使用grep和sed优化Bash数字提取脚本的建议咨询

Hey there! Let's dive into optimizing your Bash script for extracting numbers from a target statement and writing them to a file. First, I’ll start with a guess at a common "quick and dirty" temporary solution folks often use (since you didn’t share your exact code):

# Example temporary solution
echo "Sample statement with numbers 123, 45.6, and -78" | grep -o '[0-9]*\.[0-9]*\|[0-9]*' > numbers.txt

Here are actionable, practical optimization suggestions tailored to this use case:

• Cover all number types robustly
  Most basic regex patterns miss edge cases like negative numbers, scientific notation (e.g., 1.2e3), or numbers with leading/trailing whitespace. Expand your regex with extended syntax (using -E flag for cleaner grouping) to handle these:

  grep -oE '-?[0-9]+(\.[0-9]+)?([eE][+-]?[0-9]+)?' input.txt > output.txt
  

  This matches negatives, decimals, and scientific notation—way more comprehensive than basic digit matching.

• Make the script reusable and configurable
  Stop hardcoding your target statement or output file! Turn them into command-line arguments, and add error checking to avoid silent failures:

  #!/bin/bash
  # Usage: ./extract_numbers.sh "Your target statement" output_file.txt
  if [ $# -ne 2 ]; then
      echo "Error: Missing arguments!"
      echo "Usage: $0 <target_statement> <output_file>"
      exit 1
  fi
  echo "$1" | grep -oE '-?[0-9]+(\.[0-9]+)?([eE][+-]?[0-9]+)?' > "$2"
  

  Now you can run the script with any input string or file without editing the code every time.

• Skip unnecessary echo calls for file inputs
  If your target content lives in a file instead of a string, feed the file directly to grep—it’s faster and avoids extra process overhead:

  grep -oE '-?[0-9]+(\.[0-9]+)?([eE][+-]?[0-9]+)?' input.txt > output.txt
  

• Deduplicate numbers (if needed)
  If you want only unique numbers in your output, pipe the results through sort -u to remove duplicates:

  grep -oE '-?[0-9]+(\.[0-9]+)?([eE][+-]?[0-9]+)?' input.txt | sort -u > output.txt
  

• Add debugging/verbose mode
  For troubleshooting, add a -v flag to print extra context about what the script is doing:

  #!/bin/bash
  VERBOSE=false
  while getopts "v" opt; do
      case $opt in
          v) VERBOSE=true ;;
          *) echo "Invalid option: -$OPTARG" >&2; exit 1 ;;
      esac
  done
  shift $((OPTIND-1))
  
  if [ $# -ne 2 ]; then
      echo "Usage: $0 [-v] <target_statement> <output_file>"
      exit 1
  fi
  
  if $VERBOSE; then
      echo "Extracting numbers from input: '$1'"
      echo "Writing results to: $2"
  fi
  
  echo "$1" | grep -oE '-?[0-9]+(\.[0-9]+)?([eE][+-]?[0-9]+)?' > "$2"
  
  if $VERBOSE; then
      echo "Done! Extracted $(wc -l < "$2") numbers total."
  fi
  

• Use sed for extra control (alternative approach)
  If you need to clean up surrounding non-numeric characters while extracting, sed offers more flexibility. For example, to split numbers from messy text and filter valid entries:

  sed -E 's/[^0-9.eE+-]+/ /g; s/^ +| +$//g; s/ +/\n/g' input.txt | grep -E '^-?[0-9]+(\.[0-9]+)?([eE][+-]?[0-9]+)?$' > output.txt
  

  This replaces non-numeric characters with spaces, trims edges, splits into lines, then filters only valid numbers.

内容的提问来源于stack exchange，提问作者moumeneph