284167 - newegg.com - shopping cart doesn't work with FireFox

Status: RESOLVED FIXED

(Tech Evangelism Graveyard :: English US, defect)

Description

[Jeff Johnston]

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

Our customers experience more problems browsing our site using Firefox than any
other browser. My team uses Firefox almost exclusively, both internally (in the
office) and externally (at home), and are unable to recreate these issues. They
range from the site looking/functioning differently vs IE to being completely
inoperable. The bottom line here is that we love Firefox and our techie
customers do too, and we want to make them happy. Please let us know what we're
missing because this has been going on for quite awhile, even on our old site:
http://www.newegg.com/.

Reproducible: Couldn't Reproduce

Steps to Reproduce:
1.Visit http://www2.newegg.com/
2.Browse the site, try adding items to cart, etc.
3.Various bugs/errors seem to occur for our customers on a daily basis.

Actual Results:  
Unable to reproduce the daily errors we're alerted of ... only the easily
identifiable, isolated incidents.

Expected Results:  
We just want the site to run flawlessly because Firefox is the future.

I have emails from customers that I can share if you'd like to get in touch.

Comment 1

[Jeff Johnston]

Customer email from today:   I cleared out my cache and cookies and got the
following text when going to
newegg.com:

function
PopupWin(vName,vLink,vWidth,vHeight,vLeft,vTop,vScrollbar,vResizable) { var
para = ''; if(vWidth!=0) {para += ',width=' + vWidth;} if(vHeight!=0) {para
+= ',height=' + vHeight;} if(vLeft!=0) {para += ',left=' + vLeft;}
if(vTop!=0) {para += ',top=' + vTop;} if(vScrollbar!=0) {para +=
',scrollbars=' + vScrollbar;} if(vResizable!=0) {para += ',resizable=' +
vResizable;} javascript:window.open(vLink
,vName,'toolbar=0,menubar=0,location=0,directories=0,status=0' + para)

Comment 2

[Ben Fowler]

On http://www.newegg.com/ I saw these css errors in newegg.css:

{ FONT-SIZE: xx-small; FONT-FAMILY: Trebuchet MS bold & Italic } .MiddlePriceFont 
{ FONT-SIZE: x-small; FONT-FAMILY: Trebuchet MS bold & Italic } .LargePriceFont 
{ FONT-SIZE: small; FONT-FAMILY: Trebuchet MS bold & Italic } .SmallFont 

Lines 123 - 124

Error: Expected end of value for property but found '&'.  Error in parsing value
for property 'FONT-FAMILY'.  Declaration dropped.
Source File: http://www.newegg.com/newegg.css
Line: 123

but this probably isn't the cause of your presenting problem

Comment 3

[Ben Fowler]

There are no fewer than 369 reported markup errors, probably becase you have
an HTML DOCTYPE and some XML elements.

This probably accounts for comment 2.

If you really want to serve XHTML, you will probably have to make some
uncomfortable compromises.

See http://www.workingwith.me.uk/articles/scripting/mimetypes/

For internal use I use the weasally:

      HTML compatible XHTML 1.0 may be sent with the "text/html" mime type (note 
      that this applies to 1.0 only, not 1.1 or anything which comes after).

but if I were doing e-commerce, I would probably use HTML 4 sent as text/html,
having investigated this question first.

There is a lot of information about web development for standards
compliant browsers, but it may be patchy and not at all organised.
Most discussions concern rendering and display: You may want
to look at the links given in Bug 283110 comment 1.

I am not aware that mozilla.org has forums or other resources
specifically for organisations in your position - wishing to 
write a standards compliant b2p site, but there should be some
such resource. (Think of what would be a good home for posts such 
as http://forums.mozillazine.org/viewtopic.php?t=224833 ).

I would hope that get a speedy and efficent response from the
Mozilla community as there is no point is working on a browser
that is so heavenly minded as to be no earthly use.

