Facility and severity filtering are based on a simple includes() search.
As such, false positives are possible. Although for the majority of
cases, this is still useful. Filtering using logread -z/Z is possible,
but not if a static log-file is configured. logread does not yet handle
severity either. 'not' checkboxes for each invert the respective search
filter.
A raw-text filter is also included as a bonus, whose meaning can be
inverted via the 'not' checkbox.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 16beffda15)
Simplifies adding third-party repos. Functions on both 24.10 (opkg) and
main (apk). Based loosely on sshkey management.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit acdffc3d1d)
luci-mod-system: Add repo key management
follow-up fix to acdffc3d1d
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit cf1e61f10d)
luci-mod-system: repo key management
Reject PEM in OPKG; reject non-PEM in APK
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 65bdfc61db)
This fixes the missing policy option in UCI when selecting 'active-backup'
as the required policy.
This has been discussed on issue #7683.
Signed-off-by: Nicolò Veronese <nicveronese@gmail.com>
(cherry picked from commit c996de5a82)
Hardcode 'openwrt-24.10' release branch as the branch name to avoid
'LuCI (HEAD detached at 2ac26e56) branch' being shown as the branch.
Feeds are since March 2025 cloned shallow if cloned from a specific
commit, like done with a release tag. Change in OpenWrt:
32d0a57dc1
As the only cloned commit of a release tag is without history, the branch
can't be identified.
So, hardcode the branch name to get the proper branch name
'LuCI openwrt-24.10 branch'.
Fixes#7737
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
In the underlying uci system, all variables are effectively strings, so
for those configuration values which serve as 'boolean' flags, we need
to coerce the various forms into a real boolean. Only the following
string values result in true: 1|on|true|yes|enabled. Otherwise, false.
Checkboxes now fill correctly.
"Unchanged" configurations may write changed values as the Flag values
are coerced to '1' or '0' on write, but the configuration behaviour
remains synonymous.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 1355a6fa25)
follow-up fix for 0be4ad51a0
It seems to have been this way for a while, for zone selections
involving the 'any' ('*') choice. firewall forwards.js has no
'this.allowany;' property on the widgets, so the any choice is absent,
leading to a traceback, caused by the 'src' widget, even though
this code block operates on the 'dst' value. Certain combinations of
'src' and 'dst' also triggered tracebacks.
This might not be the correct behaviour for this widget, but it does
prevent tracebacks caused by null values.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit aa6924cd4c)
Tunnel devices which are up, show as down, since the config being
enumerated does not have access to any device status. The 'device' being
enumerated is e.g. 'wireguard_wg0', a configuration for a tunnel
interface, whose parent is an 'ethernet' type.
That's a bug for another day. So they shall use parent device status.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit a288354e78)
Tunnel devices which are up, show as down, since they don't have access
to any device status. The 'device' being enumerated is e.g.
'wireguard_wg0', a configuration for a tunnel interface, whose parent
is an 'ethernet' type.
That's a bug for another day. So they shall get parent device status.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 832680bb84)
build was failing b/c gcc15 defaults to c23, which lemon is not written for (and errors with).
Signed-off-by: Andrew Robbins <andrew@robbinsa.me>
(cherry picked from commit 852fc23b12)
build was failing b/c gcc15 defaults to c23, which lemon is not written for (and errors with).
Signed-off-by: Andrew Robbins <andrew@robbinsa.me>
(cherry picked from commit 7e6741fbae)
Problem:
After updating the JS-part of any package, the user is forced to clear
the browser cache.
Fix:
Each time a package is removed or installed, the contents of the
"/lib/apk/db/installed" file are changed.
This fix add the modification time of the "/lib/apk/db/installed" file
to the version of the main LuCI package (for each js-file).
Signed-off-by: Oleg S <remittor@gmail.com>
(cherry picked from commit f6f82b52eb)
For use when daemons whose interface config paradigm accept IP also.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 5be4cd129c)
handleAdd did not work correctly for a while: some state was not cleaned
up (try: wireless -> add -> no changes -> click dismiss -> press save).
