T26694#186985 states that the statistics nodes are not named properly. Since no access to the node name was given, the identifier was changed to be the UID of the image and mask.
Now that we have a multi label histogram the labels should be named properly in order to identify the correct data.
Description
Related Objects
- Mentioned Here
- T26694: [image statistics plugin] Add multithreading
Event Timeline
I still don't see why it is necessary! The arguement given in T26694 holds also true here. For the identifcation under the hood, you don't need any names or labeles.
Or I did not understand the problem you are refering to.
It's not about identification under the hood but the name of the data in the histogram, i.e. the label of the legend. But it's probably low priority for now.
OK. That makes sense, but I am not sure if i would solve it be the node name or if it doen't makes more sense to offer/implement a function that can be used by e.g. the histogramm to deduce a name from the statistics...
As you describe it, it is realy tight coupled with the UI its freedoms and what makes sense/is needed to make the charts indentifiable. It wouldn't be helpfull e.g. to encode everythin into long names to be unique and therefore unnecessary clutter the legend of the chart.