How to Trim Whitespace from String in Bash

Quick Answer: Trim Whitespace from a String

To remove leading and trailing whitespace from a string in Bash, use xargs: echo "$text" | xargs. For a pure-Bash solution, use parameter expansion with regex: ${text//[[:space:]]/ }. Both are simple and effective for cleaning user input.

Quick Comparison: Whitespace Trimming Methods

Method	Speed	Best For	Purity
xargs	Very fast	Simple trimming	External command
Parameter expansion	Very fast	Pure Bash	Built-in only
sed	Fast	Complex patterns	External command
tr	Very fast	Character removal	External command
printf %s	Very fast	Formatting	Built-in only

Bottom line: Use xargs for simplicity, use parameter expansion for pure Bash.

---

Remove leading and trailing whitespace from strings in Bash. Whitespace trimming is essential for input validation, data cleaning, and string processing in many scripts.

Method 1: Using xargs (Simplest)

The xargs method is the simplest and most readable, though it uses an external command.

text="  Hello World  "
trimmed=$(echo "$text" | xargs)
echo "[$trimmed]"  # Output: [Hello World]

# Works with any amount of whitespace
text="    text with     multiple   spaces    "
trimmed=$(echo "$text" | xargs)
echo "[$trimmed]"  # Output: [text with multiple spaces]

Example with variables:

$ text="  test  "
$ trimmed=$(echo "$text" | xargs)
$ echo "[$trimmed]"
[test]

Method 2: Using Parameter Expansion (Pure Bash)

No external commands needed, but syntax is complex.

# Remove leading and trailing whitespace
text="  Hello World  "
trimmed=${text#${text%%[![:space:]]*}}
trimmed=${trimmed%${trimmed##*[![:space:]]}}
echo "[$trimmed]"  # Output: [Hello World]

Breaking it down:

# Step 1: Remove leading whitespace
trimmed=${text#${text%%[![:space:]]*}}
# ${text%%[![:space:]]*}  = "  " (all leading spaces)
# ${text#...}            = remove that prefix

# Step 2: Remove trailing whitespace
trimmed=${trimmed%${trimmed##*[![:space:]]}}
# ${trimmed##*[![:space:]]} = "  " (all trailing spaces)
# ${trimmed%...}           = remove that suffix

Method 3: Using sed

The sed command is powerful for pattern-based trimming.

# Trim leading and trailing whitespace
text="  Hello World  "
trimmed=$(echo "$text" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//')
echo "[$trimmed]"  # Output: [Hello World]

# Alternative pattern (simpler but less portable)
trimmed=$(echo "$text" | sed 's/^ *//;s/ *$//')

# For tabs and spaces
trimmed=$(echo "$text" | sed 's/^[ \t]*//;s/[ \t]*$//')

Example:

$ text="   test string   "
$ echo "$text" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//'
test string

Method 4: Using awk

awk can trim whitespace with a simple one-liner.

# Trim both leading and trailing
text="  Hello World  "
trimmed=$(echo "$text" | awk '{$1=$1};1')
echo "[$trimmed]"  # Output: [Hello World]

# Explanation:
# $1=$1 forces awk to rebuild $0
# This removes leading/trailing spaces
# ;1 ensures the line is printed

Method 5: Remove Only Leading Spaces

Sometimes you need to trim only from one side.

text="  Hello World  "

# Remove only leading whitespace
trimmed=${text#${text%%[^ ]*}}
echo "[$trimmed]"  # Output: [Hello World  ]

# Using sed
trimmed=$(echo "$text" | sed 's/^[[:space:]]*//')
echo "[$trimmed]"  # Output: [Hello World  ]

# Using awk
trimmed=$(echo "$text" | awk '{sub(/^[[:space:]]+/, ""); print}')

Method 6: Remove Only Trailing Spaces

text="  Hello World  "

# Remove only trailing whitespace
trimmed=${text%${text##*[^ ]}}
echo "[$trimmed]"  # Output: [  Hello World]

# Using sed
trimmed=$(echo "$text" | sed 's/[[:space:]]*$//')
echo "[$trimmed]"  # Output: [  Hello World]

# Using awk
trimmed=$(echo "$text" | awk '{sub(/[[:space:]]+$/, ""); print}')

Practical Examples

Example 1: Trim User Input

#!/bin/bash

# Read user input and trim
read -p "Enter your name: " name
name=$(echo "$name" | xargs)  # Trim whitespace

read -p "Enter your email: " email
email=$(echo "$email" | xargs)

echo "Hello, $name ($email)"

Output:

Enter your name:   John Smith
Enter your email:   john@example.com
Hello, John Smith (john@example.com)

Example 2: Trim File Lines

#!/bin/bash

# Trim whitespace from all lines in file
input_file="$1"
output_file="${input_file%.txt}_trimmed.txt"

# Method 1: Using sed
sed 's/^[[:space:]]*//;s/[[:space:]]*$//' "$input_file" > "$output_file"

# Method 2: Using awk
awk '{gsub(/^[[:space:]]+|[[:space:]]+$/, ""); print}' "$input_file" > "$output_file"

echo "Trimmed file: $output_file"

Example 3: Function to Trim Strings

#!/bin/bash

# Reusable trim function
trim() {
  local text="$1"
  # Remove leading and trailing whitespace
  echo "$text" | xargs
}

# Or pure Bash version
trim_pure() {
  local text="$1"
  text=${text#${text%%[![:space:]]*}}
  text=${text%${text##*[![:space:]]}}
  echo "$text"
}

# Usage
name="  John Doe  "
echo "Trimmed: [$(trim "$name")]"
echo "Pure Bash: [$(trim_pure "$name")]"

Output:

