PY-L4-04 · CSV Files — Writing & Updating Rows

Learning Goals

3 min

Write a fresh CSV with csv.writer — header + rows.
Write with csv.DictWriter — supply fieldnames, write dicts.
Tell the difference between "w" (overwrite) and "a" (append).
Update existing rows with the read-modify-write pattern.

Warm-Up · Which Mode?

5 min

Look at these three calls to open. What happens to an existing file in each case?

open("data.csv", "r")
open("data.csv", "w")
open("data.csv", "a")

Show the answer

"r" — read. File is opened for reading; if it doesn't exist, you get FileNotFoundError.
"w" — write. Truncates the file to empty the moment you open it. Be careful — there is no undo.
"a" — append. Existing rows are preserved; new writes go at the end.

Today's big idea

To change existing rows in a CSV you don't edit in place. You read the whole file into memory, change the list, and write the whole file back out. That's the read-modify-write pattern, and it works for almost every file-update task.

New Concept · writer, DictWriter, append

14 min

csv.writer — write rows as lists

import csv

rows = [
    ["name", "age", "score"],   # header
    ["Aisyah",  13, 88],
    ["Wei Jie", 14, 75],
    ["Suresh",  13, 92],
]

with open("out.csv", "w", newline="") as f:
    writer = csv.writer(f)
    writer.writerows(rows)

out.csv
name,age,score
Aisyah,13,88
Wei Jie,14,75
Suresh,13,92

writerows writes a list of rows at once. There's also writerow for a single row.

csv.DictWriter — write rows as dicts

import csv

students = [
    {"name": "Aisyah",  "age": 13, "score": 88},
    {"name": "Wei Jie", "age": 14, "score": 75},
    {"name": "Suresh",  "age": 13, "score": 92},
]

with open("out.csv", "w", newline="") as f:
    writer = csv.DictWriter(f, fieldnames=["name", "age", "score"])
    writer.writeheader()
    writer.writerows(students)

You always supply fieldnames — they become the header AND the column order. Anything not in fieldnames is dropped; anything missing is left blank.

Append mode — add to an existing file

# Don't write the header again — it's already on disk.
with open("out.csv", "a", newline="") as f:
    writer = csv.DictWriter(f, fieldnames=["name", "age", "score"])
    writer.writerow({"name": "Mei", "age": 14, "score": 80})

Append mode is perfect for log files: each run adds one row to the end.

Read-modify-write update

To raise Wei Jie's score by 5: read, change, rewrite.

import csv

# 1. Read the whole file
with open("out.csv", newline="") as f:
    rows = list(csv.DictReader(f))

# 2. Modify
for r in rows:
    if r["name"] == "Wei Jie":
        r["score"] = str(int(r["score"]) + 5)   # CSV stays strings

# 3. Write everything back (overwrite)
with open("out.csv", "w", newline="") as f:
    writer = csv.DictWriter(f, fieldnames=rows[0].keys())
    writer.writeheader()
    writer.writerows(rows)

This is the single most common file-editing pattern in software. Memorise it.

Worked Example · Attendance Logger

12 min

An attendance app records who showed up today. New runs append; we never lose history.

# attendance.py — append-only attendance log
import csv
from datetime import date
from pathlib import Path

FILE   = Path("attendance.csv")
FIELDS = ["date", "name", "status"]

# Create with header if it doesn't exist yet
if not FILE.exists():
    with FILE.open("w", newline="") as f:
        csv.DictWriter(f, fieldnames=FIELDS).writeheader()

# Today's roll-call
today = str(date.today())
roll  = ["Aisyah", "Wei Jie", "Suresh", "Mei"]

with FILE.open("a", newline="") as f:
    writer = csv.DictWriter(f, fieldnames=FIELDS)
    for name in roll:
        status = input(f"  {name}? (p/a): ").strip().lower()
        writer.writerow({
            "date":   today,
            "name":   name,
            "status": "present" if status == "p" else "absent",
        })

print(f"\n📋 logged {len(roll)} entries to {FILE}")

Sample run

  Aisyah? (p/a): p
  Wei Jie? (p/a): a
  Suresh? (p/a): p
  Mei? (p/a): p

📋 logged 4 entries to attendance.csv

attendance.csv after three days

date,name,status
2026-05-26,Aisyah,present
2026-05-26,Wei Jie,absent
2026-05-26,Suresh,present
2026-05-26,Mei,present
2026-05-27,Aisyah,present
...

Read the diff

Three smart choices: header-on-first-run, append mode for history, date.today() from the stdlib. Run the program any day and history grows. This is how production logs work — never overwrite, always append.

