cut / paste / join
| Since: | All Linux | |
|---|---|---|
| macOS(2001 Cheetah) | ||
| Bash 1.0(1989) |
cut extracts specific columns or character ranges from each line of a file, paste merges lines from multiple files side by side, and join combines two files based on a common field as the key.
Syntax
cut [options] [file...] paste [options] [file...] join [options] file1 file2
Options
| Command / Option | Description |
|---|---|
| cut -d delimiter -f field | Splits fields by the delimiter and extracts the specified field number. |
| cut -f 1,3 | Extracts the 1st and 3rd fields. |
| cut -f 1-3 | Extracts fields 1 through 3. |
| cut -c 1-5 | Extracts characters 1 through 5 from each line. |
| paste file1 file2 | Merges corresponding lines from two files, separated by a tab. |
| paste -d ',' file1 file2 | Merges lines using the specified delimiter. |
| paste -s file | Joins all lines of a file into a single tab-separated line. |
| join file1 file2 | Performs an inner join using the first column of each file as the key (files must be pre-sorted). |
| join -1 N -2 M file1 file2 | Joins using column N of file1 and column M of file2 as the key. |
| join -a 1 file1 file2 | Also outputs unmatched lines from file1 (left outer join). |
Sample Code
The following files are used in the examples below.
data.csv
user_a,28,Tokyo,engineer user_b,15,Osaka,designer user_c,15,Nagoya,manager
names.txt
user_a user_b user_c
scores.txt
85 92 78
users.txt (sorted by ID)
001 user_a 002 user_b 003 user_c
roles.txt (sorted by ID)
001 engineer 002 designer 003 manager
Use cut -d -f to specify a delimiter and field number to extract a column.
cut -d',' -f1,3 data.csv user_a,Tokyo user_b,Osaka user_c,Nagoya
Use cut -f 1-2 to extract a range of fields.
cut -d',' -f1-2 data.csv user_a,28 user_b,15 user_c,15
Use cut -c to extract characters by position.
cut -c1-5 data.csv user_ user_ user_
Use paste to merge two files side by side (tab-separated).
paste names.txt scores.txt user_a 85 user_b 92 user_c 78
Use paste -d to specify a custom delimiter.
paste -d',' names.txt scores.txt user_a,85 user_b,92 user_c,78
Use paste -s to join all lines of a file into a single line.
paste -s names.txt user_a user_b user_c
Use join to combine two files based on a shared field (files must be sorted in advance).
join users.txt roles.txt 001 user_a engineer 002 user_b designer 003 user_c manager
Common Mistakes
Common Mistake 1: Confusing -f (field) and -c (character)
-f selects by field (delimiter-separated) and -c selects by character position. Using the wrong option produces unexpected output.
echo "id:name:age" | cut -c2 d （retrieves the 2nd character, not the 2nd field）
Use -f with -d to extract by field.
echo "id:name:age" | cut -d: -f2 name
Common Mistake 2: cut cannot handle CSV fields that contain the delimiter
If a field is quoted and contains the delimiter, cut splits at every occurrence regardless of quoting.
echo '"data_a, location",28' | cut -d, -f1 "data_a （split inside the quoted field）
Use awk or a dedicated CSV tool for fields that may contain the delimiter.
Notes
'$ cut' is convenient for simple field extraction, but it cannot handle cases where a field contains the delimiter character (such as quoted CSV). For more complex cases like these, use awk.
For counting or sorting fields, see also sort.
If you find any errors or copyright issues, please contact us.
シェアする