What is the best way to print line number as the first column

[san-r]

This is not an issue but question. As of now, csvtk sub-commands supporting the flag -n or --line-number to "print line number as the first column ("n")" are as under:

grep
filter
filter2
sample

One way to print line number as the first column at the end of workflow is as under:

csvtk sub-command ... |^
csvtk another-sub-command ... |^
csvtk grep -n -f1 -r -p ".*" |^
csvtk pretty

I'm wondering if there a better (more processor and RAM friendly) way to to print line numbers than using
csvtk grep -n -f1 -r -p ".*"

[shenwei356]

I don't think making all commands support this is a good idea.

Maybe adding a new command cat with a -n option to print the line numbers.

[san-r]

Using a sub-command like grep in the manner shown above just to print line numbers is certainly inefficient. And yes, only one sub-command cat supporting -n to print line numbers will be excellent. One more thing I've noticed is that cat sub-command does not support gzipped input unlike other sub-commands. It would be nice for it to do so.

[shenwei356]

oh, @bsipos have already created acat, can we rename it to watch? like seqkit watch?

[shenwei356]

In the next version, you can easily use

$ csvtk cut -f 1- -Z testdata/names.csv 
row,id,first_name,last_name,username
1,11,Rob,Pike,rob
2,2,Ken,Thompson,ken
3,4,Robert,Griesemer,gri
4,1,Robert,Thompson,abc
5,NA,Robert,Abel,123

$ csvtk filter2 -f '$id>2' -n  testdata/names.csv 
row,id,first_name,last_name,username
1,11,Rob,Pike,rob
3,4,Robert,Griesemer,gri

Related changes:

• add a new global flag -Z/--show-row-number, supported commands: cut, csv2tab, csv2xlsx, tab2csv, pretty.
• the colum name of row number changes from "n" to "row": csv2xlsx, csv2tab, cut, filter, filter2, grep, pretty, sample, tab2csv.