[Mac][MC] Adjust the size of the status item view based on the unread count.

[Mac][MC] Adjust the size of the status item view based on the unread count.

This also adjusts the baseline and min-X of the unread count.

BUG=243935
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=202962

[Robert Sesek, 2013-05-28 19:07:42]

Screen shot: http://cl.ly/image/0s3F1g341n0i

[sail, 2013-05-28 19:40:22]

lgtm

https://codereview.chromium.org/16159005/diff/1/ui/message_center/cocoa/statu...
File ui/message_center/cocoa/status_item_view.mm (right):

https://codereview.chromium.org/16159005/diff/1/ui/message_center/cocoa/statu...
ui/message_center/cocoa/status_item_view.mm:21: // The width of the status bar
item when it's the icon, a number, and a + sign.
I was a little surprised by this approach. I thought you'd size the view based
on the width of the unread count string.

I think this will work though since numbers and the plus sign should be the same
in all locales.

[dewittj, 2013-05-28 19:49:32]

On 2013/05/28 19:40:22, sail wrote:
> lgtm
> 
>
https://codereview.chromium.org/16159005/diff/1/ui/message_center/cocoa/statu...
> File ui/message_center/cocoa/status_item_view.mm (right):
> 
>
https://codereview.chromium.org/16159005/diff/1/ui/message_center/cocoa/statu...
> ui/message_center/cocoa/status_item_view.mm:21: // The width of the status bar
> item when it's the icon, a number, and a + sign.
> I was a little surprised by this approach. I thought you'd size the view based
> on the width of the unread count string.
> 
> I think this will work though since numbers and the plus sign should be the
same
> in all locales.

There are some inconsistencies with this as shown in the screenshot:
* In the case of no unread notifications, the screenshot shows 6 pixels to the
left of the notification bubble and 7 to the right.  Should be the same on both
sides.
* Similarly, 9+ has 6 pixels on the left and 5 on the right.
* The example with the "1" unread appears correct, with 6 on both sides.  How do
the other digits look?

[Robert Sesek, 2013-05-29 16:43:03]

PTAL. Switched to the approach Sailesh was expecting, using AppKit to provide
the subpixel bounding rectangle for the font. This yields the correct spacing in
all cases.

[sail, 2013-05-29 16:58:43]

lgtm

https://codereview.chromium.org/16159005/diff/5002/ui/message_center/cocoa/st...
File ui/message_center/cocoa/status_item_view.mm (right):

https://codereview.chromium.org/16159005/diff/5002/ui/message_center/cocoa/st...
ui/message_center/cocoa/status_item_view.mm:65: if (unreadCount_ == 0) {
Small nit. It might be better to do:
frame.size.width = kStatusItemLength;
countString = [self ...];
if (countString) {
  frame.size.width += ...;
}
This would better match the drawRect code.

https://codereview.chromium.org/16159005/diff/5002/ui/message_center/cocoa/st...
ui/message_center/cocoa/status_item_view.mm:68: // Get the subpixel bounding
rectangle for the string. -size doesn't yield
I didn't know that. Should I stop using -[NSAttributedString size] in my code?

[dewittj, 2013-05-29 17:20:30]

lgtm

[Robert Sesek, 2013-05-29 17:52:30]

https://codereview.chromium.org/16159005/diff/5002/ui/message_center/cocoa/st...
File ui/message_center/cocoa/status_item_view.mm (right):

https://codereview.chromium.org/16159005/diff/5002/ui/message_center/cocoa/st...
ui/message_center/cocoa/status_item_view.mm:65: if (unreadCount_ == 0) {
On 2013/05/29 16:58:43, sail wrote:
> Small nit. It might be better to do:
> frame.size.width = kStatusItemLength;
> countString = [self ...];
> if (countString) {
>   frame.size.width += ...;
> }
> This would better match the drawRect code.

Done.

https://codereview.chromium.org/16159005/diff/5002/ui/message_center/cocoa/st...
ui/message_center/cocoa/status_item_view.mm:68: // Get the subpixel bounding
rectangle for the string. -size doesn't yield
On 2013/05/29 16:58:43, sail wrote:
> I didn't know that. Should I stop using -[NSAttributedString size] in my code?

No, it's fine for most cases. -size basically returns the unhinted and
un-antialiased size, i.e. the typographic bounds. This call returns the size of
the glyph that will actually be drawn. In this case, since I want

[uniform space] [icon] [uniform space] [text] [uniform space]

I need to know the precise pixel size of [text] to ensure that the final space
is in fact uniform.

[commit-bot: I haz the power, 2013-05-29 17:53:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rsesek@chromium.org/16159005/15001

[commit-bot: I haz the power, 2013-05-29 20:56:33]

Change committed as 202962