PointOnPointOverlayer stores duplicates in list, how to read from Pythoncaller?

An FME List is more like a Python Dictionary, in Python a list is a group of values, where as in FME a list is a group of keys and their values.

In your first line, that will evaluate to ‘None’ so will never pass the If clause.

To get a list Attr, you need to do something like:

foo = feature.getAttribute('__list{}.attr1')

bar = feature.getAttribute('__list{}.attr2')

for f, b in zip(foo, bar):
    print(f, b)

You can then iterate over them both at the same time using zip - only works if lists are the same length

Thanks for the response!

I managed to solve the problem, but to me it doesn’t seem like the attributes that I access with getAttribute are typed as an ordinary Python lis/dictionaryt. It’s more like the list and dictonary is converted to pure strings when it enters the PyhonCaller. That’s what confused me when I read different examples on the internet. When I read 'dupl_attr{0}.Fritext' I thought that was a list when it just was a simple string. 

The only way for me to solve the problem was to loop the number of entrys in the “list”. A number that I got through another integer value in the feature. Then create the feature name I wanted to access by creating a string which at first glance, looks like a list reference, but in fact is just a plain string. 

Example: 

        altValues = []
        overlaps = feature.getAttribute('_overlaps')
        for x in range(overlaps):
            itemValues =  {}
            for item in self.featAttr:
                itemValues.update({item:feature.getAttribute('dupl_attr{' + str(x) + '}.' + item)})
            altValues.append(itemValues)

Where self.featAttr is a list created inside Python caller with names of the feature attributesI want to check. For example: ['Fritext','MarkTyp','MatOrg']

Second last row in the example creates a string that can look like this: 'dupl_attr{0}.Fritext' which is not a list type but just a string that looks like a list :)

Maybe there are other solutions but I wanted to solve this in a Pythoncaller instead of an ordinary transformer.