This commit creates the possibility that not only kernel-led-triggers can
be selected but also application-led-triggers from user space.
This is done via a plugin mechanism. The application-led-triggers are scripts
that set kernel-led-triggers on system events or services. Until now this
has not been possible. The package rssileds is a kind of
application-led-trigger.
The following new packages are added:
* luci-app-ledtrig-rssi (application-led-trigger)
* luci-app-ledtrig-switch (kernel-led-trigger) not needed on every most devices
* luci-app-ledtrig-usport (kernel-led-trigger) optional trigger
Since we have now a plugin mechanism I have added the following triggers
as a dependency. So this triggers are now installed per default on LuCI
installation.
* kmod-ledtrig-default-on
* kmod-ledtrig-heartbeat
* kmod-ledtrig-netdev
* kmod-ledtrig-timer
The kernel trigger kmod-ledtrig-usbdev was removed with the commit
d0b50c2770
So I have not ported the relevant code anymore.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
Remove the abbreviation coding from the menu item, as it does nor work
in menu and seems to prevent translation of the menu item.
(This was the only menu item in LuCI that had abbr logic.)
Reference to #3553
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
Currently LuCI can be loaded only when placed in the root of the server as the cgi-bin path are hardcoded. Change the index.html to load the cgi-bin path from the current level.
Also add a new entry in the env Object to make the cgi_base path easily accessible.
This variable will be based on the position of /cgi-bin/luci.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
While ui.js is implicitely autoloaded by other classes, we need to require
it directly in the view to avoid race conditions during rendering.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The firstboot rpc call might not return, therefor do not wait on promise
fullfillment but display modal dialog right away.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Since the modal overlay map was unable to read the current trigger value,
it mistakingly allowed the removal of the device option value.
Fix it by finding the option object through lookupOption() instead of
relying on a reference.
Fixes: #3216
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Instead of binding the checked state to the button handler function,
bind the underlying checkboxes instead to observe the actual user
choice instead of the initial value.
This fixes forcing sysupgrade or deselecting keep settings.
Reported-by: Hannu Nyman <hannu.nyman@iki.fi>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Since the advanced-reboot app is the sole remaining user of the applyreboot
template, import it into this package to avoid the need for shipping it
with the default installation.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The `/sys/devices/virtual/mtd/` information is not available on all
devices so rely on the contents of `/proc/mtd` instead to check for
the existence of the `rootfs_data` partition.
Fixes: #3136
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Some firmware images may not support preserving backup. In such cases
display a warning and disable relevant checkbox.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Some validation errors may be critical enough to prevent sysupgrade.
Check the "forceable" property and disallow forcing sysupgrade if
applicable. It would fail anyway at the "sysupgrade" call.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This new ubus method provides more properly-formatted details about
firmware file. Use it to check if uploaded image is valid.
The old "sysupgrade --test" method is left for now to provide stderr
output.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
[Added missing ACL, add default value for the ubus failure case]
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