The state is now properly removed when pressing dismiss, by stashing
addedSection correctly under the map m and not in this.
Also requires a null check in render_modal_status.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 6c061b644c)
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit cdce600aae)
luci-mod-network: give wireless.js ES2016 treatment and refactor
Follow-up fix for cdce600aae when certain
elements are empty.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 27ebe8b662)
luci-mod-network: give wireless.js ES2016 treatment and refactor
typo
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 0d610e7466)
This patch adds a badge to the associated wireless stations with the vlan id
and name. It is displayed in the same color as the network, to which it is
bridged, so the color corresponds to the color in the network configuration
page.
Signed-off-by: Tobias Waldvogel <tobias.waldvogel@gmail.com>
(cherry picked from commit 29fe3f5fda)
There can be multiple tables with the same name but different family
attribute (inet, ip, ip6, ...). The same goes for chains. So the family
attribute needs to be considered when rendering chains and rules.
Signed-off-by: Stefan Kleeschulte <stefan.kleeschulte@smk.biz>
(cherry picked from commit e79e7022e5)
Browsing the root path was problematic and the breadcrumb links behaved
inconsistently, requiring a page reload to recover.
canonicalizePath(), splitPath() and renderListing() now handle any path
location properly, including root: '/', and clicks at any breadcrumb
correctly navigate.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 5e39cf04c4)
This commit fixes wrong channel width in the 6 GHz band. Now all APs
don't show a 20 MHz channel width.
Tested on ZyXEL EX5601 with MT7921 USB adapter.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
(cherry picked from commit 62c401ca24)
Introduce addTimeLimitedNotification which extends addNotification,
with the addition of the 'timeout' parameter.
timeout: a millisecond value after which the notification will disappear
automatically. If omitted, the notification will remain until it
receives the click event.
Improved implementation of 53e36e3293
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
(cherry picked from commit 0329854049)
Any number of apps read boolean values from configuration files, then use
various inconsistent means for checking truth values. The get_bool function
allows app authors to fetch the value without regard for how it is represented
in the config file.
For example, this
let enabled = uci.get('system', 'ntp', 'enable_server');
if (enabled == '1') ...
could become the more natural
let enabled = uci.get_bool('system', 'ntp', 'enable_server');
if (enabled) ...
Signed-off-by: Eric Fahlgren <ericfahlgren@gmail.com>
(cherry picked from commit dea8d79888)
The function signature changed and all function call to addNotification
that add a CSS class does not work anymore. A direct revert is not
possible.
If this feature is needed than a new function musst be added, that does
not break the function signature.
Fixes: 53e36e3293
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
(cherry picked from commit 3f832f2512)
Sync translations.
Backport from master via merge-master script.
(At this first backport, also the backup strings get removed.)
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
removes a075566dbb (and subsequent fixes)
enforcing a datatype on this widget generates the following error when
multiple zones are selected:
Expecting: valid UCI identifier
(the error triggered for uciname). At its simplest: multiple selections
are no longer just a uciname, but multiple thereof.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
Follow-up fix to fb4cf432d2
Resolves sync errors:
standard input:682: warning: Empty msgid. It is reserved by GNU gettext:
gettext("") returns the header entry with
meta information, not the empty string.
standard input:733: warning: Empty msgid. It is reserved by GNU gettext:
gettext("") returns the header entry with
meta information, not the empty string.
standard input:761: warning: Empty msgid. It is reserved by GNU gettext:
gettext("") returns the header entry with
meta information, not the empty string.
standard input:818: warning: Empty msgid. It is reserved by GNU gettext:
gettext("") returns the header entry with
meta information, not the empty string.
Signed-off-by: Paul Donald <newtwen+github@gmail.com>
Also, the Lan and WiFi tables are half-width, we should not use the table-titles which has @media rules about 992px.
Signed-off-by: Jan Pazdziora <jan.pazdziora@code.adelton.com>
