| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In commit 50564cde49ca2e17fb7e59f36a35d61c2cbef1af, we removed support
for configuring the various "sleep state" buttons. However, the power
button might need different behaviour based on the machine, which we
cannot always detect.
In later commits, we'll hard-code the actions for tablets and virtual
machines. Note that "power off" is not an option as this would make
the default action too destructive. It is recommended that you use a
custom shortcut for this instead.
https://bugzilla.gnome.org/show_bug.cgi?id=755953
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The current default of "medium" is problematic because "medium"
and "full" are coarse descriptions that mean different things to
different font formats. This leads to widely inconsistent font rendering
and mishandling of popular (TrueType) fonts.
First, there are 2 major ways to make fonts readable on low-DPI screens:
- X-and-Y-fitting, as done on Windows for a long time before ClearType
and DirectWrite. Glyph outlines are shuffled about (read: distorted) to
fit them to the pixel grid. The goal is maximum sharpness at the
expense of outline fidelity and glyph spacing.
- Y-only-fitting: Done by ClearType, DirectWrite and Adobe's proprietary
font rendering engine to exploit the RGB-subpixel-stripe configuration
of modern LCDs. Outlines are only fitted to the Y-axis, so the
rasterizer can use tricks ("subpixel rendering") on the X-axis to
enhance contrast or perceived resolution. Since horizontal (X-axis)
metrics are unchanged, glyph spacing stays as is.
The higher the resolution (as in DPI) of a screen, the less important
hinting/grid-fitting becomes.
With the fontconfig property "autohint" set to false:
- FreeType's TrueType (.ttf) engine will use the native hinting bytecode
in the font. Depending on how much effort went into hinting the font
and whether it was optimized for ClearType, glyphs might be
grid-fitted to X-and-Y-axis or Y-axis only. ClearType is not yet
fully handled by FreeType (2.6.1), so popular fonts from the Windows
world will be more or less distorted. "Medium" seems to be a variant
of "full".
- the OpenType/CFF engine uses a different approach to rendering for
low-DPI screens. Font designers of Type 1/CFF fonts just specify
metadata that tell the hinting engine how to fit glyphs to the Y-axis
(no X-fitting!), so the details of grid-fitting are deferred to the
smarts of the hinting engine. "Medium" is equivalent to "full" here.
- The Postscript/Type 1 engine is the predecessor to the CFF engine in
spirit. Hinting works as with CFF. The Type 1 engine in FreeType is
older than Adobe's contributed CFF engine and hinting results are of
far lower quality. "Medium" is mostly "Full".
- "Slight" will always trigger the autohinter regardless of font format.
Y-fitting only.
Since the computer world is comprised of more than one typeface and more
than one font format, users will see jarringly inconsistent and
distorted fonts with "medium" and "full" due to the differences in the
engines and inconsistent quality of many typefaces in the wild. The
problem is nicely visible when you go medium or full and mix font
formats. A default installation of Fedora will have Cantarell in
OpenType/CFF (.otf) format in the UI and DejaVu (.ttf) on the console.
Cantarell is Y-fitted (and slightly emboldened), DejaVu X-and-Y-fitted.
The difference is jarring.
With the fontconfig property "autohint" set to true, FreeType's
autohinter is triggered and native hinting disabled. "Slight" will fit
to the Y-axis only, "medium" and "full" fit to X-and-Y-axes with
differing strength and usually lower quality.
From my testing, anything other than the autohinter on "slight" and the
native CFF engine on "medium" or "full" will lead to inconsistent
rendering and distortion. I recommend "slight" because the autohinter
does a good and consistent job regardless of font format, the maintainer
is also adding support for more scripts. Ubuntu has been using it for
years.
The future will soon bring high-DPI screens for the
masses, and that's where Y-only-fitting as done by the slight autohinter
and the CFF engine shines. Glyphs are rendered *much* cleaner and
inter-glyph-spacing remains untouched. A lot of native hinting done on
TrueType fonts, especially the X-and-Y variant, was made with low-DPI
screens in mind and can look bad when applied on high-DPI screens.
https://bugzilla.gnome.org/show_bug.cgi?id=756204
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The button-power, button-suspend, etc. configurations accept a
GsdPowerActionType, but not all of the variants are supported:
case GSD_POWER_ACTION_BLANK:
case GSD_POWER_ACTION_LOGOUT:
case GSD_POWER_ACTION_NOTHING:
/* these actions cannot be handled by media-keys and
* are not used in this context */
break;
Given that we usually prefer to avoid this sort of configuration which should
work out of the box, it would be better to have them act as they are labelled,
and people who want to change the configurations at system level can use udev to
apply a new keymap.
