0

I am trying to grep the specific pattern /home/mytest/data from each line of 1.txt file and when the match is found then copy the complete line to 2.txt file. But result set in 2.txt file is not the expected result.

Code

i=/home/mytest/data
IFS=',' read -ra ADDR <<< "$File_Name"
Input=1.txt
for i in "${ADDR[@]}";
do
  j=0
  while IFS= read -r line;
  do
    j=$((j+1))
    Exists=$(echo $line | grep -w "$i" )
    if [[ ! -z "$Exists" ]]; then
      cp=$(echo $line >> 2.txt)
    fi
  done <"$Input"

Expected Result - 2.txt

# file: /home/mytest/data # owner: own # group: group-sm user::r-x group::rwx mask::rwx other::--- default:user::rwx default:group::r-x default:group:smr:rwx default:group:agm:r-x default:mask::rwx default:other::---

Actual Result - 2.txt

# file: /home/mytest/data # owner: own # group: group-sm user::r-x group::rwx mask::rwx other::--- default:user::rwx default:group::r-x default:group:smr:rwx default:group:agm:r-x default:mask::rwx default:other::---
# file: /home/mytest/datasr123 # owner: own # group: group-sm user::r-x group::rwx mask::rwx other::---

Grep version: GNU grep 2.20

Improve this question
10
  • What Unix are you using? And more importantly, what grep are you using? Commented Feb 2, 2020 at 7:12
  • 1
    Strictly speaking, the string does occur on both lines. Can we assume that the string that you are trying to match is always an absolute path? Commented Feb 2, 2020 at 7:15
  • what unix ?? what grep ?? not getting you Commented Feb 2, 2020 at 7:15
  • Are you running Linux? What distribution of Linux in that case? What is the output of grep --version? Commented Feb 2, 2020 at 7:16
  • yes. GNU grep 2.20 Commented Feb 2, 2020 at 7:17

1 Answer 1

Reset to default
1

The issue is that with the (non-standard) option -w, grep will require a word boundary before and after the pattern. A "word boundary" is a transition between a word and a non-word. A "word" is a string that contains word characters, like a, b, etc., while a non-word is a string that contains characters like space, /, ?, etc.

Since / is not a "word character", there can be no word boundary before it (unless it is immediately preceded by a word character). Some versions of grep would therefore not be able to match the string in your data, ever. Other versions would be able to perform the match.

Since you seem to want to match the path on a line where it is immediately preceded # file: , and since we know that the path is followed by  # owner: on the same line, we can use these strings as delimiters for the given path:

querypath=/home/mytest/data

grep -F "# file: $querypath # owner:" 1.txt >2.txt
Improve this answer
5
  • no its not working out Commented Feb 2, 2020 at 7:25
  • ya tried it and then only i posted the above comment Commented Feb 2, 2020 at 7:29
  • @user78873 Is your script file a DOS text file? Did you write it on a Windows machine? In that case, convert it to a Unix text file with dos2unix. Commented Feb 2, 2020 at 7:31
  • @Kusalananda : no its unix bash script only. Commented Feb 2, 2020 at 7:32
  • @user78873 Well, what can I say. If I copy and paste your text and run the commands, it works on OpenBSD (with both native and GNU grep), and it also works on CentOS and on Ubuntu. Commented Feb 2, 2020 at 7:33

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.