| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
The signal handler was getting called three times on double-click: once
for the double-click, once for the first click, and once for the second
click. This resulted in the background getting shown as if the disk were
selected, even though everything else indicated it was not.
|
| |
|
| |
|
| |
|
|
|
|
| |
No reason to have this kind of stuff in two different places.
|
|
|
|
|
| |
There's nothing important in here anymore, and we need to condense these lists
into one central place.
|
|
|
|
|
|
| |
This is to draw the user's attention to them a little better. It also allows
me to kill all that CSS provider crud, which it turns out isn't even needed.
This probably changed when I started using the PangoAttrList.
|
|
|
|
|
|
|
|
|
|
| |
Doing it from the init method means we need to call gtk_widget_realize so all
the gdk and cairo stuff has something to operate on. However, that realize
call screws up glade so that the window gets displayed separately and widgets
cannot be clicked upon.
This reorganization moves the gradient setup to after the widget has been
realized.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The idea here is that by adding a gradient up top, the "Done" button looks
more like it belongs with the header than with the widgets down in the
spoke. It also subtly makes the spoke look like a subpart of the hub.
Adding this required adding an additional internal widget, which means also
changing all existing glade files to include this internal widget. Make sure
to build new glade support and install locally (or use the proper makefile
targets) before editing.
|
|
|
|
|
| |
Except on the hub, which is special and is going to need special work done
for it.
|
|
|
|
|
|
| |
Everything in the beginning of anaconda_base_window_retranslate is critical to
making sure we can change translations at runtime. Translations are still not
totally working, but this gets us a lot closer.
|
|
|
|
|
|
| |
Without this check, ENABLE_NLS will not be defined, which means gettext.h will
not do anything, which means strings in C lang will not be translated no matter
how hard we try.
|
|
|
|
|
| |
This saves us from having to type the same GTK constants again and again in
python land.
|
|
|
|
|
| |
This gets rid of some warnings on tty1. I'm surprised no one's filed bugs
about them yet.
|
| |
|
|
|
|
|
|
|
|
| |
This patch adjusts the coloring for selected mountpoints in the
mountpoint selector widget. Previously, some of the text labels appeared
grey or black on the blue selected background, making it harder to read
them. This patch adjusts those so they go white when selected, and black
or gray (depending on the label) when not selected.
|
|
|
|
|
|
|
|
|
| |
Previously, the disk overview widget (visible on the first storage
screen) would display black text on a blue-colored background for
selected items. Attempts to set the foreground color to white using GTK
overrides proved futile, so this patch removes all color overrides and
instead uses GTK state flag setting to achieve the desired affect:
unselected text is black on white; selected text is white on blue.
|
|
|
|
|
|
|
|
|
|
|
| |
The mountpoint selector has a problem where all mountpoints except the
selected one are flush with 0 margin against the right wall of the box
they're in, and the selected mountpoint gets 6 px of margin between it
and the wall.
This patch makes that margin more consistent, so both selected and
unselected mountpoints in the mountpoint selector list on the left hand
side of the custom partitioning spoke get a consistent 12 px margin.
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
This is needed for displaying UTC and GMT-X timezones. Setting
timezone to "" graphically sets "Europe/London", but doesn't add
the pin.
|
|
|
|
|
| |
Without this, translatable="yes" will not appear in the output glade files,
which means the strings will not be in anaconda.pot.
|
| |
|
|
|
|
| |
Resolves: rhbz#869388
|
|
|
|
|
|
|
| |
This is required by the previous patch. We are storing pango markup in the
language store now, and pango does not appear to support nested spans. So,
we need to set attrs instead. Also I've condensed things a bit here to make
sure the code is only in one place.
|
|
|
|
|
|
|
| |
This doesn't fix the redraw-related size problem. I think that might just
be an artifact of not using a compositing manager. This merely attempts to
fix it so we don't display a spoke entirely smashed up in the upper left
corner of the screen.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
By hardcoding a color, it makes the status text almost invisible on the KDE
live disc where the theme is darker to begin with. Instead, use the theme
colors for everything.
This isn't a perfect fix: If you focus a SpokeSelector, the text will still
be black instead of white like everything else. However, I've been unable to
make that work.
|
| |
|
| |
|
|
|
|
|
| |
This also changes the back-signal to button-signal. This is prep work for
allowing individual spokes to relabel the button if they so choose.
|
| |
|
| |
|
|
|
|
| |
This can be handled higher up in the python portion of the UI if desired.
|
|
|
|
|
| |
I discovered this searching for "iw" to make sure I deleted everything related
to the old interface.
|
|
|
|
| |
I knew I'd forgotten something when I did this for MountpointSelectors.
|
|
|
|
|
| |
This is largely just copy & paste from what msivak did elsewhere.
Thanks.
|
|
|
|
|
|
|
|
|
|
|
|
| |
This means that when you press up and down with the keyboard to navigate the
accordion, whatever MS you hit will be the one displayed on the right, and
whatever one you were previously on will have its changes saved.
This is different from how DiskOverviews and SpokeSelectors work, but I think
it makes sense. When you keyboard to a DO, you are choosing to do include it
in the install. We don't want people doing that by accident. When you
keyboard to a SS, it would jump you into the spoke and that wouldn't work
at all. The MS, however, just changes what's displayed on part of the screen.
|
| |
|
|
|
|
| |
Also, don't translate it. That doesn't make any sense.
|
| |
|
|
|
|
|
|
| |
A side effect of this is that the whole left hand side is now wider, but I
think that can be fixed by adjusting the stuff that goes in the right hand
side.
|
|
|
|
|
| |
When testing we, need to set the location of the bitmaps our TimezoneMap is
using, so let it try to get the location from the environment.
|
| |
|
|
|
|
|
| |
This is the thing on the left hand side of the custom partitioning interface,
allowing you to select a mount point and do some configuration on it.
|
|
|
|
|
|
|
| |
Really, I'd like them to be grey only when the spoke is not ready and then
turn black afterwards. But I can't make that work, and it has to be done
in C since the labels aren't exposed outside of that. So this will have
to work for now.
|
|
|
|
|
|
|
| |
Some keyboard layouts have pathologically long names, and we're already
trimming the status for network sources so just do it for all spokes. Note
that this does not change the text of the status, so it can still be used
for string comparisons.
|