https://bugzilla.gnome.org/show_bug.cgi?id=753713
|
|
|
|
| |
Resolves: https://bugzilla.gnome.org/show_bug.cgi?id=745181
|
|
|
|
|
|
| |
It's not used anymore.
https://bugzilla.gnome.org/show_bug.cgi?id=749436
|
|
|
|
|
|
|
|
| |
On sane hardware it is mainly unneeded, and the UI on g-c-c is
discoverable enough for the cases where it isn't, so let's avoid
nagging about something mostly superfluous.
https://bugzilla.gnome.org/show_bug.cgi?id=748659
|
|
|
|
|
|
|
| |
Bug 712775 adds this logic to Mutter for both wayland and X11, so remove
the now redundant handling from here.
https://bugzilla.gnome.org/show_bug.cgi?id=744343
|
|
|
|
|
|
|
|
|
|
|
|
| |
The deprecated keys are kept in isolated ".deprecated" suffixed schemas.
On plugin startup, the settings-daemon paths are opened with these legacy
schemas, and all user-modified keys are dumped to the new location and
reset.
This ensures the transition just happens once per-key, and ensures the
dconf database is left clean.
https://bugzilla.gnome.org/show_bug.cgi?id=742593
|
|
|
|
|
|
|
| |
This is part now of org.gnome.desktop.peripherals.[tablet|touchscreen],
in use now by g-s-d.
https://bugzilla.gnome.org/show_bug.cgi?id=742593
|
|
|
|
|
|
|
|
| |
Some trackballs don't have a separate scroll wheel, so one of the
buttons needs to be repurposed so pressing it will turn the main ball
into a wheel.
https://bugzilla.gnome.org/show_bug.cgi?id=645666
|
|
|
|
|
|
|
|
| |
All of this functionality is moving into mutter/gnome-shell so that
both wayland and X sessions follow the same code paths as closely as
possible.
https://bugzilla.gnome.org/show_bug.cgi?id=736436
|
| |
|
|
|
|
|
|
|
| |
Despite the name, the connection ID is a human readable string
and is not guaranteed to be unique.
https://bugzilla.gnome.org/show_bug.cgi?id=732218
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This D-Bus service tracks the current primary mean of connecting to the
Internet, or the network (NetworkManager's "primary-connection"
property). Properties exported are:
- s CurrentNetwork, the NetworkManager unique identifier for the
connection used. This name will be in the output of "nmcli c"
If empty, we're offline
- s CarrierType, the connection (and therefore media) used to access the
network. It corresponds to the TYPE column in the output of "nmcli c"
- s SharingStatus, the current status of sharing. It is one of:
- offline, there's no network so no sharing possible
- available, there is a network connection, so we could be sharing
something
- disabled-mobile-broadband, sharing is not possible because the main
network access is through a mobile device
- disabled-low-security, the network used is too insecure to offer
sharing, such as unencrypted Wi-Fi
Services supported are one of:
- rygel
- vino-server
- gnome-user-share-webdav
It offers 3 methods, one to enable (EnableService) services on the
current network, one to disable services (DisableService) on particular
networks, and one to list the networks on which each service is
enabled (ListNetworks).
https://bugzilla.gnome.org/show_bug.cgi?id=731726
|
|
|
|
| |
This is bound by default to XF86Tools
|
|
|
|
|
| |
The functionality now lives in gnome-software that has its own session
service for update polling.
|
|
|
|
|
|
| |
This reverts commit 11c0be5410caf47adaa968fb852f4a1f62210bde.
Wrong keybinding.
|
|
|
|
|
|
|
|
| |
Super+S conflicts with the "panel main menu" shortcut.
See https://bugzilla.gnome.org/show_bug.cgi?id=720994
This reverts commit f20df9907933d24519a5e69d565cf25715349fda.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
We merge the remote display object into XSettings, so that
we can override the enable-animations settings without modifying
GSettings. This was especially problematic on startup, or when
overriding user settings.
