Mastering the Sort Command in Linux: A Comprehensive Guide

The sort command in Linux is a essential command-line utility used to reorder lines in text files or streams. By default, sort processes data lexicographically (alphabetically, character-by-character). However, it provides options to handle numerical values, reverse ordering, field/column-based sorting, human-readable file sizes, and duplicate removal.

1. Default Lexicographical Sorting

When invoked without any flags, sort reads input line-by-line and orders the content alphabetically based on ASCII values.

Basic Syntax

sort [OPTION]... [FILE]...

Example Walkthrough

Consider a file named foods.txt with unordered items:

grapes
cheese
apple
zucchini

Running the default sort command:

sort foods.txt

Output:

apple
cheese
grapes
zucchini

Note: As indicated in the manual page (man sort), sort can accept multiple space-separated files directly or read target file paths from a text list using the -f / --files-from option.

2. Reversing Sort Order (-r)

To invert the standard sort sequence—arranging text from Z to A or numbers from highest to lowest—use the -r (reverse) flag.

Command Execution

sort -r foods.txt

Output:

zucchini
grapes
cheese
apple

3. String vs. Numeric Sorting (-n)

A common pitfall occurs when sorting numbers using default lexicographical mode. Because string comparison evaluates input character-by-character from left to right, numerical sequences produce counter-intuitive results.

The Naive Numeric Problem

Consider a file named numbers.txt:

09
10
2
3

Executing a standard sort:

sort numbers.txt

Output (Incorrect Order):

09
10
2
3

Why this fails: Default string comparison compares the first character of each line. '0' (in 09) comes before '1' (in 10), and '1' comes before '2' or '3'. It treats the data as plain text strings rather than arithmetic quantities.

The Solution: Numerical Evaluation (-n)

To force Linux to evaluate lines based on integer numerical values, append the -n flag:

sort -n numbers.txt

Output (Correct Numerical Order):

2
3
09
10

To sort numbers in descending numerical order, combine the numeric and reverse flags:

sort -nr numbers.txt

4. Capturing Output for Sorting Workflows

Before mastering pipeline redirection (|), intermediate terminal outputs must be saved to a temporary text file using output redirection (>) to perform multi-stage sorting operations.

Real-World Use Case: Sorting Directory Contents

To sort directory contents generated by ls -lh:

1. Redirect the command output into a storage file:
   BASH

   Copy

   ls -lh > files.txt
   

2. Verify the stored contents:
   BASH

   Copy

   cat files.txt
   

Running a basic sort files.txt at this point will fail to order by size because it defaults to evaluating column 1 (file permission strings such as -rw-r--r-- or drwxr-xr-x).

5. Column-Based Field Sorting (-k)

Directory listings and CSV-like data consist of multiple whitespace-separated columns. The -k (key) flag allows targeting a specific column field for sorting.

Analyzing Field Positions in ls -lh

-rw-r--r-- 1 user group 148M Oct 24 10:00 bigfile.zip
^          ^ ^    ^     ^
Col 1      C2 C3  Col 4 Column 5 (File Size)

1. Column 1: File permissions (-rw-r--r--)
2. Column 2: Hard link count (1)
3. Column 3: Owner (user)
4. Column 4: Group (group)
5. Column 5: File size (148M)

To sort specifically by the 5th column, specify -k 5:

sort -k 5 files.txt

6. Sorting Human-Readable File Sizes (-h)

Standard numerical sorting (-n) breaks when file sizes contain unit suffixes like K (Kilobytes), M (Megabytes), or G (Gigabytes). The system requires the human-numeric sort flag (-h) to parse these suffixes correctly.

Sorting Files by Size (Ascending & Descending)

To sort by Column 5 taking human-readable units (148M, 2K, 500B) into account:

sort -k 5h files.txt

To find the largest files at the top of the list, append r to reverse the output:

sort -k 5hr files.txt

7. Deduplicating Output (-u)

When processing text containing duplicate rows, the -u (unique) flag sorts the file while discarding duplicate entries.

Example Walkthrough

Consider fruits.txt:

apple
banana
kiwi
apple

Sorting with deduplication:

sort -u fruits.txt

Output:

apple
banana
kiwi

8. Complete Execution Flow Walkthrough

The following workflow illustrates how file size extraction, column targeting, human unit parsing, and reverse sorting operate sequentially:

9. Command Flag Summary Matrix

Summary and Key Takeaways

• Default Lexicographical Behavior: By default, sort evaluates lines character-by-character from left to right based on ASCII values.
• Character Sorting Pitfalls: Plain sort fails on numerical lists (e.g., placing 10 before 2). Always use -n for numerical data.
• Reversing Output: Pass -r to flip the output order from ascending (A-Z, lowest-to-highest) to descending (Z-A, highest-to-lowest).
• Column-Level Precision: Use -k [COLUMN_NUMBER] to isolate specific fields in whitespace-delimited outputs (such as ls -l columns).
• Handling Storage Units: Use -h when sorting file sizes containing human-readable metric suffixes (K, M, G).
• Combining Options: Flags can be concatenated for specific criteria—such as -k 5hr to sort by the 5th column numerically with unit awareness in reverse order.
• In-Line Deduplication: Use -u to automatically filter out duplicate lines while sorting.