[Question] Use run name as axis labels instead of time or alias

[ngblume]

hej hej,

more of a question than an issue..
I use the Listener integration and its works great.
But the only options I have for the labels on the x axis are timestamp or alias.
Why not run name?
Can I activate this as an option in the setting ? (didn't find it..)

The alias for me is set to "output" when using the listener integration, so all runs have the same alias.
I also looked at https://marketsquare.github.io/robotframework-dashboard/advanced-cli-examples.html#aliases-for-clean-dashboard-identification..

If no meaningful alias can be generated (e.g., all files are simply output.xml), the dashboard will generate fallback aliases such as Alias 1, Alias 2, etc.

Why not Run Name ?

Thanks!

Cheers
Niels Göran

[timdegroot1996]

Hi @ngblume, awesome to see you're looking into customizing your dashboard!

I myself came up with showing "run_starts" as that was something unique per run, and made for an easy default.
For now the request I've gotten was to allow "custom" naming through aliases. Which you found on the readme above. It could be something that we can add another setting that is similar to the existing Display Run Start/Alias Labels On Axes, but then we could make it a dropdown in which you can select Run Start, Run Name or Alias. Where run name will just be the actual project name.

Example: if you look at my example dashboard here: https://marketsquare.github.io/robotframework-dashboard/example/robot_dashboard.html it would result in:

• Tests 1 till Tests 10
• Testsuites 1 till Testsuites 3
• Other 1 and Other 2

I do still feel that this could also be achieved through the alias feature like this for example:

But it would require running your tests with specific output file names to match the project name you would want.
See also the result in the actual dashboard, you have to toggle Display Run Start/Alias Labels On Axes, to show the version in the image.

robot_dashboard.html

So the way I see it we have 2 options:

1. Extend the toggle to have run name also as a default option
2. You make use of the alias functionality to achieve the naming you want to have

Let me know what you think!

[ngblume]

Hej @timdegroot1996,

thanks for the fast response!
I am a little confused as to how what information must be set to make it work...

I agree with you, timestamps are usually a good indication...
In my usecase, though, we run tests on three different systems as part of a pipeline at the same time, so that time is rather identical...
That's why I favoured the run name...
It ouwld theregore be great, if the option "run name" could be added to the settings as part of the drop-down...

In the mean time, I am a little lost, how I would achive something similar with the alias...
I do not process the results in dividually,but use the listener (see other issue)...
Is it than possible to do that and maybe a pointer, as to how, if it is possible?

Thanks!

Cheers
Niels Göran

[timdegroot1996]

Hi @ngblume,

I get your confusion. What I proposed above were 2 possibilities:

1. I make the run_name an option that is available in the settings, then you just have to wait for me to build it, and you can simply change a setting in that case.
2. You can use aliases, which are set through the filename of the output.xml files. Below an example, but this might be a bit more complex, but then you can solve what you want to achieve right away.

I'm not entirely sure how you run your tests, but I would assume it's in a pipeline of sorts. In that pipeline you would have to update your commands to actually change the filenames of the log/outputs manually, so they are unique, and alias "ready".

Here an example for a bash pipeline script, when running with app1, app2, app3:

timestamp=$(date +%s)
robot --output app1-$(timestamp).xml --log app1-$(timestamp).html --listener robotdashboardlistener.py
robot --output app2-$(timestamp).xml --log app2-$(timestamp).html --listener robotdashboardlistener.py
robot --output app3-$(timestamp).xml --log app3-$(timestamp).html --listener robotdashboardlistener.py

This should result in unique AND matching output/log file names which will still be picked up by the listener correctly.
The aliases will then also be app1-2026-04-13-11:23:00 or some other timestamp, but you get the idea. Same for the other files.

EDIT: I did not try this bash script, I just quickly searched for some bash command to get a timestamp, be sure to double check what you require/want for your runs 😉

[ngblume]

I implemented it along those lines...
What I am not sure about yet, is, wether I can still use "log" or "output" to the filepath, since otherwise, the only difference would be the extension...

Thanks !

Cheers
Niels Göran

[timdegroot1996]

@ngblume, output/log are not required in the name. They can be added if you want but it can match based on the extension if the remaining parts match exactly. Just like you implemented it.

[timdegroot1996]

@ngblume I played around a bit with the idea I had above to make run name also an option instead of run_start and alias which already exist. You can find it in the settings under Run Label Display.

Let me know what you think!

robot_dashboard.html

[ngblume]

Hej hej,

will give it a try..
I noticed, that I would also need to change my run-name (set via argument files) before that options makes sense for me across multiple nightly runs... As soon as I got that set, I will give it a try...

Thanks!

Cheers
Niels Göran

[ngblume]

One follow-up question regarding the "Customoize Layout" option....
Is there a way to set options like "Only last Run" persistently like the layout itself?

[timdegroot1996]

@ngblume Currently most toggles to "steer" the graphs are not saved. This is the same for the toggle/dropdown in the test statistics.

The only thing that is saved is the "Section" filters. This is something that could be added if you really want it. But please make a new issue for that! 👍