Trimmed: [John Doe]
Pure Bash: [John Doe]

Example 4: Trim CSV Fields

#!/bin/bash

# Trim whitespace from CSV fields
csv_file="$1"

# Process CSV, trimming each field
awk -F',' '{
  for (i=1; i<=NF; i++) {
    gsub(/^[[:space:]]+|[[:space:]]+$/, "", $i)
  }
  print $0
}' OFS=',' "$csv_file" > "${csv_file%.csv}_trimmed.csv"

echo "Trimmed CSV: ${csv_file%.csv}_trimmed.csv"

Example 5: Trim and Validate Input

#!/bin/bash

# Function to trim and validate input
validate_input() {
  local input="$1"
  local min_length="${2:-1}"
  local max_length="${3:-100}"

  # Trim whitespace
  input=$(echo "$input" | xargs)

  # Check if empty
  if [ -z "$input" ]; then
    echo "Error: Input cannot be empty"
    return 1
  fi

  # Check length
  if [ ${#input} -lt $min_length ]; then
    echo "Error: Input too short (minimum $min_length chars)"
    return 1
  fi

  if [ ${#input} -gt $max_length ]; then
    echo "Error: Input too long (maximum $max_length chars)"
    return 1
  fi

  echo "$input"
  return 0
}

# Usage
read -p "Enter username: " username
if username=$(validate_input "$username" 3 20); then
  echo "Username accepted: $username"
else
  echo "Invalid username"
fi

Output:

Enter username:    john
Username accepted: john

Example 6: Trim and Normalize Whitespace

#!/bin/bash

# Normalize: trim AND reduce multiple spaces to single space
normalize_whitespace() {
  local text="$1"

  # Trim and collapse multiple spaces
  echo "$text" | xargs

  # Or using sed:
  # echo "$text" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//;s/[[:space:]]\+/ /g'
}

# Usage
text="  This   has    multiple     spaces  "
echo "Original: [$text]"
echo "Normalized: [$(normalize_whitespace "$text")]"

Output:

Original: [  This   has    multiple     spaces  ]
Normalized: [This has multiple spaces]

Example 7: Trim Configuration File Values

#!/bin/bash

# Parse config file, trimming values
config_file="$1"

while IFS='=' read -r key value; do
  # Skip comments and empty lines
  [[ "$key" =~ ^#.*$ ]] && continue
  [ -z "$key" ] && continue

  # Trim whitespace from key and value
  key=$(echo "$key" | xargs)
  value=$(echo "$value" | xargs)

  echo "Config: $key = $value"
done < "$config_file"

Input (config.conf):

# Database Configuration
DB_HOST   =   localhost
DB_PORT   =   5432
DB_NAME   =   mydb

Output:

Config: DB_HOST = localhost
Config: DB_PORT = 5432
Config: DB_NAME = mydb

Performance Comparison

For trimming whitespace:

Method	Speed	Memory	Portable
xargs	Very Fast	Low	Yes
Parameter Expansion	Fastest	Low	Yes
sed	Fast	Low	Yes
awk	Fast	Low	Yes

Best choice: Use xargs for simplicity, parameter expansion for pure Bash.

Important Considerations

Distinguishing Whitespace Types

Different types of whitespace require different handling:

# Spaces only
text="  hello  "
trimmed=${text#${text%%[^ ]*}}
trimmed=${trimmed%${trimmed##*[^ ]}}

# Spaces and tabs
text="	hello	"
trimmed=$(echo "$text" | sed 's/^[ \t]*//;s/[ \t]*$//')

# All Unicode whitespace (using xargs)
trimmed=$(echo "$text" | xargs)

Performance with Large Strings

Parameter expansion can be slow with very long strings:

# For large strings, use sed or awk
if [ ${#text} -gt 10000 ]; then
  trimmed=$(echo "$text" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//')
else
  trimmed=${text#${text%%[![:space:]]*}}
  trimmed=${trimmed%${trimmed##*[![:space:]]}}
fi

Preserving Internal Whitespace

Be careful not to remove internal whitespace:

text="  hello   world  "

# This removes leading/trailing but keeps internal spaces:
trimmed=$(echo "$text" | xargs)
echo "[$trimmed]"  # Output: [hello world] (note single space)

# To preserve exact internal spacing:
trimmed=${text#${text%%[![:space:]]*}}
trimmed=${trimmed%${trimmed##*[![:space:]]}}
echo "[$trimmed]"  # Output: [hello   world]

Key Points

• Use echo "$text" | xargs for simplest, most readable solution
• Use parameter expansion for pure Bash without external commands
• Use sed for complex patterns and file processing
• Use awk when already processing text
• Watch out for tabs vs spaces
• Remember xargs collapses internal whitespace
• Test with various whitespace types in your data

Quick Reference

# Simplest: trim both sides
echo "$text" | xargs

# Pure Bash: trim both sides
text=${text#${text%%[![:space:]]*}}
text=${text%${text##*[![:space:]]}}

# Sed: trim both sides
sed 's/^[[:space:]]*//;s/[[:space:]]*$//'

# Awk: trim both sides
awk '{$1=$1};1'

# Lead only
sed 's/^[[:space:]]*///

# Trailing only
sed 's/[[:space:]]*$//'

Recommended Pattern

#!/bin/bash

# Simple and portable approach
trim() {
  echo "$1" | xargs
}

# Pure Bash version (no external command)
trim_bash() {
  local text="$1"
  text=${text#${text%%[![:space:]]*}}
  text=${text%${text##*[![:space:]]}}
  echo "$text"
}

# Usage
name="  John Doe  "
echo "Trimmed: $(trim "$name")"