Fix DPI scaling on Linux with GTK3

Fix DPI scaling on Linux with GTK3

The device scale was determined by querying the value of the
"gtk-xft-dpi" property from GtkSettings. In GTK2, this always returned
the value of the "Xft/DPI" XSettings property. However, in GTK3, this
returns the value of "Gdk/UnscaledDPI" if it is set, resulting in the
wrong device scale being calculated on high DPI screens.

Instead, we now query the "gdk-window-scaling-factor" setting from
GdkScreen, which is derived from the value of the
"Gdk/WindowScalingFactor" XSettings property. If this property is not
set then we fall back to the previous code path, as environments that
don't provide Gdk/WindowScalingFactor also won't provide
Gdk/UnscaledDPI.

BUG=716135

[Chris Coulson, 2017-04-27 23:36:21]

chris.coulson@canonical.com changed reviewers:
+ erg@chromium.org

[Chris Coulson, 2017-04-27 23:43:31]

On 2017/04/27 23:36:22, Chris Coulson wrote:

Note, the last time I submitted something for review it was pointed out that my
account is not part of the CCLA from Canonical. I'm still not sure who manages
that internally (I'm trying to find out).

[Tom (Use chromium acct), 2017-04-27 23:49:02]

Description was changed from

==========
Fix DPI scaling on Linux with GTK3

The device scale was determined by querying the value of the
"gtk-xft-dpi" property from GtkSettings. In GTK2, this always returned
the value of the "Xft/DPI" XSettings property. However, in GTK3, this
returns the value of "Gdk/UnscaledDPI" if it is set, resulting in the
wrong device scale being calculated on high DPI screens.

Instead, we now query the "gdk-window-scaling-factor" setting from
GdkScreen, which is derived from the value of the
"Gdk/WindowScalingFactor" XSettings property. If this property is not
set then we fall back to the previous code path, as environments that
don't provide Gdk/WindowScalingFactor also won't provide
Gdk/UnscaledDPI.

BUG=716135
==========

to

==========
Fix DPI scaling on Linux with GTK3

The device scale was determined by querying the value of the
"gtk-xft-dpi" property from GtkSettings. In GTK2, this always returned
the value of the "Xft/DPI" XSettings property. However, in GTK3, this
returns the value of "Gdk/UnscaledDPI" if it is set, resulting in the
wrong device scale being calculated on high DPI screens.

Instead, we now query the "gdk-window-scaling-factor" setting from
GdkScreen, which is derived from the value of the
"Gdk/WindowScalingFactor" XSettings property. If this property is not
set then we fall back to the previous code path, as environments that
don't provide Gdk/WindowScalingFactor also won't provide
Gdk/UnscaledDPI.

BUG=716135
==========

[Chris Coulson, 2017-04-28 16:27:49]

On 2017/04/27 23:43:31, Chris Coulson wrote:
> On 2017/04/27 23:36:22, Chris Coulson wrote:
> 
> Note, the last time I submitted something for review it was pointed out that
my
> account is not part of the CCLA from Canonical. I'm still not sure who manages
> that internally (I'm trying to find out).

The CCLA issue *should* be resolved now.

[Elliot Glaysher, 2017-04-28 18:12:14]

erg@chromium.org changed reviewers:
+ oshima@chromium.org

[Elliot Glaysher, 2017-04-28 18:12:15]

+oshima who actually understands dpi stuff

lgtm, but oshima should also probably look at this since i'm not a dpi
specialist.

[oshima, 2017-04-28 18:31:18]

thank you for fixing this!

lgtm with one nit

https://codereview.chromium.org/2852593002/diff/1/chrome/browser/ui/libgtkui/...
File chrome/browser/ui/libgtkui/gtk_ui.cc (right):

https://codereview.chromium.org/2852593002/diff/1/chrome/browser/ui/libgtkui/...
chrome/browser/ui/libgtkui/gtk_ui.cc:328: float GetScaleFromDPI() {
nit: GetScaleFromXftDPI

[Chris Coulson, 2017-04-28 18:40:50]

https://codereview.chromium.org/2852593002/diff/1/chrome/browser/ui/libgtkui/...
File chrome/browser/ui/libgtkui/gtk_ui.cc (right):

https://codereview.chromium.org/2852593002/diff/1/chrome/browser/ui/libgtkui/...
chrome/browser/ui/libgtkui/gtk_ui.cc:328: float GetScaleFromDPI() {
On 2017/04/28 18:31:18, oshima wrote:
> nit: GetScaleFromXftDPI
> 

Acknowledged.

