Remove duplicates in place

A sorted list has duplicates next to each other. Remove them without making a new list, using two positions: a reader that visits every item, and a writer that marks where the next kept item goes. Whenever the reader sees a value different from the last kept one, copy it to the writer's position and advance the writer. At the end, the first writer items are the unique values.

Write dedupe(sorted_nums) that rearranges the list so its first part holds the unique values in order, and returns how many there are. The point is to practise the reader-and-writer method rather than build a new list or use a set.

nums = [1, 1, 2, 3, 3, 3]
dedupe(nums)  ->  3     and nums now starts [1, 2, 3, ...]

def dedupe(sorted_nums):
    ...