Reading a specific line from CSV file

Question

I have a ten line CSV file. From this file, I only want the, say, fourth line. What's the quickest way to do this? I'm looking for something like:

with open(file, 'r') as my_file:
    reader = csv.reader(my_file)
    print reader[3]

where reader[3] is obviously incorrect syntax for what I want to achieve. How do I move the reader to line 4 and get it's content?

Martijn Pieters · Accepted Answer · 2014-10-04 21:10:41Z

6

If all you have is 10 lines, you can load the whole file into a list:

with open(file, 'r') as my_file:
    reader = csv.reader(my_file)
    rows = list(reader)
    print rows[3]

For a larger file, use itertools.islice():

from itertools import islice

with open(file, 'r') as my_file:
    reader = csv.reader(my_file)
    print next(islice(reader, 3, 4))

answered Oct 4, 2014 at 21:10

Martijn Pieters

1.1m326 gold badges4.2k silver badges3.4k bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Mencia Over a year ago

Is the time complexity of printing one line from a CSV file (done this way) O(1)? If not, is there a way to read a line from a CSV file, in constant time?

Martijn Pieters Over a year ago

@Mencia: your question is muddled. First you talk about printing, then reading a line. Next, Big O deals in asymptotic behaviour, stating how the time taken changes as the number of items processed changes, if you were to increase the number of items to infinity. So for one line out of N lines, there is usually little point in talking about O(...) performance. If it's one line, it's one line, so always O(1) compared to the whole file.

Martijn Pieters Over a year ago

@Mencia: Next, file I/O involves lots of buffering. So next(reader), producing one line from the CSV file triggers a buffered read, reading a block of data into the Python process, to then be decoded from bytes, and then the next line separator needs to be located, and then the CSV module needs to split the line of text into columns. The I/O buffer read is going to be the slowest and as long as the first row fits in that block (about 8 kliobytes, typically), you'll not see any difference in performance regardless of file size or row length. There is no point in worrying about this.

Collectives™ on Stack Overflow

Reading a specific line from CSV file

1 Answer 1

3 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related