What is not a row in a csv file?

Question

I came upon this code in a tutorial:

 for row in csv_reader:
        if not row:
            continue
        dataset.append(row)

Which I take to mean that if the code encounters something other than a row, just skip and continue. Is that correct?

What defines 'not row'?

An empty line. Maybe you have a blank line and you want to ignore it. This would give you all non-empty lines. — idjaw, Jul 23 '18 at 02:19
It's checking to see of the "truthiness" of `row` is `False` which would indicate the value returned was an empty sequence of elements. — martineau, Jul 23 '18 at 02:20
I thought it might be an empty row among other things, but I can't seem to find definitive literature. @ martineau, you sound pretty definitive so I'll go with that until something reveals that it is not. Thanks. — spacedustpi, Jul 23 '18 at 02:24
Possible duplicate of [What is Truthy and Falsy in python? How is it different from True and False?](https://stackoverflow.com/questions/39983695/what-is-truthy-and-falsy-in-python-how-is-it-different-from-true-and-false) — Merik, Jul 23 '18 at 02:26

score 1 · Accepted Answer · answered Jul 23 '18 at 02:22

This allows you to skip over empty lines in a CSV file.

not row just means "row is falsey".

In Python, the following things are falsey:

False
None
Numeric zeroes
Empty containers

(Of course you can write your own class with a __bool__ method that does anything you want—but by convention, it should follow the same rule.)

This is all explained in Boolean operations in the docs.

--

The rows iterated by a csv.reader are lists. (And lists are containers, so they're falsey iff they're empty.)

An empty line produces an empty list; a line with text but no delimiters produces a list of one string; a line with delimiters produces a list of two or more strings.

This is covered in the csv module docs… but not really all in one place.

rafaelc · Answer 2 · 2018-07-23T02:28:08.820

1

row is just a variable name in this context. When you do if row, you actually checking if there is any content to the variable that python considers to be True.

Take a look at this answer from Patrick Haugh in which he highlights lots of examples of what is Falsy in python.

To illustrate in a minimal example:

import csv
for row in csv.reader(['row1,foo','', 'row3,bar']):
    print(row)

yields

['row1', 'foo']
[]
['row3', 'bar']

But if you do

for row in csv.reader(['row1,foo','', 'row3,bar']):
    if row:
        print(row)

Then output is

['row1', 'foo']
['row3', 'bar']

and thus basically the empty row is filtered out.

edited Jul 23 '18 at 02:28

answered Jul 23 '18 at 02:25

rafaelc

57,686
15
58
82

1

Surely you wanted to use the same list in your second example as in your first? Because that will just print out the line `['one', 'two', 'three', '']`. – abarnert Jul 23 '18 at 02:27
Perhaps it is more efficient to write: for row in csv_reader: if row: dataset.append(row) – spacedustpi Jul 23 '18 at 02:30

What is not a row in a csv file?

2 Answers2