Is there any tutorial?

[philippludwig]

So I have been running arbtt-capture for two weeks now, and I see that it has recorded lots of data via arbtt-stats --dump-samples; but I still don’t understand how to do anything useful with the data without writing what feels like 1000 lines of configuration.
I have read through the user guide, but still don’t know where to start. At the moment, arbtt-stats just outputs:

Processing data [==================================================================================] 100%
Total time per tag
==================
Tag_|_Time_|_Percentage_

My configuration is:

{$idle > 60 ==> tag inactive}

---

My main issue is that the configuration example is not really helpful, at least for me. It is huge and I don’t understand at the moment what I need of that and what I don’t. Is there may be a step-by-step guide for creating a config, showing how you can build something from the bottom up? For example, just to get started by adding a few lines that show how much time I spent in Firefox would be helpful to understand how this actually works.

[nomeata]

I don’t think we have much more than the users’s guide. Did you have a look at this section?
https://arbtt.nomeata.de/doc/users_guide/effective-use.html

[philippludwig]

Yes, and I don’t get it I am afraid. When I try this command:

arbtt-stats --categorizefile=foo --m=0 --filter='$sampleage <24:00' --dump-samples

The output looks a lot different from mine; that’s where I am unsure to proceed/what to do.

---

Anyway, I am closing this now. Thanks for your time, I don’t want to waste it further, just hoped for a tutorial or something.

[waldyrious]

I'd say that starting off with a more basic example in http://arbtt.nomeata.de/doc/users_guide/configuration.html would be useful indeed. Not necessarily a tutorial, but just a selection of a handful of the rules from the larger example, adapted to be more generic and applicable to a broader range of people. In fact, I would even propose removing the full example from the docs, and instead link to the version in the source tree as an example of a real-world, personalized configuration file.

[nomeata]

I agree, that example is not really that useful (and probably not even very “real-world” either).

@philippludwig, I wonder if there is something else wrong in your case. What does arbtt-stats --info say?

[philippludwig]

Here you go:

Processing data [===================================================================================] 100%
General Information
===================
                       FirstRecord | 2023-03-10 17:51:47.045115822 UTC
                        LastRecord | 2023-03-19 08:51:23.572679778 UTC
                 Number of records |                              4900
               Total time recorded |                       3d09h40m00s
               Total time selected |                       2d09h00m00s
   Fraction of total time recorded |                               39%
   Fraction of total time selected |                               28%
Fraction of recorded time selected |                               70%

[nomeata]

I would suggest you start with this configuration

-- Simple rule that just tags the current program
$idle > 60 ==> tag inactive,
tag Program:$current.program,

so that you get a categorization based on the current program. And then, as you want more or different analyses, add more rules.