Try It Yourself

13 min

01 🟢 Write a fresh CSV

Make a list of three favourite movies (title, year, rating). Save it as movies.csv with a header.

Hint

import csv
movies = [
    {"title": "Up",       "year": 2009, "rating": 5},
    {"title": "Inside Out","year": 2015, "rating": 5},
    {"title": "Coco",     "year": 2017, "rating": 4},
]
with open("movies.csv", "w", newline="") as f:
    w = csv.DictWriter(f, fieldnames=["title", "year", "rating"])
    w.writeheader()
    w.writerows(movies)

02 🟡 Append two more

Append two more movies to movies.csv WITHOUT writing the header again.

Hint

extra = [
    {"title": "Toy Story 3", "year": 2010, "rating": 5},
    {"title": "Brave",       "year": 2012, "rating": 4},
]
with open("movies.csv", "a", newline="") as f:
    w = csv.DictWriter(f, fieldnames=["title", "year", "rating"])
    w.writerows(extra)        # NO writeheader()

03 🔴 Update a rating

Bump every 4-star movie up to 5 stars. Use read-modify-write.

Hint

with open("movies.csv", newline="") as f:
    rows = list(csv.DictReader(f))

for r in rows:
    if int(r["rating"]) == 4:
        r["rating"] = "5"

with open("movies.csv", "w", newline="") as f:
    w = csv.DictWriter(f, fieldnames=rows[0].keys())
    w.writeheader()
    w.writerows(rows)

Mini-Challenge · Daily Mood Tracker

8 min

Build mood.py. Each run:

Asks "how are you today? (1-5)" and an optional one-line note.
Appends a row to mood.csv with today's date, the rating and the note.
Reads the whole file and reports: total entries, average rating, the day with the highest rating so far.

Show one possible solution

# mood.py — append + summarise
import csv
from datetime import date
from pathlib import Path

FILE   = Path("mood.csv")
FIELDS = ["date", "rating", "note"]

if not FILE.exists():
    with FILE.open("w", newline="") as f:
        csv.DictWriter(f, fieldnames=FIELDS).writeheader()

rating = int(input("Mood today (1-5): "))
note   = input("One line about it: ").strip()

with FILE.open("a", newline="") as f:
    csv.DictWriter(f, fieldnames=FIELDS).writerow({
        "date": str(date.today()), "rating": rating, "note": note,
    })

# Summarise
with FILE.open(newline="") as f:
    rows = list(csv.DictReader(f))

ratings = [int(r["rating"]) for r in rows]
best    = max(rows, key=lambda r: int(r["rating"]))
print(f"\nEntries: {len(rows)}")
print(f"Average: {sum(ratings)/len(ratings):.2f}")
print(f"Best day so far: {best['date']} (rating {best['rating']})")

Non-negotiables: file header on first run, append for new entries, read-back for the summary stats.

Recap

3 min

csv.writer takes lists; csv.DictWriter takes dicts (cleaner, name-based). Mode "w" truncates and rewrites; "a" appends. To change existing rows, read the whole file, modify the in-memory list, write everything back. Tomorrow we use these skills to build a CSV ↔ JSON converter.

Vocabulary Card

writeheader / writerow / writerows: DictWriter methods — write the header line, one row, many rows.
append mode: open(path, "a"). Existing content kept; new writes go to the end.
read-modify-write: Standard pattern for editing a file: load everything, change in memory, save everything back.

Homework

4 min

Build todo.py:

Stores tasks in todo.csv with columns id, task, done.
The program offers a menu: 1) list tasks, 2) add task, 3) mark task done, 4) quit.
"done" is stored as the string "yes" or "no".
Use append for adds; use read-modify-write for "mark done".

Sample · todo.py

# todo.py — CSV-backed to-do list
import csv
from pathlib import Path

FILE   = Path("todo.csv")
FIELDS = ["id", "task", "done"]

def load():
    if not FILE.exists():
        return []
    with FILE.open(newline="") as f:
        return list(csv.DictReader(f))

def save(rows):
    with FILE.open("w", newline="") as f:
        w = csv.DictWriter(f, fieldnames=FIELDS)
        w.writeheader()
        w.writerows(rows)

def list_tasks(rows):
    if not rows:
        print("  (no tasks yet)")
        return
    for r in rows:
        tick = "✅" if r["done"] == "yes" else "⬜"
        print(f"  {tick} #{r['id']}  {r['task']}")

def add_task(rows):
    task = input("  task: ").strip()
    new_id = str(max([int(r["id"]) for r in rows], default=0) + 1)
    rows.append({"id": new_id, "task": task, "done": "no"})
    save(rows)
    print(f"  added #{new_id}")

def mark_done(rows):
    tid = input("  id to mark done: ").strip()
    for r in rows:
        if r["id"] == tid:
            r["done"] = "yes"
    save(rows)

while True:
    rows = load()
    print("\n1) list  2) add  3) done  4) quit")
    choice = input("> ").strip()
    if   choice == "1": list_tasks(rows)
    elif choice == "2": add_task(rows)
    elif choice == "3": mark_done(rows)
    elif choice == "4": break

