root@vm-learning
~
$
open
ch-1-7
UNIT 1 ▪ CHAPTER 7
07
CSV Files
csv module · writer · writerow · writerows · reader · DictReader · Worked Programs
A CSV file (Comma-Separated Values) is a plain text file that stores tabular data — rows and columns — with a comma (or another separator) between the columns. It is the universal translator between Python, Excel, Google Sheets, and almost every database in the world.
Real-life analogy. Think of CSV as spreadsheet paper in plain English. If Excel and Python want to exchange data, they both speak CSV. No fancy format, no hidden bytes — just lines where commas mark the column borders.
7.1 What is a CSV file?
Each line is one record (one row). Inside a line, fields (columns) are separated by a comma. Here is marks.csv as it sits on disk — open it in Notepad and you’ll see exactly this:
roll,name,marks
1,Asha,92
2,Rahul,78
3,Riya,85
4,Ankit,71
The first line is an optional header — it names the columns. The rest are data rows. Open the same file in Excel / LibreOffice and it appears as a proper table.
7.2 Why do we use CSV?
- Readable — any editor can show it; no special software needed.
- Portable — works on every operating system and every programming language.
- Compact — no XML tags, no extra metadata.
- Directly understood by Excel / Google Sheets — double-click and it opens as a table.
- Easy to back up, diff, email — it is just text.
7.2.1 CSV vs Text vs Binary — at a glance
| Text (.txt) | CSV (.csv) | Binary (.dat) | |
|---|---|---|---|
| Structure | Free-form lines | Rows & columns | Raw bytes |
| Opens in Excel? | No (plain view) | Yes (as a table) | No |
| Keeps Python types? | No — always str | No — always str | Yes (with pickle) |
| Ideal for | Logs, notes | Spreadsheet exchange | Images, Python objects |
7.3 The csv Module
Python ships with a built-in module called csv that reads and writes CSV correctly — handling edge cases like commas inside a quoted field, newlines inside a field, and different separators.
import csv
7.3.1 Why not just use split(",")?
For the simple files in your CBSE practicals you technically could. But consider a single row like:
4,"Ankit Kumar, Jr.",71
"Ankit Kumar, Jr." has a comma inside quotes — it is still one field. Calling .split(",") would wrongly break it into four pieces. The csv module understands quoting, so always prefer it.
7.4 Writing a CSV File
7.4.1 Step 1 — open the file and create a writer
import csv
with open("marks.csv", "w", newline="") as f: # note: newline=""
w = csv.writer(f)
w.writerow(["roll", "name", "marks"]) # header
w.writerow([1, "Asha", 92])
w.writerow([2, "Rahul", 78])
w.writerow([3, "Riya", 85])
Always pass
newline="" when opening a CSV file (both for reading and writing). On Windows, without it, Python writes extra \r characters — producing blank lines between every row when Excel opens the file.
7.4.2 writer.writerow(row_list) — write one row
Takes a list (or tuple) of values and writes them as one comma-separated line, adding the newline for you.
7.4.3 writer.writerows(list_of_rows) — write many rows
import csv
rows = [
["roll", "name", "marks"],
[1, "Asha", 92],
[2, "Rahul", 78],
[3, "Riya", 85],
[4, "Ankit", 71],
]
with open("marks.csv", "w", newline="") as f:
csv.writer(f).writerows(rows) # one call writes every row
writerow = one row at a time. writerows = everything in one shot (takes a list of rows). Do not mix them up — writerows with a single row writes each character as a separate column.
7.4.4 Appending rows — mode "a"
import csv
new_rows = [
[5, "Karan", 66],
[6, "Meera", 95],
]
with open("marks.csv", "a", newline="") as f:
csv.writer(f).writerows(new_rows) # appended at the end
7.5 Reading a CSV File
7.5.1 csv.reader — iterate over rows
import csv
with open("marks.csv", "r", newline="") as f:
r = csv.reader(f)
for row in r:
print(row)
['roll', 'name', 'marks']
['1', 'Asha', '92']
['2', 'Rahul', '78']
['3', 'Riya', '85']
['4', 'Ankit', '71']
Every value read from a CSV is a string — even numbers. Convert with
int() / float() before doing math.
7.5.2 Skipping the header row
import csv
with open("marks.csv", "r", newline="") as f:
r = csv.reader(f)
header = next(r) # read the first line into 'header'
for row in r: # remaining rows
print(row[1], "scored", row[2])
7.5.3 Converting columns as you read
import csv
total = 0
count = 0
with open("marks.csv", "r", newline="") as f:
r = csv.reader(f)
next(r) # skip header
for roll, name, marks in r: # unpacking
total += int(marks) # string → int
count += 1
print(f"Average marks = {total/count:.2f}")
7.6 Custom Delimiters — TSV, semicolons, pipes
CSV’s “C” stands for comma, but the separator is configurable. European spreadsheets often use ; because they write decimals with a comma; log files sometimes use TAB (\t) or |.
import csv
# write using ';' as separator
with open("marks_eu.csv", "w", newline="") as f:
w = csv.writer(f, delimiter=";")
w.writerow(["roll", "name", "marks"])
w.writerow([1, "Asha", 92])
# read it back — tell the reader to expect the same delimiter
with open("marks_eu.csv", "r", newline="") as f:
for row in csv.reader(f, delimiter=";"):
print(row)
7.6.1 Other useful csv options
| Option | What it does |
|---|---|
delimiter | Column separator (default ",") |
quotechar | Character that wraps text fields (default '"') |
quoting | When to quote — csv.QUOTE_ALL, QUOTE_MINIMAL, … |
lineterminator | String written at the end of each row |
7.7 DictReader & DictWriter — readable, self-documenting code
Instead of accessing columns by index (row[0], row[1] …) — which is easy to confuse — you can access them by column name.
