[Suggestion] Add a tail command

[apcamargo]

As far as I know, there's no way to easily inspect the last N records of a .csv file in a way that includes the header. I assume this could be achieved by reading the first line (the header) and skip to the last N lines of the file.

[tetedange13]

Hi @apcamargo ,

I managed to do this with awk and tac (inspired from "awk Solution" of this answer : https://unix.stackexchange.com/a/139099)

Input CSV

==> test.csv <==
id,age,gender
1,39,M
2,25,M
3,36,F
4,45,F

One-liner

tac test.csv | awk -v nbr=2 'NR<=nbr { prev=$0; print } END { if($0!=prev){print} }' | tac

Output (pretty)

id   age   gender
--   ---   ------
3    36    F
4    45    F

Explanations

• "N last rows" in your question <--> nbr awk variable defined
• Mandatory to store previous line (-> prev=$0 in above cmd) and test for equality before printing last line => To avoid double printting header if nbr > nb_records
  

Hope this helps ! Have a nice day, Felix.

[shenwei356]

Thank you @tetedange13