Non-negotiables: persistent storage, growing ids, mark-done updates the file, file is created if missing.

import csv rows = [ ["name", "age", "score"], # header ["Aisyah", 13, 88], ["Wei Jie", 14, 75], ["Suresh", 13, 92], ] with open("out.csv", "w", newline="") as f: writer = csv.writer(f) writer.writerows(rows)

import csv students = [ {"name": "Aisyah", "age": 13, "score": 88}, {"name": "Wei Jie", "age": 14, "score": 75}, {"name": "Suresh", "age": 13, "score": 92}, ] with open("out.csv", "w", newline="") as f: writer = csv.DictWriter(f, fieldnames=["name", "age", "score"]) writer.writeheader() writer.writerows(students)

# Don't write the header again — it's already on disk. with open("out.csv", "a", newline="") as f: writer = csv.DictWriter(f, fieldnames=["name", "age", "score"]) writer.writerow({"name": "Mei", "age": 14, "score": 80})

import csv # 1. Read the whole file with open("out.csv", newline="") as f: rows = list(csv.DictReader(f)) # 2. Modify for r in rows: if r["name"] == "Wei Jie": r["score"] = str(int(r["score"]) + 5) # CSV stays strings # 3. Write everything back (overwrite) with open("out.csv", "w", newline="") as f: writer = csv.DictWriter(f, fieldnames=rows[0].keys()) writer.writeheader() writer.writerows(rows)

# attendance.py — append-only attendance log import csv from datetime import date from pathlib import Path FILE = Path("attendance.csv") FIELDS = ["date", "name", "status"] # Create with header if it doesn't exist yet if not FILE.exists(): with FILE.open("w", newline="") as f: csv.DictWriter(f, fieldnames=FIELDS).writeheader() # Today's roll-call today = str(date.today()) roll = ["Aisyah", "Wei Jie", "Suresh", "Mei"] with FILE.open("a", newline="") as f: writer = csv.DictWriter(f, fieldnames=FIELDS) for name in roll: status = input(f" {name}? (p/a): ").strip().lower() writer.writerow({ "date": today, "name": name, "status": "present" if status == "p" else "absent", }) print(f"\n📋 logged {len(roll)} entries to {FILE}")

import csv movies = [ {"title": "Up", "year": 2009, "rating": 5}, {"title": "Inside Out","year": 2015, "rating": 5}, {"title": "Coco", "year": 2017, "rating": 4}, ] with open("movies.csv", "w", newline="") as f: w = csv.DictWriter(f, fieldnames=["title", "year", "rating"]) w.writeheader() w.writerows(movies)

extra = [ {"title": "Toy Story 3", "year": 2010, "rating": 5}, {"title": "Brave", "year": 2012, "rating": 4}, ] with open("movies.csv", "a", newline="") as f: w = csv.DictWriter(f, fieldnames=["title", "year", "rating"]) w.writerows(extra) # NO writeheader()

with open("movies.csv", newline="") as f: rows = list(csv.DictReader(f)) for r in rows: if int(r["rating"]) == 4: r["rating"] = "5" with open("movies.csv", "w", newline="") as f: w = csv.DictWriter(f, fieldnames=rows[0].keys()) w.writeheader() w.writerows(rows)

# mood.py — append + summarise import csv from datetime import date from pathlib import Path FILE = Path("mood.csv") FIELDS = ["date", "rating", "note"] if not FILE.exists(): with FILE.open("w", newline="") as f: csv.DictWriter(f, fieldnames=FIELDS).writeheader() rating = int(input("Mood today (1-5): ")) note = input("One line about it: ").strip() with FILE.open("a", newline="") as f: csv.DictWriter(f, fieldnames=FIELDS).writerow({ "date": str(date.today()), "rating": rating, "note": note, }) # Summarise with FILE.open(newline="") as f: rows = list(csv.DictReader(f)) ratings = [int(r["rating"]) for r in rows] best = max(rows, key=lambda r: int(r["rating"])) print(f"\nEntries: {len(rows)}") print(f"Average: {sum(ratings)/len(ratings):.2f}") print(f"Best day so far: {best['date']} (rating {best['rating']})")