how to replace a specific char occurrences in string after a given substring

Question

I have a string Contain key=value format separated by # I am trying to replace the '=' char occurrences with ':' in the value of TITLE using BASH script.

"ID=21566#OS=Linux#TARGET_END=Synchronica#DEPENDENCY=Independent#AUTOMATION_OS=Linux#AUTOMATION_TOOL=JSystem#TITLE=Session tracking. "DL Started" Status Reported.Level=none"

later on i am parsing this string to execute the eval operation

eval $(echo $test_line | sed 's/"//g' | tr '#' '\n' | tr ' ' '_' | sed 's/=/="/g' | sed 's/$/"/g')

When the sed 's/=/="/g' section will also change ..Level=none to Level="none This leads to

eval: line 52: unexpected EOF while looking for matching `"'

What will be right replace bash command to replace my string ?

Just remove the /g modifier from sed 's/=/="/g'.

choroba
– choroba

2019-11-03 18:56:21 +00:00
Commented Nov 3, 2019 at 18:56 — choroba
– choroba, Commented Nov 3, 2019 at 18:56

dash-o · Accepted Answer · 2019-11-03 20:13:05Z

1

As an alternative, consider pure-bash solution to bring the variables into bash, avoiding the (risky) eval.

IFS=# read -a kv <<<"ID=21566#OS=Linux#TARGET_END=Synchronica#..."

for kvp in "${kv[@]}" ; do
    declare "$kvp"
done

answered Nov 3, 2019 at 20:13

dash-o

14.6k1 gold badge14 silver badges43 bronze badges

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

2 Comments

Boris Katz Over a year ago

Thanks for the suggestion. Why the eval is risky?

dash-o Over a year ago

I'm assuming the input comes from external system. This open your script to shell injection. e.g. , if the input will contain " ; rm -rf / (quotes, followed by any command) - it will execute that command - in the rm - it can wipe out part of your system.

Boris Katz · Accepted Answer · 2019-11-03 20:11:24Z

0

I found the way to solve it.
I will add sed 's/=/:/8g' to my eval command. It will replace 8th to nth occurrences of '='.
The action will only effect the value of TITLE as expected.

 eval $(echo $test_line | sed 's/=/:/8g' | sed 's/"/"/g' | tr '#' '\n' | tr ' ' '_' | sed 's/=/="/g' | sed 's/$/"/g')

answered Nov 3, 2019 at 20:11

Boris Katz

217 bronze badges

Comments

Mika Feiler · Accepted Answer · 2019-11-04 22:30:37Z

0

I did it like this :

echo '"ID=21566#OS:Linux#TARGET_END:Synchronica#DEPENDENCY:Independent#AUTOMATION_OS:Linux#AUTOMATION_TOOL:JSystem#TITLE:Session tracking. "DL Started" Status Reported.Level=none"' \
 | 
sed -E 's/(#)?([A-Z_]+)(=)/\1\2:/g'

Let me know if it works for you.

edited Nov 4, 2019 at 22:30

Mika Feiler

5385 silver badges23 bronze badges

answered Nov 3, 2019 at 20:13

Matias Barrios

5,0864 gold badges31 silver badges71 bronze badges

Collectives™ on Stack Overflow

how to replace a specific char occurrences in string after a given substring

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related