Split a CSV file using AWK, while reading from a different file for naming the output files

Question

I have a CSV file as the following:

group1, item1
group1, item2
group2, item3
group1, item4
.....

I have managed to split this file by groups to separate csv files (group1.csv.dat, group2.csv.dat etc.). Each file contains all items belonging to a specific group.

group1.csv.dat:

item1, true
item2, true
item4, true
.....

group2.csv.dat:

item3, true
.....

I have used the following AWK:

awk -F, '{print $2",true" > $1".csv.dat"}' file1

Now, I have a second file (let's say file 2), as follows:

group1, GRFS+NC, 4
group2, GRTU+NC, 6 
....

How can I read this file using AWK in order to name the files created in the first step as GRFS4.csv.dat, GRTU6.csv.dat instead of group1.csv.dat, group2.csv.dat? Preferably, I would like to incorporate processing into the first step. Many thanks...

Vijay · Accepted Answer · 2014-01-13 11:04:30Z

1

awk -F, '{split($2,a,"+");print $2",true" > a[1]""$3".csv.dat"}' file2

answered Jan 13, 2014 at 11:04

Vijay

67.7k94 gold badges238 silver badges327 bronze badges

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

3 Comments

fedorqui Over a year ago

Be careful, trailing spaces will appear in file names.

Spyros Agrafiotis Over a year ago

Thank you for your answer. This will produce files based on information from file2 only. What I would like to do is essentially rename the files produced by the AWK in my question (e.g. take group1.csv.dat, lookup for "group1" in file2 and rename group1.csv.dat to GRFS4.csv.dat by concatenating the relevant fields).

Ed Morton Over a year ago

As discussed many times previously, un-parenthesized concatenation on the right side of output redirection produces undefined behavior. You need to put brackets around it. Also, the null string ("") between a[1] and $3 is doing absolutely nothing.

Ed Morton · Accepted Answer · 2014-01-13 15:48:40Z

0

You need something like this, untested:

awk '
NR==FNR{ name[$1] = $3 $6 ".csv.dat"; next }
{ print $2 ",true" > name[$1] }
' FS='[, +]' file2 FS=',' file1

Just count the fields in file2 to make sure $3 and $6 are the correct fields. Add a debugging for loop to print them all to see if you're not sure.

answered Jan 13, 2014 at 15:48

Ed Morton

209k18 gold badges90 silver badges212 bronze badges

Collectives™ on Stack Overflow

Split a CSV file using AWK, while reading from a different file for naming the output files

2 Answers 2

3 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related