[Tom (Use chromium acct), 2017-05-01 05:43:06]

The CQ bit was checked by thomasanderson@google.com

[Tom (Use chromium acct), 2017-05-01 05:43:06]

The patchset sent to the CQ was uploaded after l-g-t-m from erg@chromium.org,
oshima@chromium.org
Link to the patchset: https://codereview.chromium.org/2852593002/#ps20001
(title: "Rename GetScaleFromDPI to GetScaleFromXftDPI")

[commit-bot: I haz the power, 2017-05-01 05:43:22]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-01 05:43:23]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-01 05:43:24]

The author chris.coulson@canonical.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[oshima_google, 2017-05-01 17:41:05]

The CQ bit was checked by oshima@google.com

[oshima_google, 2017-05-01 17:41:22]

The CQ bit was unchecked by oshima@google.com

[oshima, 2017-05-01 17:41:29]

The CQ bit was checked by oshima@chromium.org

[commit-bot: I haz the power, 2017-05-01 17:41:53]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-01 17:41:54]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-01 17:41:55]

The author chris.coulson@canonical.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[Tom (Use chromium acct), 2017-05-03 00:15:55]

pinging Chris, did you sign the CLA?

I wouldn't want to ship Gtk3 without this, so we need to land this so it can be
merged to M59.

[Chris Coulson, 2017-05-03 08:57:09]

On 2017/05/03 00:15:55, Tom Anderson wrote:
> pinging Chris, did you sign the CLA?
> 
> I wouldn't want to ship Gtk3 without this, so we need to land this so it can
be
> merged to M59.

As far as I'm aware, I was added to the list of employees authorized to submit
contributions on behalf of Canonical, which has signed the corporate CLA. If
that hasn't actually happened yet, I'll chase it up today.

[oshima_google, 2017-05-03 14:16:59]

On 2017/05/03 08:57:09, Chris Coulson wrote:
> On 2017/05/03 00:15:55, Tom Anderson wrote:
> > pinging Chris, did you sign the CLA?
> > 
> > I wouldn't want to ship Gtk3 without this, so we need to land this so it can
> be
> > merged to M59.
> 
> As far as I'm aware, I was added to the list of employees authorized to submit
> contributions on behalf of Canonical, which has signed the corporate CLA. If
> that hasn't actually happened yet, I'll chase it up today.

FWIW, canonical isn't listed in AUTHORS files. (not sure how/if it's linked to
CLA though)

[Tom (Use chromium acct), 2017-05-03 17:30:51]

The CQ bit was checked by thomasanderson@google.com

[commit-bot: I haz the power, 2017-05-03 17:31:17]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-03 17:31:18]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-03 17:31:19]

The author chris.coulson@canonical.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[Chris Coulson, 2017-05-03 19:35:38]

On 2017/05/03 08:57:09, Chris Coulson wrote:
> On 2017/05/03 00:15:55, Tom Anderson wrote:
> > pinging Chris, did you sign the CLA?
> > 
> > I wouldn't want to ship Gtk3 without this, so we need to land this so it can
> be
> > merged to M59.
> 
> As far as I'm aware, I was added to the list of employees authorized to submit
> contributions on behalf of Canonical, which has signed the corporate CLA. If
> that hasn't actually happened yet, I'll chase it up today.

So, I pinged someone on our legal team today, and I'm basically waiting on them
to notify Google that I'm to be added to this list. Sorry about this, I didn't
realize it would take so long :/

[Chris Coulson, 2017-05-05 13:33:44]

On 2017/05/03 19:35:38, Chris Coulson wrote:
> On 2017/05/03 08:57:09, Chris Coulson wrote:
> > On 2017/05/03 00:15:55, Tom Anderson wrote:
> > > pinging Chris, did you sign the CLA?
> > > 
> > > I wouldn't want to ship Gtk3 without this, so we need to land this so it
can
> > be
> > > merged to M59.
> > 
> > As far as I'm aware, I was added to the list of employees authorized to
submit
> > contributions on behalf of Canonical, which has signed the corporate CLA. If
> > that hasn't actually happened yet, I'll chase it up today.
> 
> So, I pinged someone on our legal team today, and I'm basically waiting on
them
> to notify Google that I'm to be added to this list. Sorry about this, I didn't
> realize it would take so long :/

This has now been done.

[Tom (Use chromium acct), 2017-05-05 15:33:20]

The CQ bit was checked by thomasanderson@google.com

[commit-bot: I haz the power, 2017-05-05 15:33:41]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-05 15:33:42]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-05 15:33:43]

The author chris.coulson@canonical.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[Elliot Glaysher, 2017-05-05 17:18:31]

On 2017/05/05 15:33:43, commit-bot: I haz the power wrote:
> The author mailto:chris.coulson@canonical.com has not signed Google
Contributor License
> Agreement. Please visit https://cla.developers.google.com to sign and manage
> CLA.

