Shell Scripting

Shell Programming and Scripting
1

I'm trying to write a unix script to handle this function
we are in the ksh env, but force bash in scripting
I have three files as input and need to concatenate them onto one line at a time

For example: The first record of each file is
File 1 =a
File 2 =b
File 3 =c

Output should be a;b;c

Second line of each file is

File 1 =d
File 2 =e
File 3 =f

output should be d;e;f

I would like to eventually open this file in excel. That is why I added the delimiters.

2 
awk -F= '{ORS=NR%3?";":"RS"}' file > file.csv
3

not familiar with awk

can you tell me what this is doing or how do I represent my files

4 
cut -f2 -d"=" sample.txt | tr "\n" ";"

cut command gives the secong coloumn with delimiter "="

tr -- translates the \n to ;

5

Sorry I forgot to print the values:

awk -F= '{ORS=NR%3?";":RS}{print $2}' file > file.csv
6

I think I've misrepresented my input files. I have three files with data in them. I want to string the data in those files together on one line, so I tried a cat statement, but it just lists all the files end to end. I need unix to read one record from each file and string them together on one line with delimeters, then go to the next line on each input file and create a new line on the output file with second record. Each file has approx 10,000 lines on it. All the reconstructed records need to be on one output file in a csv format
sorry for the confusion or maybe I'm just confused on what you guys are showing me

7 
awk '{getline a < "file2";getline b < "file3";print $0,a,b}' OFS=";" file1 > file.csv
8

I just replaced your file names with my actual names and got some errors.

awk '{getline a < "scomp2p8_2010_03_01.txt";getline b < "scomp2p6_2010_03_01.txt";print $0,a,b}' OFS=";" scomp2p5_2010_03_01.txt > scomp2p9_2010_03_01.txt''

awk: syntax error near line 1
awk: illegal statement near line 1
awk: syntax error near line 1
awk: illegal statement near line 1

---------- Post updated at 03:36 PM ---------- Previous update was at 03:34 PM ----------

[/COLOR]that double quote at the end isn't actually part of the awk statement

9

Try nawk or /usr/xpg4/bin/awk on Solaris.

10

I got it to work with nawk thanks, but for some reason when I replace the hardcoded file names with variables. It won't accept RUNDATE variable.

RUNDATE="2010_03_24";

nawk '{getline a < "scomp2p8_${RUNDATE}.txt";getline b < "scomp2p6_${RUNDATE}.txt";print $0,a,b}' OFS=";" scomp2p5_${RUNDATE}.txt > scomp2p9_2010_03_24.txt

Also the usr/xpg4/bin/awk seems to exist on the box, but this statement won't work. Just curious why?

usr/xpg4/bin/awk '{getline a < "scomp2p8_2010_03_01.txt";getline b < "scomp2p6_2010_03_01.txt";print $0,a,b}' OFS=";" scomp2p5_2010_03_01.txt > scomp2p9_2010_03_01.txt

bash: usr/xpg4/bin/awk: No such file or directory

thanks for all the help

11

With shell variables the command should be:

nawk -v rd=${RUNDATE} '{getline a < "scomp2p8_"rd".txt";getline b < "scomp2p6_"rd".txt";print $0,a,b}' OFS=";" scomp2p5_${RUNDATE}.txt > scomp2p9_2010_03_24.txt

You forgot the slash before usr.

12

... deleted ... redundant solution...

13

still doesn't recognize the rd variable. doesn't read in the files within the double quotes.

14 
nawk -v f1="scomp2p8_"${RUNDATE}".txt" -v f2="scomp2p6_"$(RUNDATE)".txt" '{getline a < f1;getline b < f2;print $0,a,b}' OFS=";" scomp2p5_${RUNDATE}.txt > scomp2p9_2010_03_24.txt
15

Thanks that worked. I don't know who came up with this idea of sharing the knowledge, but I love it.

Thanks big time

Hopefully sometime I can return the favor