7.7.1 csv.DictReader
import csv
with open("marks.csv", "r", newline="") as f:
r = csv.DictReader(f) # uses first row as keys
for row in r:
print(row["name"], "->", row["marks"])
7.7.2 csv.DictWriter
import csv
rows = [
{"roll": 1, "name": "Asha", "marks": 92},
{"roll": 2, "name": "Rahul", "marks": 78},
]
with open("marks.csv", "w", newline="") as f:
w = csv.DictWriter(f, fieldnames=["roll", "name", "marks"])
w.writeheader() # writes the header row
w.writerows(rows) # every dict becomes one row
Use
DictReader / DictWriter whenever the file has a header row — your code becomes self-documenting, and adding / reordering columns later does not break it.
7.8 CBSE-style Worked Programs
7.8.1 Create a CSV of student records
import csv
rows = []
while True:
roll = int(input("Roll : "))
name = input("Name : ")
marks = int(input("Marks: "))
rows.append([roll, name, marks])
if input("Another? (y/n): ").lower() != "y":
break
with open("students.csv", "w", newline="") as f:
w = csv.writer(f)
w.writerow(["roll", "name", "marks"])
w.writerows(rows)
print("File written.")
7.8.2 Display every record in the CSV
import csv
with open("students.csv", "r", newline="") as f:
for row in csv.reader(f):
print(row)
7.8.3 Count records where marks > 75
import csv
count = 0
with open("students.csv", "r", newline="") as f:
r = csv.reader(f)
next(r) # skip header
for row in r:
if int(row[2]) > 75:
count += 1
print("High scorers:", count)
7.8.4 Search a student by roll number
import csv
target = input("Roll to find: ")
with open("students.csv", "r", newline="") as f:
r = csv.reader(f)
next(r)
for row in r:
if row[0] == target:
print("Found:", row)
break
else:
print("No match.")
7.8.5 Append a new student interactively
import csv
rec = [
int(input("Roll : ")),
input("Name : "),
int(input("Marks: ")),
]
with open("students.csv", "a", newline="") as f:
csv.writer(f).writerow(rec)
print("Appended.")
7.8.6 Update marks for a given roll number
Like binary files, CSV updates follow the read-all → modify → write-all pattern.
import csv
target = "2"
new_marks = "88"
# 1) read
with open("students.csv", "r", newline="") as f:
rows = list(csv.reader(f))
# 2) modify (skip header)
for row in rows[1:]:
if row[0] == target:
row[2] = new_marks
# 3) write back
with open("students.csv", "w", newline="") as f:
csv.writer(f).writerows(rows)
print("Updated.")
7.8.7 Copy passing students (marks ≥ 33) to another CSV
import csv
with open("students.csv", "r", newline="") as src, \
open("passed.csv", "w", newline="") as dst:
r = csv.reader(src)
w = csv.writer(dst)
header = next(r)
w.writerow(header)
for row in r:
if int(row[2]) >= 33:
w.writerow(row)
print("Copied passing students.")
7.8.8 Class topper from the CSV
import csv
with open("students.csv", "r", newline="") as f:
r = csv.DictReader(f)
topper = max(r, key=lambda row: int(row["marks"]))
print("Topper:", topper["name"], "—", topper["marks"])
7.8.9 Merge two CSVs into one
import csv
with open("all.csv", "w", newline="") as out:
w = csv.writer(out)
for fname in ("section_A.csv", "section_B.csv"):
with open(fname, "r", newline="") as f:
w.writerows(csv.reader(f))
print("Merged.")
7.8.10 CSV → text summary report
import csv
with open("students.csv", "r", newline="") as src, \
open("report.txt", "w") as dst:
r = csv.DictReader(src)
for row in r:
line = f"{row['roll']:>3}. {row['name']:<15} {row['marks']}\n"
dst.write(line)
print("Report saved.")
7.9 Common Mistakes to Avoid
| # | Mistake | Fix |
|---|---|---|
| 1 | Missing newline="" while opening | Always pass it — blank lines vanish |
| 2 | Forgetting that every value read is a string | Convert with int() / float() before math |
| 3 | writerows(single_row) — each character becomes a column | Use writerow for one row |
| 4 | Forgetting next(r) — header is treated as data | Call next(r) once to skip it |
| 5 | Using split(",") on rows with quoted commas | Use the csv module |
| 6 | Writing in mode "w" when you meant "a" | "w" wipes; "a" appends |
| 7 | Wrong delimiter (file uses ;) | Pass delimiter=";" to reader / writer |
| 8 | Trying to update a row in the middle of the file | Use the read-all / modify / write-all pattern |
Quick-revision summary
- A CSV is a plain-text, comma-separated table — the universal format between Python, Excel and databases.
import csv; open the file withnewline=""— always.csv.writer(f).writerow(row)writes one row;writerows(list_of_rows)writes many at once.csv.reader(f)yields each row as a list of strings; convert numeric columns before math.- Skip the header with
next(reader). DictReader/DictWriterlet you access columns by name — recommended when the file has a header.- Custom separators: pass
delimiter=";"(or any other character). - Updating rows follows the read-all → modify → write-all pattern, just like binary files.