Theory: Aggregation

Data aggregation is the most essential operation when working with trees. Examples of data aggregation are:

• Calculating the total number of files in the directory or the overall size of all files
• Getting a list of all files
• Finding all files by template

The point in aggregating operations is accumulating the result. Traversing the tree in depth using a recursive process, which we discussed in detail in the previous lesson, is well-suited for this. Using it, we can go through all the tree nodes and collect the result, starting from the lowest level.

Let's look at aggregation using a recursive process by counting the total number of nodes in a tree as an example. Essentially, we want to find out how many files and directories are in our file tree:

from hexlet import fs

tree = fs.mkdir('/', [
    fs.mkdir('etc', [
        fs.mkfile('bashrc'),
        fs.mkfile('consul.cfg'),
    ]),
    fs.mkfile('hexletrc'),
    fs.mkdir('bin', [
        fs.mkfile('ls'),
        fs.mkfile('cat'),
    ]),
])

# Getting to the tree bottom using a recursive process in our implementation
def get_nodes_count(node):
    if fs.is_file(node):
        # Returning `1` to account for the current file
        return 1
    # If the node is a directory, we get its children
    children = fs.get_children(node)
    # Here is the hardest part
    # Counting the descendants for each child by recursively calling our function
    descendant_counts = list(map(get_nodes_count, children))
    # Returning current directory `1` plus overall number of descendants
    return 1 + sum(descendant_counts)

get_nodes_count(tree)
# 8

There's not much code here, but it's pretty tricky. There are several key points:

1. The function checks the node type:
   • If it is a file, the function returns the node
   • If the node is a directory, the function returns the child nodes, so we call our function again for each child node and repeat the algorithm
2. Calls to this function on children return the number of their descendants, so we get a list of the numbers that we need to combine
3. At the end, we get the total number of all descendants of the node plus one calculated as the current node itself

Before moving on, have a play around with this code. It's the only way to get to know how it works.