Hi all,

I’m working on an FME script and I’ve hit a wall I can’t seem to get over.

Basically, I’m using a custom transformer (PointClusterer) to cluster a bunch of points by proximity. The number of clusters I want is however many it takes to have groups of 300 plus remainder. So if I have 1500 points, I have five groups, 1501 points, I have six, 1499 points, I have five etc.

The problem is that the PointClusterer transformer wants the number of clusters, not the cluster divider, and I want this field to be filled in dynamically based on the characteristics of the data I'm loading in.

Theoretically, this shouldn’t be a problem. There are a bunch of different ways to get FME to calculate the number of clusters I want based on the total feature count I load in. Here’s an example of a really basic way I cooked up. I take the total count of features, divide that number by 300, round down, then add 1 and write that number to a field as an integer.

Unfortunately, when I connect this feature to the PointClusterer transformer and tell it to use my ClusterValue field, I get this error:

I suspect this is because by selecting that attribute for that parameter I'm feeding in each row of the feature instead of just a single value.

I've run into situations in the past where I've needed to use just a single value from a feature as a parameter (the min, the max, sum, whatever, but I've never been able to figure out how and I've always had to come up with a clumsy work around. My question is: is there a way to take the value I calculated by doing operations on the total feature count, and use it to fill in a transformer parameter that is just looking for a single value?

Thank you for your time.