Currently, I'm unable to test on MacOS, so I'm not sure how well this will work, if at all. I'll try to source MacOS hardware to test this application.
* ``-h``, ``--help`` to show the help screen and exit (basically has all of the below CLI option info).
* ``-a``, ``--avgcpu`` enables showing the average CPU usage on rustop
* ``-c``, ``--celsius`` displays the temperature type in Celsius. This is the default.
* ``-f``, ``--fahrenheit`` displays the temperature type in Fahrenheit. This is the default.
* ``-k``, ``--kelvin`` displays the temperature type in Kelvin. This is the default.
* ``-v``, ``--version`` displays the version number and exits.
* ``-r <RATE>``, ``--rate <RATE>`` will set the refresh rate in *milliseconds*. Pick a range from 250ms to ``UINT_MAX``. Defaults to 1000ms, and higher values may take more resources due to more frequent polling of data, and may be less accurate in some circumstances.
* ``Up/k``, ``Down/j``, ``Left/h``, ``Right/l`` to navigate between panels. This currently doesn't have much functionality but will change in the future.
* ``dd`` to kill the selected process (currently only on Linux) - **I would highly recommend you to be careful using this, lest you accidentally kill the wrong process**.
* ``c`` to sort by CPU usage. Sorts in descending order by default. Press again to reverse sorting order.
* ``m`` to sort by memory usage. Sorts in descending order by default. Press again to reverse sorting order.
* ``p`` to sort by PID. Sorts in ascending order by default. Press again to reverse sorting order.
* ``n`` to sort by process name. Sorts in ascending order by default. Press again to reverse sorting order.
[* Scrolling either scrolls through the list if the panel is a table (Temperature, Disks, Processes), or zooms in and out if it is a chart.]: <>
* Scrolling currently only scrolls through the list if you are on the Processes panel. This will change in the future.
## Regarding Process Use Percentage (on Linux)
I cannot guarantee whether they are 100% accurate. I'm using a technique I found online which seems to be a better indicator of process use percentage at the current time, rather than pulling from, say, ``ps`` (which is average CPU usage over the *entire lifespan* of the process). I found the options from the library I used to get other data (heim) to be a bit too inaccurate in my testing.
For reference, the formula I am using to calculate CPU process usage is along the lines of:
```rust
let idle = idle + iowait;
let non_idle = user + nice + system + irq + softirq + steal + guest;
let total = idle + non_idle;
let prev_total = prev_idle + prev_non_idle; // Both of these values are calculated using the same formula from the previous polling
let total_delta : f64 = total - prev_total;
let idle_delta : f64 = idle - prev_idle;
let final_delta : f64 = total_delta - idle_delta;
//...
// Get utime and stime from reading /proc/<PID>/stat
let after_proc_val = utime + stime;
(after_proc_val - before_proc_val) / cpu_usage * 100_f64; //This gives your use percentage. before_proc_val comes from the previous polling
```
Any suggestions on more accurate data is welcome! Note all other fields should be accurate.
* [tui-rs](https://github.com/fdehau/tui-rs) (note I used a fork due to some issues I faced, you can find that [here](https://github.com/ClementTsang/tui-rs))