0

I want to parse data of particular columns from several files(aprilPlate.txt, mayPlate.txt, junePlate.txt, julyPlate.txt, augustPlate.txt) using For loop.

The data of input files (aprilPlate.txt, mayPlate.txt, junePlate.txt, julyPlate.txt, augustPlate.txt) look like:

Incl Cal Ps Name Q Con Std Status True 255 A1 Sample 1 35.86 0 True 255 A2 Sample 2 36.06 0 True 255 A3 Sample 3 17.45 0 True 255 A4 Sample 4 17.56 0 True 255 A5 Sample 5 17.55 0 True 255 A6 Sample 6 40.00 0 True 255 A7 Sample 7 36.38 0 True 255 A8 Sample 8 27.98 0 True 255 A9 Sample 9 27.95 0 True 255 A10 Sample 10 28.19 0 True 255 A11 Sample 11 36.93 0 True 255 A12 Sample 12 37.74 0 True 255 A13 Sample 13 17.88 0 True 255 A14 Sample 14 17.82 0 True 255 A15 Sample 15 17.90 0 . . .

By using FOR loop, I able to get the task done. However, in order to get desire data from all files, I have to modify the file name in the script (below) manually. 

#!/bin/bash # parse the data of desire columns from target file # rename the column name # redirect the stdoutput to a text file for z in A B; do for i in 3 4 5 13 14 15; do grep $z$i aprilPlate.txt | awk -F "\t" '{print $3 "\t" $5}' | sed -e 's/A[3-5]/st_SWC/g;s/A[1][0-9]/st_SWD/g;s/B[3-5]/st_TZC/g;s/B[1][0-9]/st_TZD/g;' >> stone.txt; done; done for z in E F; do for i in 8 9 10 18 19 20; do grep $z$i aprilPlate.txt | awk -F "\t" '{print $3 "\t" $5}' | sed -e 's/E[8-9]\|E[1][0]/su_SWC/g;s/E[1][0-9]\|E[2][0]/su_SWD/g;s/F[8-9]\|F[1][0]/su_TZC/g;s/F[1][0-9]\|F[2][0]/su_TZD/g;' >> suy.txt; done; done paste -d'\t' stone.txt suy.txt >> aprilPlate.data.txt
  • My apologies, I made a mistake in my previous coding. The correction was done.

The output of parsed data file should looks like:

st_SWC 17.45 su_SWC 28.85 st_SWC 17.56 su_SWC 28.79 st_SWC 17.55 su_SWC 28.82 st_SWD 17.88 su_SWD 29.24 st_SWD 17.82 su_SWD 29.18 st_SWD 17.90 su_SWD 29.23 st_TZC 18.06 su_TZC 25.99 st_TZC 18.09 su_TZC 25.98 st_TZC 18.13 su_TZC 26.02 st_TZD 17.75 su_TZD 25.00 st_TZD 17.70 su_TZD 25.01 st_TZD 17.69 su_TZD 24.98

I would like to ask, is there anyway I can have the files as 3rd variable in the script? Other solutions are welcome.

Improve this question

2 Answers 2

Reset to default
0

Something like will help you

for file in aprilPlate.txt mayPlate.txt junePlate.txt julyPlate.txt augustPlate.txt; do for z in A B; do for i in 3 4 5 13 14 15; do grep $z$i $file | awk -F "\t" '{print $3 "\t" $5}' | sed -e 's/A[3-5]/SWC/g;s/A[1][0-9]/SWD/g;s/B[3-5]/TZC/g;s/B[1][0-9]/TZD/g;' >> stone.txt; done; done <snip> done
Improve this answer
1
  • Of course w/o commas between filenames (my mistake, edited) Commented Apr 30, 2015 at 10:10
0

Not sure exactly what you are trying to do since you have no input or output examples, but the following should work.

#!/bin/bash awksrc='BEGIN{FS=OFS="\t"} { gsub(/A[345]|E[89]|E10/, "SWC"); gsub(/A1[0-9]|E1[1-9]|E20/, "SWD"); gsub(/B[345]|F[89]|F10/, "TZC"); gsub(/B1[0-9]|F1[1-9]|F20/, "TZD"); } /SW[CD]{print $3, $5 >"stone.txt"} /TZ[CD]{print $3, $5 >"suy.txt"}' for file in aprilPlate.txt mayPlate.txt junePlate.txt julyPlate.txt ... do awk "$awkscr" $file >/dev/null paste -d'\t' stone.txt suy.txt >> ${file%.txt}.data.txt done

The ${file%.txt} gets the base name, without the extension to be able to add the .data.txt extension.

This reduces the number of shell loops to just one, and processes each file once. The output of the awk script is unneeded since the files are getting written to inside it.

Again, without example input, it's hard to say if this is correct for your purposes.

Improve this answer
1
  • I have the input and output examples included in my question now. Thanks for your suggestion. Commented Apr 30, 2015 at 7:58

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.