Class 7 - Introduction to tables

Date: September 23, 2026

⚠️Homework

Homework 2 due September 27, 2026

🆕New Skills

First week for skills: 0

Reference

For all work with tables, refer to the Tables page in the menu at the top of the page!

Skills: 3

Notes (review before class)

Tables

• A table has rows that share the same attributes, its columns: a workout journal, a recipe index, a library catalog.
• Tables are values: they can be named, passed to functions, and returned. PLL provides them; there's nothing to import.
• Code that works with tables follows their structure of rows and columns.

Literal Tables

A literal table is written out with table, given the column names, in order, and then a list of rows:

people = table(
    ["name", "age"],
    [
        ["Alicia", 30],
        ["Meihui", 40],
        ["Jamal", 25],
    ],
)

• Each row has one value per column, in order. (Try: remove a value or a comma, and read the error.)
• The directory gets just people, for the whole table: no column names.
• Tables are equal only if their columns and rows are in the same order. A test function can check a fact, rather than a function (!= is "not equal"):

def test_column_order_matters():
    assert people != table(
        ["age", "name"],
        [[30, "Alicia"], [40, "Meihui"], [25, "Jamal"]],
    )

Aside: Lists

Square brackets make a list: values, in order, separated by commas, like [30, 40, 25]. table is given a list of column names and a list of rows, each itself a list. More in Class 13.

Loading Tables from CSV Files

• load_table loads a CSV ("comma-separated values") file, which any spreadsheet can save, from a URL or a file in your program's folder: load_table("riders.csv"). The column names come from its first line.
• Every value loaded from a CSV is a string, even "4". Convert numeric columns, into a new table, with transform_column (Class 9) and int ("4" to 4) or float ("4.5" to 4.5): riders = riders_raw.transform_column("riders", int).
• Table operations like this are methods (like the string methods of Class 2): written after the table, with a dot.

Rows and Cells

t.row(n) is row n, counting from 0 at the top (so the last is one less than the number of rows). A row is a dict: a column's name, a string, in square brackets gives its value.

shuttle = table(
    ["month", "riders"],
    [
        ["Jan", 1123],
        ["Feb", 1045],
        ["Mar", 1087],
        ["Apr", 999],
    ],
)

shuttle.row(2)            # -> {'month': 'Mar', 'riders': 1087}
shuttle.row(2)["riders"]  # -> 1087
march_row = shuttle.row(2)
march_row["riders"]       # -> 1087

• Values work in any computation: shuttle.row(2)["riders"] >= 1000.
• Without quotes, shuttle.row(2)[riders] is an error: riders isn't in the directory.
• Only the named version adds an entry, march_row, to the directory.
• Running a table's name shows it (its first rows, if it is long).

Aside: Dictionaries

A dictionary (dict), written {key: value, ...}, maps keys to values: you look up a value by its key, not its position.

capitals = {"France": "Paris", "Japan": "Tokyo", "Peru": "Lima"}
capitals["Japan"]   # -> 'Tokyo'

• Each key appears once. Keys needn't be strings: {411: "Pauline"} maps a room number to a person.
• So a row's values are looked up by column name; a table's rows, by position.
• A missing key, like capitals["Chile"], is an error, not a quiet nothing.

Summarizing Columns

sum, mean, min, and max give a column's total, average, smallest, and largest value (more in Tables): shuttle.max("riders") is 1123.

Recap (15 mins)

A workout journal is small enough to type in as a table; a recipe index is a CSV file on the web. Read through the program:

workouts = table(
    ["date", "activity", "duration"],
    [
        ["2025-04-01", "Running", 30],
        ["2025-04-02", "Yoga", 45],
        ["2025-04-03", "Cycling", 60],
    ],
)

def test_row_order_matters():
    assert workouts != table(
        ["date", "activity", "duration"],
        [
            ["2025-04-03", "Cycling", 60],
            ["2025-04-01", "Running", 30],
            ["2025-04-02", "Yoga", 45],
        ],
    )

second_workout = workouts.row(1)
second_activity = second_workout["activity"]
average_duration = workouts.mean("duration")

recipes_raw = load_table("https://raw.githubusercontent.com/neu-pdi/cs2000-public-resources/refs/heads/main/static/support/5-recipes.csv")
recipes_with_servings = recipes_raw.transform_column("servings", int)
recipes = recipes_with_servings.transform_column("prep time", int)
average_prep_time = recipes.mean("prep time")

• What are second_activity and average_duration? What expression would give the date of the first workout?
• Why does test_row_order_matters pass?
• Why does the program compute average_prep_time from recipes, rather than from recipes_raw? How are those two tables different?

Class Exercise (40 mins)

1. Browse and pick a dataset on the London DataStore. Once you found a dataset to work with, open the page for details. Scroll down until you see a data set in CSV format. Right-click the "Download" button so you can copy (not download) the link; PLL will download it. Use the copied link (URL) with load_table in your code to create a table from it. (If PLL reports that it could not reach the URL, the site may not allow other sites to read its data. In that case, download the CSV, drag it into your cs2000-scratch repository in the Explorer on the left, and give load_table its file name instead.)
2. Look at the columns PLL loaded: which of them should be converted to numbers? Use transform_column to do that for one of them.
3. Use the interactions panel, .row, and square brackets to explore the data a little bit.
4. Check the total number of rows with table_identifier.length(). What happens if you try to do table_identifier.row(M) for M that is greater than the total number of rows?
5. Similarly, try extracting a column that doesn't exist.

Wrap-up (5 mins)

• Tables are an extremely common and powerful form of data. Today we just learned how to look at them; in the upcoming days we will see how to program with them!