awk : merge multiple rows into one row per first column record value

Question

I have one text file which contains some records like below,

100,a
b
c
101,d,e
f
102,g
103,h
104,i
j
k

so,some rows start with number,some rows start with string ,and I want to merge rows which rows are order by number and merge rows like below:

100,a,b,c
101,d,e,f
102,g
103,h
104,i,j,k

How can I use awk to do this ? Thanks

SO is not a code writing service; you're expected to show your effort (eg, research, code); consider reviewing how do I ask a good question and then come back and update the question accordingly; in particular, show the code you've tried and the (wrong) output generated by your code — markp-fuso
– markp-fuso, Commented Jul 22, 2022 at 15:57

JNevill · Accepted Answer · 2022-07-22 15:23:50Z

1

You can do something like:

awk '/^[0-9]/{if(buf){print buf};buf=$0}/^[a-zA-Z]/{buf=buf","$0}END{print buf}' yourfile.txt

This will

If so then it will print out what is stored in variable buf if that variable has some value in it if (buf){print buf}
It will then reset the variable buf to the value of the current line buf=$0

Then it will add the current line to the value in the variable buf with a comma separator buf=buf","$0

Finally when it reaches the end of the file, it prints out whatever is left in the buf variable. END{print buf}

answered Jul 22, 2022 at 15:23

JNevill

50.6k4 gold badges46 silver badges72 bronze badges

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

1 Comment

I didn't think about it that in depth @EdMorton Is that a concern from OPs input file or some other awk eccentricity I'm not taking into account?