Informational guide

Convert TXT to CSV in Python With pandas

If you convert text to CSV more than once, stop doing it by hand. A few lines of Python turn it into a repeatable, correct process that handles quoting for you. This guide covers the pandas approach for real-world files, the standard-library approach when you cannot install anything, and the specific arguments that fix the errors you will actually hit.

How do you convert TXT to CSV in pandas?

Read the text with read_csv, set sep to your source delimiter, and write it out with to_csv using index=False. That is the whole job for a clean file:

import pandas as pd

df = pd.read_csv("input.txt", sep="\t")   # tab-separated source
df.to_csv("output.csv", index=False)

read_csv does the parsing and to_csv writes an RFC 4180 compliant file, so a value like Smith, John is automatically wrapped in quotes and your columns never shift. The index=False argument stops pandas from adding an unwanted row-number column. The official argument list lives in the pandas documentation.

Handling different separators

The sep argument is where most files differ. Match it to your source:

The whitespace case is the one people miss. Console output and fixed-width exports often use a variable number of spaces to line up columns. sep=r"\s+" treats any run of whitespace as a single separator, which collapses that ragged spacing into clean fields.

Files with no header

By default pandas treats the first line as column names. When your text file starts straight into data, that eats your first row. Turn it off and name the columns yourself:

import pandas as pd

df = pd.read_csv("input.txt", sep=r"\s+", header=None,
                 names=["name", "age", "city"])
df.to_csv("output.csv", index=False)

Fixing encoding errors

A UnicodeDecodeError means the file is not UTF-8. Tell pandas the real encoding on the way in, and force a BOM on the way out so Excel opens the result cleanly:

df = pd.read_csv("input.txt", sep="\t", encoding="latin-1")
df.to_csv("output.csv", index=False, encoding="utf-8-sig")

Without pandas: the csv module

If you cannot install pandas, the standard library does the job. This reads a tab-separated file and writes a comma CSV with correct quoting:

import csv

with open("input.txt", newline="", encoding="utf-8") as fin, \
     open("output.csv", "w", newline="", encoding="utf-8") as fout:
    reader = csv.reader(fin, delimiter="\t")
    writer = csv.writer(fout)          # defaults to comma, quotes as needed
    writer.writerows(reader)

The csv.writer applies quoting automatically, so you get the same safe output as pandas without the dependency. The newline="" argument prevents blank lines between rows on Windows, which is a classic gotcha.

Batch converting a folder

To convert every text file in a directory, loop over them:

import pandas as pd
from pathlib import Path

for txt in Path("data").glob("*.txt"):
    pd.read_csv(txt, sep="\t").to_csv(txt.with_suffix(".csv"), index=False)

When a script is overkill

For a single file you will convert once, spinning up Python is more work than the task deserves. Paste it into the browser converter instead: it applies the same quoting rules, auto-detects the delimiter, and runs locally so nothing is uploaded. Use Python when the job repeats, and the tool when it does not. The full method guide compares every option, and the Excel guide covers the spreadsheet route.