cut

What it does

cut extracts sections from each line of input and writes them to standard output. It operates in one of two modes: by delimited field number (-f, using -d to set the separator) or by raw character position (-c). It's a lightweight tool for pulling a column out of structured text like /etc/passwd, CSV exports, or fixed-width command output, without needing the full power of awk.

Beginner examples

• cut -d':' -f1 /etc/passwd — print the first colon-delimited field (usernames)
• cut -d',' -f2 data.csv — print the second comma-delimited field
• cut -c1-5 file.txt — print the first 5 characters of each line
• cut -f1,3 data.tsv — print the first and third tab-delimited fields (default delimiter)

cut -d':' -f1,3 /etc/passwd

Advanced examples

• Extract a range of fields: cut -d',' -f2-4 sales.csv
• Print everything except a field using an open-ended range: cut -d':' -f2- /etc/passwd
• Combine with sort to build a unique list of usernames: cut -d':' -f1 /etc/passwd | sort
• Use a different output delimiter than the input one (GNU cut): cut -d',' -f1,2 --output-delimiter='|' data.csv
• Pull a fixed-width column from ps output: ps aux | cut -c1-10

awk -F',' '{print NF}' data.csv | sort -u   # sanity-check field count before cutting

Common mistakes

• Assuming cut handles quoted CSV fields correctly — it splits naively on the delimiter character, so a comma inside a quoted field will throw off the field count.
• Forgetting -d and getting no useful output because the default delimiter is TAB, not comma or space.
• Using multiple spaces as a delimiter with cut -d' ' and getting empty fields, since cut doesn't collapse repeated delimiters the way awk does by default.
• Expecting cut -f3,1 to reorder output — cut always prints fields in original left-to-right order regardless of the order listed.

Tips

• For space-separated data with inconsistent spacing (like ps aux), prefer awk '{print $2}' over cut, since awk collapses repeated whitespace automatically.
• Use cut -s when processing mixed input so lines without the delimiter are dropped instead of printed as-is.
• Check field counts first with awk -F',' '{print NF}' file | sort -u before trusting cut -f output on messy CSV data.

Best practices

• Reach for cut only on genuinely simple, single-character-delimited data; switch to awk once you need quoting, multiple delimiters, or field reordering.
• Always quote the delimiter argument (-d',') to avoid the shell interpreting special characters like ; or |.
• When scripting, pin down the exact field number with a quick manual check (head -1 file | cut -d',' -f2) rather than guessing from eyeballing the file.

Real-world use cases

• Building a list of system usernames for an audit: cut -d':' -f1 /etc/passwd | sort.
• Extracting the PID column from ps output to feed into kill: ps -ef | grep myapp | cut -c10-15.
• Pulling a single column out of a CSV export for a quick calculation: cut -d',' -f5 sales.csv | paste -sd+ | bc.

Common interview questions

• What's the difference between cut -c and cut -f? -c selects by raw character position regardless of content; -f splits the line into fields on a delimiter (set with -d) and selects by field number.
• Why might cut give wrong results on a CSV file? cut splits naively on the delimiter character, so commas inside quoted fields break the field alignment; a CSV-aware tool or awk with proper parsing handles that correctly.
• How do you select all fields from the third one onward? cut -d',' -f3- — an open-ended range with no upper bound selects from field 3 to the end of the line.