Handle changes to the property _NET_WM_STRUT_PARTIAL by reallocating
screen space in the dock area when the dock client requests a new
position.
This will allow changes to the bar config directive `position` to be
effective on the `reload` command.
fixes#1294
This introduces a "separator_symbol" property for the i3bar configuration.
If set, the specified string will be used as a separator instead of a vertical line. Since it is an optional configuration, complete backwards compatibility is given.
fixes#1472
Otherwise, placeholder windows would accumulate these over restarts. The
worst part is that they don’t get dumped properly, meaning an empty
swallows specification ends up in the state, which matches any window,
so other random windows would be matched into the placeholder (e.g.
i3bar).
fixes#1502
This reverts commit cbd51ce664.
If you’ve started using this feature, I’m very sorry, but I think it’s
better to revert it entirely, as it will only cause more problems in the
future as more and more people upgrade and get bitten by this if we
didn’t revert it…
fixes#1536
Add `markup` to the i3bar protocol as a block member.
This is a string that determines how the block should be parsed as
markup. "pango" indicates the block should be parsed as Pango markup.
"none" indicates the block should not be parsed as markup.
When `new_window` is given in the config, config validation with `i3 -C`
would segfault.
Add a NULL check in logical_px() to check for the case when the config
is being validated without an X server to prevent this.
This combines detecting of current and clicked workspaces into one cycle
and then checks if the x coordinate of the clicked point is greater than
the workspace buttons width.
This allows "--workspace 3" for numbered workspaces, e.g., if the name is actually
"3:foo". This introduces the same functionality the IPC already offers in many places.
This draws the statusline after drawing (and calculating width of) the
workspace buttons and fixes calculation of its maximum visible width.
This (hopefully) guarantees that these two will never overlap.
This fixes the bug when statusline pixmap wasn't entirely cleared that
caused random artifacts to appear when the statusline width is greater
than the screen width.
Michael Stapelberg