Hi all, I've stumbled about this puzzling and seemingly simple task: Iterate over a a bundle and print the key-value pairs. Here a simple example: <hansl> bundle B = _(a = 1, b = 2) loop foreach i B printf "key: %s, value: %d\n", "$i", $i # OR: printf "key: $i, value: %d\n", , $i endloop </hansl> That works as all elements in B are numeric. However, I always found this confusing: For the first printf, the key is accessed by means of "$i" as an argument while in the second line one accesses it via $i inside the 'format' part. ALSO NOTE: The parser is "robust" against the two comma at the end of the second printf command. Bug or feature? ;-) But lets augment the bundle and add a third string object. HOW TO, iterate in this case across the bundle _without_ having to check the data type? Is there such a way? The following fails. <hansl> string B.c = "foo" B loop foreach i B # FAILS AS value linked to "c" is of type string: printf "key: %s, value: %d\n", "$i", $i # ALSO FAILS: printf "key: $i, value: %d\n", $i endloop </hansl> In Julia, similar to Python, a loop over a dictionary constructs two separate variables: one for the key and another one for the value so it's simple to remember on how to access the key or bundle. <Julia> for (k,v) in d println("key: $k, value: $v") end </Julia> I am wondering whether this is a side-product of these languages such that k or v can take every data type on-the-fly. Of course, I know that we don't have that. But I just wanted to mention that it feels like a solid approach for printing key-value pairs. Oh, here is another issue which I do not understand: <hansl> bundle B = _(a = 1, b = 2) loop foreach i B eval B["$i"] # print value works printf "key: %i, value: %d\n", B["$i"] # FAILS: "Syntax error" endloop </hansl> Thanks, Artur