it to 'ShouldExec.html'.

To reproduce, simply load TestFrames.html. You will first see an alert which
shows the html that will be dynamically written into the 'test1' by the script
executing in WriteFrames.html. Next, it depends on the line you're executing to
try to access 'buffer_0'. If you leave it the way it is in the attachment, the
browser will fail silently, never executing the code in ShouldExecute.html. If
you comment that line out and uncomment the line below it that says:

target.buffer_0.location = 'ShouldExec.html';

Then you get the following message in the JavaScript console:

'setting a property that has only a getter'

and the browser never executes the code in ShouldExecute.html. This description
is also documented in the comments in the testcase.

Actual Results:

The code in ShouldExecute.html doesn't get executed.

Expected Results:

The code in ShouldExecute.html gets executed. This happens in .92 and earlier.

This is occurring in the Mozilla .93 milestone release. Sorry I don't have the
build date, but I would have to quit and restart that version (currently running
.92 to file this). If you need specific build date, I can supply in a followup.

Comment 1

[William J. Edney]

Attachment: 1 of 3 to showcase this bug

Comment 2

[William J. Edney]

Attachment: 2 of 3 to showcase this bug

Comment 3

[William J. Edney]

Attachment: File 3 of 3 to showcase this bug.

Comment 4

[Brendan Eich [:brendan]]

jst, is this a dup of bug 92912?

/be

Comment 5

[Johnny Stenback (:jst)]

Yup, looks like a dup.

*** This bug has been marked as a duplicate of 92912 ***