2

I have a sed command that I've been using to format the date in datasets. I've run into an issue and I'm sure it's just my lack of understanding on the syntax but was wondering if someone could help clarify what's going on.

input:

26,082619,1300,1,2,0
26,082619,1400,332,8,190
26,082619,1500,442,10,118
26,082619,1600,530,151,12
26,082619,1700,534,164,0
26,082619,1800,538,176,0

command

sed -E "s/,(.{2})(.{2})(.{2}),/,\1\/\2\/20\3,/g" input.csv > output.csv

output (note line 3)

26,08/26/2019,1300,1,2,0
26,08/26/2019,1400,332,8,190
26,08/26/2019,1500,44/2,/2010,118
26,08/26/2019,1600,530,151,12
26,08/26/2019,1700,534,164,0
26,08/26/2019,1800,538,176,0

expected output

26,08/26/2019,1300,1,2,0
26,08/26/2019,1400,332,8,190
26,08/26/2019,1500,442,10,118
26,08/26/2019,1600,530,151,12
26,08/26/2019,1700,534,164,0
26,08/26/2019,1800,538,176,0
Improve this question
1
  • Try sed -E 's/,([0-9]{2})([0-9]{2})([0-9]{2}),/,\1\/\2\/20\3,/' input.csv Commented Jan 11, 2021 at 18:27

1 Answer 1

Reset to default
3

The issue is two fold:

  1. in a line-oriented context, . matches any character - including ,

  2. you used the g flag, which tells sed to make as many substitutions as possible in a line

so it is matching all non-overlapping sequences of six characters that are bounded by commas.

For your example data, it should be sufficient to drop the g flag so that only the first matching sequence is substituted - however you should also consider replacing . by something more specific such as [^,] (any character except comma) or [0-9] (any of the Latin decimal digits).

Improve this answer
1
  • Very helpful, thank you! You are correct, removal of g was sufficient in my case. I will explore increasing the specificity as well to avoid other potential issues with this and my other cmds. Thanks again! Commented Jan 11, 2021 at 18:03

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.