Speaking personally, I would be pleased for problems
such as you describe to be as widely known as possible as great
software is made out of defect reports! Taking your statement 
that problems needing resolution crop up
on a daily basis, at face value, would you consider creating 
a blog of how you get on: What problems does one face
with a b2p application in these circumstances? You might find
a home for it on http://feedhouse.mozillazine.org/ or 
http://kb.mozillazine.org/Main_Page . I would be pleased for
mozilla to publicise real world stories of success or otherwise 
in your field. 

From your point of view, you might not want to be giving tuition to 
your competitors. 
http://www.everypoet.com/archive/poetry/Rudyard_Kipling/kipling_the_mary_gloster.htm

Comment 4

[Jeff Johnston]

USER COMMENTS FROM TODAY ~
- dislays good in ie. overlaping images/words in firefox
- In Firefox, increasing font sizes helped (it didn’t in IE), but then there
were overlap problems between fields of text.  
- I'm using Firefox in WinXP Pro @ 1600x1200. I've found that if I click outside
of the "main area" in the center, some random part gets selected within the main
area.  It's not a huge problem but somewhat annoying.
- The KEYWORD box in the Search section at the top of the page is too small.
When I type in the window the bottom of my text is not visible. I'm using
Firefox browser.
- I am using FireFox.  When I went into Cases and External Closures, there 
was a visual anomaly.  The green price (15.99 FedEx Saver Shipping) 
meshed behind the description.  I saw this on the VA3000BWA Black 
Computer Case Side Panel Window item.

Comment 5

[Kevin Brosnan]

One of the big issues you will run into is that your customers will likely be
running more extensions and browser tweaks than the average Firefox user.

common extensions that mess with pages
http://adblock.mozdev.org/
http://flashblock.mozdev.org/

common tweaks that users do that mess with pages via about:config
http://kb.mozillazine.org/About:config
Set "network.http.pipelining" to "true"
Set "network.http.proxy.pipelining" to "true"
Set "network.http.pipelining.maxrequests" to some number like 30.
-This means it will make 30 requests at once
right-click anywhere and select New-> Integer. Name
"nglayout.initialpaint.delay" and set its value to "0"

Comment 6

[Ben Fowler]

(In reply to comment #4)
> USER COMMENTS FROM TODAY ~

If you think that bugzilla is the right place for these reports then they should
be reported separately and marked as dependencies of this bug.

Each report should contain (as a minimum) a receipe for reproduction and a note of
the observed and expected behaviour.

Comment 7

[Ben Fowler]

(In reply to comment #4)
> - dislays good in ie. overlaping images/words in firefox

There is probably too little to go on. Do you have a URL where this
bad render can be seen? A reduced testcase?

Comment 8

[Ben Fowler]

Attachment: Testcase for segment 2 of comment 4

(In reply to comment #4)
> USER COMMENTS FROM TODAY ~
> - In Firefox, increasing font sizes helped (it didn’t in IE), but 
> then there were overlap problems between fields of text.  

Again URL? Testcase?

I can, in fact, see this on the index page. This page has 202 validation 
errors.

Now, whilst I have been criticised for asking for web authors to provide
valid mark-up before coming to Bugzilla with a bug report 
( Bug 283548 Comment 2 ), I find it a lot easier to work with clean code,
and it saves me very often from making silly errors such as Bug 281429
Comment 5 .

I enclose a shorter testcase with a more friendly name, one third the
size, valid HTML et cetera, et cetera.

I suspect that you have applied styles that restrict the vertical size
of layout blocks and require overlap. Certainly it displays well if
all styles are removed.

Without insight into your design goals, any progress of mine will be 
slow. Could you look at the mark up I have provided, note how your
styles get applied to it and compare with your deisgn, if you are still
having difficulty getting an end-to-end publication (your design -> your
customers' display) that works, owing to problems with Firefox, then
post again.

Comment 9

[Ben Fowler]

(In reply to comment #4)
> USER COMMENTS FROM TODAY ~
> - I'm using Firefox in WinXP Pro @ 1600x1200. I've found that if I click outside
> of the "main area" in the center, some random part gets selected within the main
> area.  It's not a huge problem but somewhat annoying.

