30030 - Infinite loop if back button pressed leads to JS redirect

Status: RESOLVED FIXED

(Core :: DOM: Navigation, defect, P3)

Description

[Jerry Baker]

Step to reproduce:

1) Go to the URL http://www.camtechsystems.com.

2) Let the redirect happen (you will go to the "Old Browser" page).

3) Press 'Back'

4) When the "Old Browser" page loads again, press 'Back'

5) Continue this until Mozilla enters an infinite loop of reloads (it usually 
only takes twice or three times).

Occurs with M14 final and debug and opt builds pulled from CVS this morning at 
8:30.

Comment 1

[rogerl (gone)]

The JS code is executing fine each time the page is loaded. There's no script on 
the target page so I don't know why it's immediately going back again after the 
first couple of times.

cc'ing radha since I don't know where this goes, sorry.

Comment 2

[Warren Harris]

Sounds like one for Radha.

Comment 3

[Sean Richardson]

Looks conceptually similar to bug 20283, "Infinite loop when anchor has 
javascript:history.go(0)"

Comment 4

[Claudius Gayle]

bug 20283 has a lot to it. I think maybe radha should take a look and decide whether she wants to dupe it or not.

Comment 5

[Jerry Baker]

Well, the JavaScript history does not seem to be 0 here, and it is not in an 
anchor. But, if it is a dup then mark it. You guys know more than me ;-)

Comment 6

[leger]

Putting on [nsbeta2+][6/01] radar.  This work must be done by 06/01 or we may 
pull this for PR2.

Comment 7

[Claudius Gayle]

I can no longer reproduce this because the page has changed (no redirect) does anyone have (or can suggest) an independent 
testcase?

Comment 8

[don]

Move to M19.

Comment 9

[leger]

Due to slip in schedule, moving this bug from [6/01] to [Will be minus on 6/15] 
for fix deadline.

Comment 10

[Claudius Gayle]

Has anyone seen this recently or know of a testcase? I'd like to argue for this bug to be nsbeta2+(no date) but w/o a reproducible 
testcase my argument would have no merit (and rightly so).

Comment 11

[Radha on family leave (not reading bugmail)]

I belive the problem is still there. However, I dunno of a test case. 

Comment 12

[Jerry Baker]

Well, I tried to make a testcase, but since there are some other bugs in JS now, 
I cannot. Specifically, redirect pages that use document.location are not placed 
in the history. Oddly enough, pages that use location.replace() remain in the 
history.

Comment 13

[Jim Roskind]

Cleaning up status whiteboard marking beta2 minus (we're past 6/15)
Note that the discussion suggests that there is no longer a test case to be 
had... and perhaps this should be just marked "fixed" until/unless there is a 
current example.

Comment 14

[Jerry Baker]

Ummm. There can't be a testcase until someone fixes the JavaScript history bugs 
that exist. These bugs are blocking any testing on this.

Comment 15

[verah (gone)]

Nav triage team: [nsbeta3-]; marking dependency on 18321

Comment 16

[Radha on family leave (not reading bugmail)]

*** Bug 33311 has been marked as a duplicate of this bug. ***

Comment 17

[Radha on family leave (not reading bugmail)]

I don't see any redirection to teh old page in the above url. Please provide a 
good url

Comment 18

[Jerry Baker]

Bug 42723 is preventing me from testing this (read second comment about location 
= foo.bar acting like location.replace)

Comment 19

[Daniel Veditz [:dveditz]]

Adding nsbeta3 keyword to bugs which already have nsbeta3 status markings so 
the queries don't get all screwed up.

Comment 20

[Radha on family leave (not reading bugmail)]

42723 is fixed. Most of the JS hookup with SH is done. Marking this fixed. 
Reopen if the problem still exists.

Comment 21

[Claudius Gayle]

I need a hand here. If someone can accurately verify this fix please do. If someone can create
a JS testcase that demonstrates the fix - even better.

Comment 22

[Jerry Baker]

Mass removing self from CC list.

Comment 23

[Jerry Baker]

Now I feel sumb because I have to add back. Sorry for the spam.

Comment 24

[Tracy Walker [:tracy]]

mass remove verifyme requests greater than 4 months old