- 05 Jan, 2021 1 commit
-
-
Jonathan Riddell authored
GIT_SILENT
-
- 02 Jan, 2021 2 commits
-
-
Script Kiddy authored
In case of conflict in i18n, keep the version of the branch "ours" To resolve a particular conflict, "git checkout --ours path/to/file.desktop"
-
Script Kiddy authored
-
- 08 Dec, 2020 1 commit
-
-
Script Kiddy authored
In case of conflict in i18n, keep the version of the branch "ours" To resolve a particular conflict, "git checkout --ours path/to/file.desktop"
-
- 01 Dec, 2020 1 commit
-
-
Jonathan Riddell authored
GIT_SILENT
-
- 10 Nov, 2020 3 commits
-
-
Jonathan Riddell authored
GIT_SILENT
-
Script Kiddy authored
In case of conflict in i18n, keep the version of the branch "ours" To resolve a particular conflict, "git checkout --ours path/to/file.desktop"
-
Script Kiddy authored
-
- 09 Nov, 2020 1 commit
-
-
David Redondo authored
BUG:427204 (cherry picked from commit 57374fe5)
-
- 01 Nov, 2020 1 commit
-
-
Script Kiddy authored
-
- 29 Oct, 2020 2 commits
-
-
David Redondo authored
In instantiatePlugins the plugins are ordered according to the configured libraries paths of the environment. When we load the plugins we are currently doing the opposite of what we want. We load first the plugin from the directory with the highest priority, some time later we load the same plugin from a folder with a lower priority and then use the containers from this plugin and throw the existing containers away! (cherry picked from commit ee18830a)
-
David Redondo authored
We do not need the other ones and I had already removed it on other places of this file but overlooked it here. (cherry picked from commit 2abddc2d)
-
- 27 Oct, 2020 1 commit
-
-
Jonathan Riddell authored
GIT_SILENT
-
- 20 Oct, 2020 1 commit
-
-
Jonathan Riddell authored
GIT_SILENT
-
- 16 Oct, 2020 1 commit
-
-
Arjen Hiemstra authored
We update the network speed every 2s to match the daemon update rate. However, this means that when calculating the difference between the current and the previous, we get twice the amount since we are a calculating the difference over 2s. So divide by 2 to get the byte rate per second. (cherry picked from commit cf19c92b)
-
- 08 Oct, 2020 1 commit
-
-
Jonathan Riddell authored
GIT_SILENT
-
- 01 Oct, 2020 1 commit
-
-
Arjen Hiemstra authored
They were apparently missing units.
-
- 24 Sep, 2020 2 commits
-
-
Script Kiddy authored
In case of conflict in i18n, keep the version of the branch "ours" To resolve a particular conflict, "git checkout --ours path/to/file.desktop"
-
Script Kiddy authored
-
- 19 Sep, 2020 2 commits
-
-
Script Kiddy authored
In case of conflict in i18n, keep the version of the branch "ours" To resolve a particular conflict, "git checkout --ours path/to/file.desktop"
-
Script Kiddy authored
-
- 18 Sep, 2020 1 commit
-
-
Script Kiddy authored
-
- 17 Sep, 2020 4 commits
-
-
Bhushan Shah authored
GIT_SILENT
-
Bhushan Shah authored
GIT_SILENT
-
Bhushan Shah authored
GIT_SILENT
-
Bhushan Shah authored
GIT_SILENT
-
- 16 Sep, 2020 1 commit
-
-
Laurent Montel authored
-
- 15 Sep, 2020 1 commit
-
-
Arjen Hiemstra authored
ksystemstats expects sensors to consist of at least three parts: subsystem, object, property. Sensors like "system/processors" break this assumption, causing them to not be properly readable. Rather than trying to come up with an elaborate way to handle this in ksystemstats, simply expose them also as a three-part sensor so the mapping doesn't need to change. These sensors are not very dynamic anyway so duplicating them does not do much harm. The two-part version is kept around for backward compatibility with the ksysguard UI.
-
- 03 Sep, 2020 6 commits
-
-
David Redondo authored
Clients and plugins are both children of the daemon. The Client destructor dereferences SensorProperties to unsubscribe from them. Because usually the object tree looks something like plugin -> sensorcontainer -> sensorobject -> sensorproperty we can not rely on the properties still being around when the clients are deleted. This fixes the segfault in the unit test detected by the address sanitizer.
-
Adriaan de Groot authored
The label can be edited in the dialog, but this information wasn't being set again on the footers in the graph display. BUG: 373935
-
Adriaan de Groot authored
-
Adriaan de Groot authored
In the table of sensors (for bar charts) show the label in column 1 (second, right after host) rather than hiding it at the end; this makes it easier to distinguish the sensors which are **typically** all "localhost, System/CPU Lo.../, %" and then the UI is cut off (scroll right to see the remaining columns, including label).
-
Adriaan de Groot authored
-
David Redondo authored
The returned map also contains infos about containers and objects.
-
- 01 Sep, 2020 1 commit
-
-
Jonathan Riddell authored
GIT_SILENT
-
- 31 Aug, 2020 1 commit
-
-
David Redondo authored
Saves parsing all these files in /proc/net. For now the old code path is retained as fallback when there is some error communicating with the kernel.
-
- 14 Aug, 2020 1 commit
-
-
Arjen Hiemstra authored
Unconnected devices do not have any useful data we can expose, so simply hide them completely instead of showing devices that do not provide useful data.
-
- 13 Aug, 2020 1 commit
-
-
David Redondo authored
The return value of compare is 0 if both strings match https://en.cppreference.com/w/cpp/string/basic_string/compare
-
- 03 Aug, 2020 1 commit
-
-
David Edmundson authored
If ksgrd changes the monitors then all IDs are invalid. This can happen when a new device is plugged in and removed. SensorObjects can remain valid and are re-matched up.
-
- 31 Jul, 2020 1 commit
-
-
David Redondo authored
-