I am using Mac OS X at 1152 by 768, so I will let this one by.

Comment 10

[Ben Fowler]

(In reply to comment #4)
> USER COMMENTS FROM TODAY ~
> - The KEYWORD box in the Search section at the top of the page is too small.
> When I type in the window the bottom of my text is not visible. I'm using
> Firefox browser.

This WFM, but the Category Drop done just below is both too small and too 
narrow.

You have probably set a style on these input elements somewhere that you
are unhappy with! It is not a constraint with HTML or a feature/bug
in Firefox.

Comment 11

[Ben Fowler]

(In reply to comment #4)
> USER COMMENTS FROM TODAY ~
> - I am using FireFox.  When I went into Cases and External Closures, there 
> was a visual anomaly.  The green price (15.99 FedEx Saver Shipping) 
> meshed behind the description.  I saw this on the VA3000BWA Black 
> Computer Case Side Panel Window item.

This WFM, though I can't be sure that I followed the receipe completely.

Comment 12

[hfritze]

Just wanted to verify this bug as of 8/01/05. I'm using Firefox 1.05 and
Newegg's shopping cart remains empty after selecting "add to cart." Cookies was
enabled with "ask" option.

This also happens at www.lloydhearingaid.com and several other sites I've tried.

I alway have to revert to IE and I hate that.

Comment 13

[Ben Fowler]

(In reply to comment #12)
> Just wanted to verify this bug as of 8/01/05. I'm using Firefox 1.05 and
> Newegg's shopping cart remains empty after selecting "add to cart." 

This bug is mainly about about CSS semantics and rendering, and to a lesser
extent, about developing business-to-public applications. "shopping cart
remains empty" looks like a new problem, and if you think that it is a defect
in Firefox, it is probably a new Report.

Otherwise, I suspect that you want to look through the relevant mozillazine
forums.

Comment 14

[Ryuichi KUBUKI]

I think it should be moved to tech evangelism, though it's a rare case that the administrator of the web site reports it.

Comment 15

[Thomas K. (:tom)]

I too can't reproduce these, but I would like to note that the slideouts on the side of the page which work fine in Firefox, IE, and Safari fail in Camino.  These are basically extra menu items found when waving over a major category, and has really saddened my camino usage.

I doubt the Newegg team lacks at least one mac user, but if they do I'm willing to sell them a top of the line mini mac on a g4 processor with dvd burning for less than apple does ;).

Using Camino 1.0 Release, there's no about: page in this browser so that's all I got, except Version 2006021400 (1.0)

Comment 16

[C. Fox]

I'm seeing this same issue in Firefox 2.0, but I seem to recall it working properly in 1.5, I'll have to test some more.

Another area this crops up:
Sign in.
Make a permanent "Wish List"
Try to remove an item from the wish list (by changing quantity to "0" and hitting "update".
Expected behavior: List refreshes with the item removed.
Actual Behavior: The user is taken back to the login screen (and upon logging in the wish list has not been changed).

Comment 17

[Thomas K. (:tom)]

Jeff:

If you're still following this bug, I'd like to note that for some reason you have the 'gift card' name and password fields tagged the same as the actual name and password fields, for Firefox and Camino try to autofill them (and successfully do so, with the user name and password, which breaks things later).  This problem really becomes noticeable because you don't try to verify the gift card code until the very end of the transaction, and there's no way to remove it then.  Please consider changing their ids.

Comment 18

[Chris Lawson (gone)]

Is anyone still seeing the *original* bug here?

It doesn't sound like it to me. Comment 15 is probably bad browser-sniffing (which should be filed as a separate bug), and comment 17 should *definitely* be filed as a separate bug because it has nothing whatsoever to do with this one.

It also sounds like comment 16 is a separate bug.

I'm going to mark this one FIXED; if people who wrote comments 15-17 wish to file new bugs on those problems, please do so.

Jeff, thanks for your help thus far.