Skipping lines, csv.DictReader

Question

I have a file that has an obnoxious preface to the header. So it looks like this:

Review performed by:    

Meeting:    

Person:     

Number:     

Code: 



Confirmation    

Tab Separated Header Names That I Want To Use

I want to skip past everything and use the tab sep header names for my code. This is what I have so far:

reader = csv.DictReader(CSVFile)
for i in range(14): #trying to skip the first 14 rows
    reader.next()
for row in reader:
    print(row)
    if args.nextCode:
        tab = (row["Tab"])
        sep = int((row["Separated"]))

This code gets this error:

File "/usr/local/Cellar/python/2.7.5/Frameworks/Python.framework/Versions/2.7/lib/python2.7/csv.py", line 104, in next
    row = self.reader.next()
StopIteration

I tried to print the rows, to see where I was in the file, and I changed the “range(14)” to range 5, but when I print the row, I get this:

{'Review performed by:': 'Tab/tSeparated/tHeader/tNames/tThat/tI/tWant/tTo/tUse'}
Traceback (most recent call last):
  File "program.py", line 396, in <module>
    main()
  File "program.py", line 234, in main
    tab = (row["Tab"])
KeyError: 'Tab'

So I am not really sure the right way to skip those top lines. Any help would be appreciated.

Asked By: Stephopolis

||

Source

Answer 1

A csv.DictReader reads the first line from the file when it’s instantiated, to get the headers for subsequent rows. Therefore it uses Review performed by: as the header row, then you skip the next 14 rows.

Instead, skip the lines before creating the DictReader:

for i in range(14):
    CSVFile.next()
reader = csv.DictReader(CSVFile)
...

Answered By: jonrsharpe

Answer 2

You could wrap the CSVFile with an itertools.islice iterator object to slice-off the lines of the preface when creating the DictReader, instead of providing it directly to the constructor.

This works because the csv.reader constructor will accept "any object which supports the iterator protocol and returns a string each time its __next__() method is called" as its first argument according to the csv docs. This also applies to csv.DictReaders because they’re implemented via an underlying csv.reader instance.

Note how the next(iterator).split() expression supplies the csv.DictReader with a fieldnames argument (so it’s not taken it from the first line of the file when it’s instantiated) — so what @jonrsharpe said about the first line being read when the DictReader is instantiated isn’t strictly true.)

iterator = itertools.islice(CSVFile, 14, None)  # Skip header lines.
for row in csv.DictReader(CSVFile, next(iterator).split(), delimiter='t'):
    # process row ...

Answered By: martineau

Skipping lines, csv.DictReader

Question:

Answers: