My transpose is easy but cannot figure out how to do it in FME. I checked this article already
Transpose a Table Using FME (safe.com)
I have my data with unique rows, but I want to create one row per column type instead.
From my data:
TO result
I tried the custom Attribute Transposer from the article but get missing values and columns
Use 2 AttributeManagers in parallel.
Use 2 AttributeManagers in parallel.
Hello, this is good but in reality I have over 80 columns and don't want to create 80 parallel transformers I want the processing to be automatic without getting my hands on each column. My attribute managers are already very slow and a pain. I did however solve it, I used attribute explode like in the guide, and with this example basically sorted out all attribute.name !="Price Class" with regular expression. Then renamed attribute.name= Class, and attribute.value=Value. I can upload part of script if anyone is interested.
I had 686 features going in and after the explode I had
around 70k rows, and after some clearing and removing empty values I ended up with around 30k features which was expected by the customer that it would end up on that amount since its one row per class for each feature ID. So if you open in excel for example you can now sort for all values for class B and which features have a value for class B with their ID.
Use 2 AttributeManagers in parallel.
Hi, good to see that you have found a solution that is better suited for your data.
80 parallel AttributeManagers doesn't sound appealing at all!