UNTESTED
https://bugzilla.gnome.org/show_bug.cgi?id=694692
|
|
|
|
|
|
|
| |
The percentage and time levels to use for low battery, critically
low battery and when to take action is now handled within UPower.
https://bugzilla.gnome.org/show_bug.cgi?id=709736
|
|
|
|
|
|
|
|
|
| |
The latest update is from the Dell website and dates back from
5 years ago (2009), and the recalls 8 years ago (2006) by the time
GNOME 3.12 is released. Even airlines don't check for exploding
batteries anymore.
https://bugzilla.gnome.org/show_bug.cgi?id=709782
|
|
|
|
| |
No new translations in GNOME 3.10 without a string freeze break request.
|
|
|
|
|
|
|
| |
Don't hardcode an arbitary 30 seconds limit for the maximum screencast
length but read the value out of gsettings (which defaults to 30 seconds).
https://bugzilla.gnome.org/show_bug.cgi?id=708660
|
|
|
|
|
|
| |
As we do not have a UI for this configuration, and that people
are used to handling it through xsetwacom, remove the setting,
and let people run xsetwacom instead of trampling over their choice.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For now, this takes care of automatic timezone updating based on
geolocation. It relies on the new GeoClue2 D-Bus service for getting the
location information and updates the system TZ using the
org.freedesktop.timedate1 service.
In the future, it is also going to take care of displaying notifications
for upcoming time related changes ("Daylight Saving Starts Tonight") and
notifying the user about timezone changes.
https://bugzilla.gnome.org/show_bug.cgi?id=706979
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=705929
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=704435
|
|
|
|
|
|
| |
Based on a patch by Jasper St. Pierre <jstpierre@mecheye.net>
https://bugzilla.gnome.org/show_bug.cgi?id=703410
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
In particular, we can't get the locale value out of GSettings
in order to set LC_PAPER etc, since calling into GSettings
initializes the dconf backend which in turn uses gdbus, which
starts a worker thread.
As a simple workaround, set up the locale environment in
a small wrapper script that then exec's the g-s-d binary.
https://bugzilla.gnome.org/show_bug.cgi?id=701322
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=701089
|
|
|
|
|
|
| |
Mutter and gnome-shell will handle that task going forward.
https://bugzilla.gnome.org/show_bug.cgi?id=700349
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=671072
|
|
|
|
|
|
|
|
| |
When the screen resolution changes and the last known calibrated
resolution doesn't match the new one, a notification is shown
with an action to trigger the calibration immediately.
https://bugzilla.gnome.org/show_bug.cgi?id=677095
|
|
|
|
|
|
|
| |
We're generally trying to move all desktop-wide keybindings to use
Super.
https://bugzilla.gnome.org/show_bug.cgi?id=667327
|
|
|
|
|
|
| |
For both vertical and horizontal scrolls.
https://bugzilla.gnome.org/show_bug.cgi?id=651134
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=695579
|
|
|
|
|
|
|
|
| |
Move the default priority to the plugin files. The GSettings
is kept as a way to tweak the default value.
A few priorities were close to the default 100 priority, so those
were changed to use 100 as the priority.
|
| |
|
|
|
|
|
| |
Instead of checking min/max values by hand, tell GSettings
the range of values we allow.
|
|
|
|
|
|
|
| |
The background drawing has now moved to gnome-shell, allowing
smoother integration, animated backgrounds, etc.
https://bugzilla.gnome.org/show_bug.cgi?id=686549
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is very commonly requested in any situation with
shared resources: servers, labs, etc.
We either log out, or force the logout when the maximum idle time
has been detected.
The feature is controlled by the sleep-inactive-*-timeout and
sleep-inactive-*-type keys.
http://bugzilla.gnome.org/show_bug.cgi?id=597030
|
|
|
|
|
|
|
| |
And make the default time for the dimming be based upon the idle timeout.
If there is no idle timeout, dim at one minute of idle.
If there is an idle timeout, ensure that the dim time is at least 10
seconds, otherwise a third of the idle-delay.
|
|
|
|
|
| |
We have a single setting in the power panel, so it doesn't make sense
to have two separate ones here.
|
|
|
|
|
|
|
|
|
|
|
|
| |
The blanking will only work as designed if:
session.idle-delay ==
power.sleep-display-ac ==
power.sleep-display-battery
So nuke the sleep-display* configuration keys, and rely solely
on the idle-delay to blank the screen.
https://bugzilla.gnome.org/show_bug.cgi?id=691002
|
|
|
|
|
|
|
|
|
|
| |
The purpose of that OSD window is to give users a simple way
to display the current pad button functions.
The on-screen buttons also show when a pad button is pressed
so that users can visually check the matching button/function.
https://bugzilla.gnome.org/show_bug.cgi?id=679062
|
|
|
|
|
|
|
|
| |
These are now provided by gnome-shell.
This reverts commit fc58c347900f906e304591135ca6c13d9950ab4d.
https://bugzilla.gnome.org/show_bug.cgi?id=690106
|
|
|
|
|
|
|
|
| |
Disable animations when on slow VNC or spice remote displays.
Currently only monitors if the display can be exported via VNC.
https://bugzilla.gnome.org/show_bug.cgi?id=680195
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=689447
|