A Table of Two Things: pivot_table in Python

A group-by answers a question about one thing, like items sold per category. Sometimes you want two things at once: items per category, split by status. A pivot table lays that out as a grid: one row for each value of one column, one column for each value of another, and in each cell a summary of the rows that have both.

df.pivot_table(...) gives back that new table. You name four things: index, the column whose values become the rows; columns, the column whose values become the table's columns; values, the column that fills the cells; and aggfunc, how each cell combines its rows, like "sum" or "mean" (without it you get the mean). For example, the average spent per city, split by newsletter:

by_city = customers.pivot_table(
    index="city",
    columns="newsletter",
    values="spent",
    aggfunc="mean",
)

A pair with no rows at all gets NaN, "no value". fill_value=0 puts 0 in those cells instead. That's true for a sum (no lines means 0 items), but not for a mean, where an empty cell has no answer.

The starting code keeps the March 2026 lines of orders.csv in march: .str.startswith("2026-03") gives back True for each date that begins with 2026-03. Your program must print the items sold in March per category and status:

status    cancelled  paid
category                 
gifts             1     7
home              0    27
kitchen           0    24
office            0    44

Your task: build table from march with pivot_table: one row per category, one column per status, the sum of quantity in each cell and 0 in the empty cells, then print it with .to_string().

Assignment
Follow the Coding Tutorial to build the March table of items per category and status with pivot_table.

Hint
Write each argument on its own line, ending with a comma, between march.pivot_table( and the closing ). Without values, pandas tries to average every other column, and text can't be averaged.