Closed Bug 360854 Opened 18 years ago Closed 17 years ago

Current tab can get "pushed" offscreen by the tab scroll buttons coming up.

Categories

(Camino Graveyard :: Tabbed Browsing, defect, P1)

PowerPC
macOS
defect

Tracking

(Not tracked)

RESOLVED FIXED
Camino1.6

People

(Reporter: froodian, Assigned: stuart.morgan+bugzilla)

Details

(Keywords: fixed1.8.1.8)

Attachments

(1 file, 1 obsolete file)

This can happen with a real use-case, I'm just using a very narrow window for illustration's purposes.

1. Open a Camino window, resize to 525 px
2. Open five tabs (max that will fit in a window that size)
3. With the rightmost tab focused, open a link in the background

What happens:  Currently focused tab is now off-screen
What should happen:  Implementation should keep the current tab visible
Status: UNCONFIRMED → NEW
Ever confirmed: true
Confirming bug, working on fix.
Status: NEW → ASSIGNED
Gives us selected tab is always visible for free which is quite a nice bonus.

The maths might be a little bit assumptive, but I think it covers all cases.

Spaced out and commented this part of the method to make it easier to read - it took me a few minutes to understand it... and I wrote it!
Attachment #248792 - Flags: review?(stuart.morgan)
Comment on attachment 248792 [details] [diff] [review]
Makes sure that overflow tabs doesn't displace the currently selected tab

> Gives us selected tab is always visible for free which is quite a nice bonus.

Except that preventing people from scrolling away from the selected tab defeats part of the purpose of scrolling tabs.
Attachment #248792 - Flags: review?(stuart.morgan) → review-
The way this should probably be done is to check if the current tab is visible before recomputing the visible tab count, and then adjusting if it's not visible by the end.
When tabs go offscreen it is also IMPOSSIBLE to close them without closing others. Please make scrolling arrows to navigate additional tabs.
We have them on trunk; this bug is about a specific minor bug in that implementation.
Assignee: des.elliott → froodian
Status: ASSIGNED → NEW
Depends on: 374623
No longer depends on: 374623
No longer blocks: 355493
Target Milestone: --- → Camino1.6
Setting priority per 1.6 roadmap.
Priority: -- → P1
Attached patch fixSplinter Review
Sorry Ian, this annoyed me enough recently that I just fixed it. This also fixes the related problem of having the selected tab fall off the edge when shrinking a window.
Assignee: froodian → stuart.morgan
Attachment #248792 - Attachment is obsolete: true
Status: NEW → ASSIGNED
Attachment #271682 - Flags: review?
Attachment #271682 - Flags: review? → review?(froodian)
Attachment #271682 - Flags: review?(froodian) → review?(murph)
(In reply to comment #8)
> Created an attachment (id=271682) [details]
> fix

I wonder if |layoutButtonsPreservingVisibility:| is a little misleading in that visibility can still be preserved even when supplying 'NO' as the argument.  This confused me upon first glance, as it would seem that explicitly specifying a certain value would dictate only that behavior.  I needed to scan through the method implementation to learn how it actually worked.  

What was unclear to me at first was that preserveVisibility = NO does not mean never maintain visibility, but rather 'maintain visibility if needed'.  YES, more predictably, forces visibility no matter what.

In defense, you did comment inside the method explaining |keepCurrentTabVisible| and it is private and therefore usually called with implementation knowledge in mind.

This method name's self-documenting aspect aside, the actual visibility preservation itself works perfectly in every single tab use case I performed.  I'll wait to hear your thoughts about what I mentioned above; If we decide that's not a problem, r=me.
Comment on attachment 271682 [details] [diff] [review]
fix

I see your point, but I wasn't able to come up with a better name (and my earlier attempt to have that logic outside the layout method was ugly). I'm not too concerned, since I think "Yes, I care about preserving visibility" and "No, I don't care, so do whatever" are reasonable; I'll clarify the method-level comment if no-one has any suggestions for a clearer name.
Attachment #271682 - Flags: superreview?(mikepinkerton)
Attachment #271682 - Flags: review?(murph)
Attachment #271682 - Flags: review+
Comment on attachment 271682 [details] [diff] [review]
fix

sr=pink
Attachment #271682 - Flags: superreview?(mikepinkerton) → superreview+
Landed on trunk and MOZILLA_1_8_BRANCH.
Status: ASSIGNED → RESOLVED
Closed: 17 years ago
Keywords: fixed1.8.1.7
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.