regexp-cut

Uses awk to provide cut like syntax for field extraction. The command name is rcut.

⚠️ ⚠️ Work under construction!

Motivation

cut's syntax is handy for many field extraction problems. But it doesn't allow multi-character or regexp delimiters. So, this project aims to provide cut like syntax for those cases. Currently uses mawk in a bash script.

ℹ️ Note that rcut isn't feature compatible or a replacement for the cut command. rcut helps when you need features like regexp field separator.

Features

Default field separation is same as awk
Both input (-d) and output (-o) field separators can be multiple characters
Input field separator can use regular expressions
- this script uses mawk by default
- you can change it to gawk for better regexp support with -g option
If input field separator is a single character, output field separator will also be this same character
Fixed string input field separator can be enabled by using the -F option
- if -o is not used, value passed to the -d option will be set as the output field separator
Field range can be specified by using - separator (same as cut)
- - by itself means all the fields (this is also the default if -f option isn't used at all)
- if start of the range isn't given, default is 1
- if end of the range isn't given, default is last field of a line
Negative indexing is allowed if you use -n option
- -1 means the last field, -2 means the second-last field and so on
- you'll have to use : to specify field ranges
Multiple fields and ranges can be separated using , character (same as cut)
Unlike cut, order matters with the -f option and field/range duplication is also allowed
- this assumes -c (complement) is not active
Using -c option will print all the fields in the same order as input except the fields specified by -f option
Using -s option will suppress lines not matching the input field separator
Minimum field number is forced to be 1
Maximum field number is forced to be last field of a line

⚠️ ⚠️ Work under construction!

Examples

$ cat spaces.txt
   1 2	3  
x y z
 i          j 		k	

# by default, it uses awk's space/tab field separation and trimming
# unlike cut, order matters
$ rcut -f3,1 spaces.txt
3 1
z x
k i

# multi-character delimiter
$ echo 'apple:-:fig:-:guava' | rcut -d:-: -f2
fig

# regexp delimiter
$ echo 'Sample123string42with777numbers' | rcut -d'[0-9]+' -f1,4
Sample numbers

# fixed string delimiter
$ echo '123)(%)*#^&(*@#.[](\\){1}\xyz' | rcut -Fd')(%)*#^&(*@#.[](\\){1}\' -f1,2 -o,
123,xyz

# multiple ranges can be specified, order matters
$ printf '1 2 3 4 5\na b c d e\n' | rcut -f2-3,5,1,2-4
2 3 5 1 2 3 4
b c e a b c d

# last field
$ printf 'apple ball cat\n1 2 3 4 5' | rcut -nf-1
cat
5

# except last two fields
$ printf 'apple ball cat\n1 2 3 4 5' | rcut -cnf-2:
apple
1 2 3

# suppress lines without input field delimiter
$ printf '1,2,3,4\nhello\na,b,c\n' | rcut -sd, -f2
2
b

# -g option will switch to gawk
$ echo '1aa2aa3' | rcut -gd'a{2}' -f2
2

See Examples.md for many more examples.

Tests

You can use script.awk to check if all the example code snippets are working as expected.

$ cd examples/
$ awk -f script.awk Examples.md

TODO

Step value other than 1 for field range
What to do if start of the range is greater than end?
And possibly more...

Similar tools

hck — close to drop in replacement for cut that can use a regex delimiter, works on compressed files, etc
choose — negative indexing, regexp based delimiters, etc

Contributing

Please open an issue for typos/bugs/suggestions/etc
Even for pull requests, open an issue for discussion before submitting PRs
In case you need to reach me, mail me at echo 'bGVhcm5ieWV4YW1wbGUubmV0QGdtYWlsLmNvbQo=' | base64 --decode or send a DM via twitter

License

This project is licensed under MIT, see LICENSE file for details.

learnbyexample/regexp-cut