417413 - can't scroll search engines with mouse wheel

Status: RESOLVED WONTFIX

(Firefox :: Search, enhancement)

Description

[Hugo Osvaldo Barrera]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9b3) Gecko/2008020514 Firefox/3.0b3
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9b3) Gecko/2008020514 Firefox/3.0b3

the summary just about says it all. i can no longer scroll though the search engines in my list with my mouse wheel (i verified that they're all there.. but wheel does nothing)

Reproducible: Always

Steps to Reproduce:
1.click on search bar, just where u type the search text
2.move wheel to see it no longer works
3.
Actual Results:  
nothing happens

Expected Results:  
should scroll though search engines like beta2

--

Comment 1

[Stephen Donner [:stephend] Not actively reading bugmail]

Intentional change from bug 410603; see the modified key combination there.

Comment 2

[Hugo Osvaldo Barrera]

ability to be able to enable/disable this feature from the options menu or "about:config" would be nice, since holding down ctrl is not as comfortable or fast, AND many are already very used to using the mouse wheel by now

Comment 5

[(no longer active)]

This is a valid enhancement request, which many people seem to share.  Taking over to add a pref for this purpose.

Comment 6

[(no longer active)]

Oops, I made a mistake in setting the dependency.  Sorry for the bugspam :-/

Comment 8

[(no longer active)]

Attachment: Patch (v1)

Added the browser.search.scrollSwitchKey pref which enables any modifier key to be required for switching search engines via mouse scroll.  Setting this pref to 0 restores the old behavior (before bug 410603).  The defaults are 8 (meta) for Mac and 2 (ctrl) for other platforms.

Requesting review from gavin who reviewed my patch on bug bug 410603.  This should not take a lot of time to review.

Comment 9

[(no longer active)]

Gavin: any chance to get a review here, so that I could try getting an approval on this patch?  Thanks!

Comment 10

[Nahor]

It seems that the unselected key are "optional" instead of "required unset". I mean, if the required key is <ctrl>, the <ctrl>+<shift> will work too.

That's actually the current behavior but is it correct UI-wise? For instance <ctrl>+<wheel> on the HTML page zoom the page in and out but <ctrl>+<shift> doesn't.

Comment 11

[(no longer active)]

(In reply to comment #10)
> It seems that the unselected key are "optional" instead of "required unset". I
> mean, if the required key is <ctrl>, the <ctrl>+<shift> will work too.

Yes.

> That's actually the current behavior but is it correct UI-wise? For instance
> <ctrl>+<wheel> on the HTML page zoom the page in and out but <ctrl>+<shift>
> doesn't.

As long as the latter form does not do something else, I don't see why that would be a problem.

Comment 12

[Nahor]

(In reply to comment #11)
> > That's actually the current behavior but is it correct UI-wise? For instance
> > <ctrl>+<wheel> on the HTML page zoom the page in and out but <ctrl>+<shift>
> > doesn't.
> 
> As long as the latter form does not do something else, I don't see why that
> would be a problem.
> 

The problem is that people may learn a bad habit. The day the others forms are used, they will need to unlearn that habit (which is hard and thus this bug report).

I personally think it's a bad idea to be too lax just because one can. This tends to pose problems on the long run.


Also, my earlier example was bad. I didn't notice that <ctrl>+<shift>+<wheel> was actually used on the HTML page to scroll through the history (there wasn't any history when tried). 
<ctrl>+<alt>+<wheel> does behave like <ctrl>+<wheel>. And <ctrl>+<alt>+<shift>+<wheel> behaves like <ctrl>+<shift>+<wheel>. 

So at least this behavior is consistent (well sort of, <ctrl>+<shift>+<n> doesn't do anything while <ctrl>+<n> opens a new window, but then again, it's about the keyboard, not the mouse).

Anyway, that was just my personal opinion. Feel free to disregard :)

Comment 13

[(no longer active)]

Let's hear what Gavin and others think about this.  The necessary code changes for this behavior are simple, and I'm not opposed to any of these two behaviors per se.

Comment 15

[:Gavin Sharp [email: [e-mail]]]

We're not going to take this for RC2, and I'm not really sure we should take it at all. Seems like overkill.

Comment 16

[(no longer active)]

(In reply to comment #15)
> We're not going to take this for RC2, and I'm not really sure we should take it
> at all. Seems like overkill.

I don't have a strong motive here, it just seemed to me that some users think this would be desirable.  Feel free to WONTFIX this if you feel otherwise.

Comment 17

[Mike Connor [:mconnor]]

We have enough hidden prefs as is, we should be having fewer, not more.

WONTFIX.