Hi Florian,

> On Jun 16, 2020, at 14:18, Florian Eckert <[email protected]> wrote:
> 
> Hi Adrian,
> 
> thanks for your comment
> 
>>> -           "0xd00")        s="resync" ;;
>>> -           "0x3c0")        s="short init entry" ;;
>>> -           "")             s="not running daemon"; ls="0xfff" ;;
>>> -           *)              s="unknown" ;;
>>> +           "0x0")          s="not initialized"
>>> +                           n=1 ;;
>> Wouldn't it be more user-friendly to just use the decimal number
>> equivalent of the hex code?
> 
> I've been thinking the same thing.
> But then I have a problem when I want to display the values over time 
> (collectd/gravana).
> From my point of view the distances are then too small or too large it does 
> not give equal gradations.

        I guess lantiq (or rather maxlinear nowadays) might introduce new hex 
codes into existing gaps which will cause a distruption of the nice monotonic 
order oc the enumerating decimals you went for. Is there no other data type in 
graphana that is ordered yet not scaled?

Best Regards
        Sebastian


> 
>>> -           "0x1000000")    s="test" ;;
> 
> Especially the test value jumps out of line there. I think this is unlikely 
> and could be neglected,
> but I would like to have a linear gradation.
> 
> 
>>> -           "")             s="not running daemon"; ls="0xfff" ;;
>>> -           *)              s="unknown" ;;
>> Empty and error could be modelled with negative numbers then.
> 
> We can still do that. So we have for "" and * the value -1 and -2.
> 
> Best regards
> 
> Florian
> 
> _______________________________________________
> openwrt-devel mailing list
> [email protected]
> https://lists.openwrt.org/mailman/listinfo/openwrt-devel


_______________________________________________
openwrt-devel mailing list
[email protected]
https://lists.openwrt.org/mailman/listinfo/openwrt-devel

Reply via email to