Getting values from file if match found for lines with same id in Python

Question

I have a file with lines of data. Each line starts with an id, followed by fixed set of attributes separated by comma.

123,2,kent,...,
123,2,bob,...,
123,2,sarah,...,
123,8,may,...,

154,4,sheila,...,
154,4,jeff,...,

175,3,bob,...,

249,2,jack,...,
249,5,bob,...,
249,3,rose,...,

I would like to get an attribute if the conditions are met. The conditions are if 'bob' appears within the same id, get the value of the second attribute that follows.

For example:

id: 123
values returned: 2, 8

id: 249
values returned: 3

Java has a double loop I can use, but I would like to try this in Python. Any suggestions would be great.

Why the `values returned` of the id 249 is `3` instead of `2, 5, 3`? — aIKid, Feb 20 '14 at 05:32

Steinar Lima · Accepted Answer · 2014-02-20T06:45:17.713

1

I came up with a (perhaps) more pythonic solution which uses groupby and dropwhile. This method yields the same result as the below method, but I think it's prettier.. :) Flags, "curr_id" and stuff like that is not very pythonic, and should be avoided if possible!

import csv
from itertools import groupby, dropwhile

goal = 'bob'
ids = {}

with open('my_data.csv') as ifile:
    reader = csv.reader(ifile)
    for key, rows in groupby(reader, key=lambda r: r[0]):
        matched_rows = list(dropwhile(lambda r: r[2] != goal, rows))
        if len(matched_rows) > 1:
            ids[key] = [row[1] for row in matched_rows[1:]]

print ids

(first solution below)

from collections import defaultdict
import csv

curr_id = None
found = False
goal = 'bob'
ids = defaultdict(list)

with open('my_data.csv') as ifile:
    for row in csv.reader(ifile):
        if row[0] != curr_id:
            found = False
            curr_id = row[0]
        if found:
            ids[curr_id].append(row[1])
        elif row[2] == goal:
            found = True

print dict(ids)

Output:

{'123': ['2', '8'], '249': ['3']}

edited Feb 20 '14 at 06:45

answered Feb 20 '14 at 06:00

Steinar Lima

7,644
2
39
40

+1 Ah that's better than my answer which is similar however it has an empty list for 175 – mechanical_meat Feb 20 '14 at 06:02
@bernie Thanks man, but I'm not satisfied with this solution - I don't like using flags, curr_id and stuff... :) – Steinar Lima Feb 20 '14 at 06:05
@bernie I found another solution if it interests you :) – Steinar Lima Feb 20 '14 at 06:21
1

Very clean. Takes advantage of the data already being sorted by ID. – mechanical_meat Feb 20 '14 at 06:29
Thanks Steinar Lima for the answer! It worked exactly as I want and you were able to figure out the file I was trying to read is actually a csv. – Cryssie Feb 20 '14 at 08:35

score 0 · Answer 2 · answered Feb 20 '14 at 05:37

0

Just set a flag or something as you loop through:

name = 'bob'
id = '123'
found = False

for line in file:
    l = line.split(',')
    if l[0] == id:
        if l[2] == name:
            found = True
        if found:
            print l[1]

answered Feb 20 '14 at 05:37

aIKid

26,968
4
39
65

score 0 · Answer 3 · answered Feb 20 '14 at 05:58

import csv, collections as co, cStringIO as StringIO

s = '''123,2,kent,...,
123,2,bob,...,
123,2,sarah,...,
123,8,may,...,
154,4,sheila,...,
154,4,jeff,...,
175,3,bob,...,
249,2,jack,...,
249,5,bob,...,
249,3,rose,...,'''

filelikeobject = StringIO.StringIO(s)
dd = co.defaultdict(list)
cr = csv.reader(filelikeobject)
for line in cr:
  if line[2] == 'bob':
    dd[line[0]]; continue
  if line[0] in dd:
    dd[line[0]].append(line[1])

Result:

>>> dd
defaultdict(<type 'list'>, {'175': [], '123': ['2', '8'], '249': ['3']})

Getting values from file if match found for lines with same id in Python

3 Answers3