I verified internally that Canonical Limited signed the GCLA. It looks like the
check is failing because Canonical isn't listed in the AUTHORS file. I've opened
https://codereview.chromium.org/2860413002 to fix this.

[Chris Coulson, 2017-05-05 17:22:07]

On 2017/05/05 17:18:31, Elliot Glaysher wrote:
> On 2017/05/05 15:33:43, commit-bot: I haz the power wrote:
> > The author mailto:chris.coulson@canonical.com has not signed Google
> Contributor License
> > Agreement. Please visit https://cla.developers.google.com to sign and manage
> > CLA.
> 
> I verified internally that Canonical Limited signed the GCLA. It looks like
the
> check is failing because Canonical isn't listed in the AUTHORS file. I've
opened
> https://codereview.chromium.org/2860413002 to fix this.

Thanks, and sorry this has ended up being such a pain.

[Chris Coulson, 2017-05-05 17:22:08]

On 2017/05/05 17:18:31, Elliot Glaysher wrote:
> On 2017/05/05 15:33:43, commit-bot: I haz the power wrote:
> > The author mailto:chris.coulson@canonical.com has not signed Google
> Contributor License
> > Agreement. Please visit https://cla.developers.google.com to sign and manage
> > CLA.
> 
> I verified internally that Canonical Limited signed the GCLA. It looks like
the
> check is failing because Canonical isn't listed in the AUTHORS file. I've
opened
> https://codereview.chromium.org/2860413002 to fix this.

Thanks, and sorry this has ended up being such a pain.

[Elliot Glaysher, 2017-05-05 17:22:37]

On 2017/05/05 17:22:08, Chris Coulson wrote:
> On 2017/05/05 17:18:31, Elliot Glaysher wrote:
> > On 2017/05/05 15:33:43, commit-bot: I haz the power wrote:
> > > The author mailto:chris.coulson@canonical.com has not signed Google
> > Contributor License
> > > Agreement. Please visit https://cla.developers.google.com to sign and
manage
> > > CLA.
> > 
> > I verified internally that Canonical Limited signed the GCLA. It looks like
> the
> > check is failing because Canonical isn't listed in the AUTHORS file. I've
> opened
> > https://codereview.chromium.org/2860413002 to fix this.
> 
> Thanks, and sorry this has ended up being such a pain.

No, this part is our fault.

[Elliot Glaysher, 2017-05-05 18:12:27]

The CQ bit was checked by erg@chromium.org

[commit-bot: I haz the power, 2017-05-05 18:13:03]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-05 18:13:04]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-05 18:13:05]

The author chris.coulson@canonical.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[Elliot Glaysher, 2017-05-05 18:24:47]

The CQ bit was checked by erg@chromium.org

[commit-bot: I haz the power, 2017-05-05 18:25:20]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-05 18:25:22]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-05 18:25:26]

The author chris.coulson@canonical.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[johnf, 2017-05-08 09:13:51]

On 2017/05/05 18:25:26, commit-bot: I haz the power wrote:
> The author mailto:chris.coulson@canonical.com has not signed Google
Contributor License
> Agreement. Please visit https://cla.developers.google.com to sign and manage
> CLA.

Hi,

Can someone kick off commit-bot again? The AUTHORS patch landed 10 mins after
the last time it ran here. Bad timing!!

Cheers,
John

[oshima, 2017-05-08 09:57:24]

The CQ bit was checked by oshima@chromium.org

[commit-bot: I haz the power, 2017-05-08 09:58:05]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-08 09:58:06]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-08 09:58:07]

The author chris.coulson@canonical.com has not signed Google Contributor License
Agreement. Please visit https://cla.developers.google.com to sign and manage
CLA.

[oshima, 2017-05-12 00:34:04]

On 2017/05/08 09:58:07, commit-bot: I haz the power wrote:
> The author mailto:chris.coulson@canonical.com has not signed Google
Contributor License
> Agreement. Please visit https://cla.developers.google.com to sign and manage
> CLA.

erg@, would you mind if I land this instead for now?

[oshima, 2017-05-12 00:35:00]

On 2017/05/12 00:34:04, oshima wrote:
> On 2017/05/08 09:58:07, commit-bot: I haz the power wrote:
> > The author mailto:chris.coulson@canonical.com has not signed Google
> Contributor License
> > Agreement. Please visit https://cla.developers.google.com to sign and manage
> > CLA.
> 
> erg@, would you mind if I land this instead for now?

Ah, nvm. thomasanderson@ did it already.