Pick up lines from a file based on line numbers in another file

Question

I have two files - one contains the addresses (line numbers) and the other one data, like this:

address file:

data file

1.000451451
2.000589214
3.117892278
4.479511994
5.484514874
6.784499874
7.021239396

I want to randomize the data file based on the numbers of address files so I get:

2.000589214
4.479511994
6.784499874
7.021239396
1.000451451
3.117892278
5.484514874

I want to do it either in python or in bash, but didn't yet find any solution.

Does the address file contains line number or the `int`s of numbers from the data file? — heemayl, Jun 09 '17 at 04:07
`I want to do it either in python or in bash` then try something and add the code in question... — Sundeep, Jun 09 '17 at 04:45

codeforester · Answer 1 · 2017-06-09T18:51:08.753

3

If you don't mind sed, we can use process substitution to achieve this easily:

sed -nf <(sed 's/$/p/' addr.txt) data.txt

-n suppresses the default printing
-f makes sed read commands from the process substitution <(...)
<(sed 's/$/p/' addr.txt) creates sed print commands based on line numbers in addr.txt

Gives the output:

2.000589214
4.479511994
6.784499874
7.021239396
1.000451451
3.117892278
5.484514874

edited Jun 09 '17 at 18:51

answered Jun 09 '17 at 04:42

codeforester

39,467
16
112
140

score 2 · Accepted Answer · edited Jun 09 '17 at 04:27

2

With awk:

awk 'NR==FNR {a[NR]=$0; next} {print a[$0]}' data.txt addr.txt

NR==FNR {a[NR]=$0; next} creates an associative array a with keys being the record (line) number and values being the whole record, this would be applicable only for the first file (NR==FNR), which is data.txt. next makes awk to go to the next line without processing the record any further
{print a[$0]} prints the value from the array with keys being the currect file's (addr.txt) line (record) number

Example:

% cat addr.txt 
2
4
6
7
1
3
5

% cat data.txt 
1.000451451
2.000589214
3.117892278
4.479511994
5.484514874
6.784499874
7.021239396

% awk 'NR==FNR {a[NR]=$0; next} {print a[$0]}' data.txt addr.txt
2.000589214
4.479511994
6.784499874
7.021239396
1.000451451
3.117892278
5.484514874

edited Jun 09 '17 at 04:27

codeforester

39,467
16
112
140

answered Jun 09 '17 at 04:13

heemayl

39,294
7
70
76

2

Nice solution and great explanation. – codeforester Jun 09 '17 at 04:27
Very flexible use in awk. – CWLiu Jun 09 '17 at 05:34
nice quality answer with explanation for future use. PS :I didn't DV. – P.... Jun 09 '17 at 07:28

Chiheb Nexus · Answer 3 · 2017-06-09T05:00:12.040

0

You can do it, also, within Python, like this example:

with open("address_file", 'r') as f1, open("data_file", "r") as f2:
    data1 = f1.read().splitlines()
    data2 = f2.read().splitlines()

for k in data1:
    # Handle exceptions if there is any
    try:
        print(data2[int(k)-1])
    except Exception:
        pass

Edit: As suggested @heemayl, here is another solution using only one list:

with open("file1", 'r') as f1, open("file2", 'r') as f2:
    data = f2.read().splitlines()

    for k in f1.read().splitlines():
        print(data[int(k)-1])

Both will output:

2.000589214
4.479511994
6.784499874
7.021239396
1.000451451
3.117892278
5.484514874

edited Jun 09 '17 at 05:00

answered Jun 09 '17 at 04:35

Chiheb Nexus

9,104
4
30
43

2

You don't need to have two lists. Just create the list for datafile and iterate over the lines for the file with just line numbers. – heemayl Jun 09 '17 at 04:38
Yes, i know. But i think it's easy for the OP to catch whats happening inside the code. I do'nt think he had a good knowledge how to answer his question using Python. But still your comment is correct. – Chiheb Nexus Jun 09 '17 at 04:39

Pick up lines from a file based on line numbers in another file

3 Answers3