Bash find输出重定向与xargs管道及脚本输出优化问询

Hey there! Let's break down your two Bash questions with practical examples and clear explanations:

In Bash, there are a few reliable ways to pipe find output to xargs, and it's crucial to handle filenames with spaces or special characters properly to avoid errors:

• Basic usage (for filenames without special characters)
  Simply pipe the two commands together—find's output will be passed directly to xargs:

  find /target/directory -name "*.log" | xargs rm
  

  This command locates all .log files and passes them to xargs for deletion.

• Safe usage (handles spaces/special characters)
  If your filenames contain spaces, newlines, or special symbols, the basic method will fail. Use find's -print0 flag and xargs' -0 flag to separate filenames with null characters:

  find /target/directory -name "*.log" -print0 | xargs -0 rm
  

  -print0 appends a null character to each filename output by find, and -0 tells xargs to recognize null as the delimiter—this eliminates issues with special characters.

• Redirect to file first, then pipe to xargs
  If you need to save the find results to a file before processing with xargs, try this:

  # Save find results to a file
  find /target/directory -name "*.log" > /tmp/found_files.txt
  # Pass the file content to xargs
  xargs rm < /tmp/found_files.txt
  
  # Safe version for special characters
  find /target/directory -name "*.log" -print0 > /tmp/found_files_null.txt
  xargs -0 rm < /tmp/found_files_null.txt
  

For your disk cleanup script, we'll address two key requirements: showing find output on the terminal while saving it to a temp file, and getting pre-compression file lists instead of gzipped ones:

2.1 Show find output on terminal AND save to a temp file #

Use the tee command—it duplicates input to both standard output (your terminal) and a specified file:

# Example: Locate old logs, output to terminal and save to temp file
find /var/log -name "*.log" -mtime +7 | tee /tmp/cleanup_candidates.txt

/tmp/cleanup_candidates.txt is the temp file storing results; adjust the path/filename as needed.

2.2 Get pre-compression file lists (exclude gzipped files) #

To avoid listing .gz files and target only the original uncompressed files, add an exclusion flag to find:

# Locate uncompressed .log files (exclude .gz), output to terminal and temp file
find /var/log -name "*.log" ! -name "*.gz" -mtime +7 | tee /tmp/pre_compress_files.txt

The ! -name "*.gz" flag filters out already compressed files, ensuring you get the pre-compression file list.

Full integrated disk cleanup script example #

Here's how to combine these features into your existing script:

#!/bin/bash

# Define the disk partition to monitor
MONITOR_PART="/"
# Extract numeric disk usage percentage
USAGE=$(df -h "$MONITOR_PART" | grep -v Filesystem | awk '{print $5}' | sed 's/%//g')

# Trigger cleanup if usage exceeds 85%
if [ "$USAGE" -gt 85 ]; then
    echo "Disk usage exceeds 85%, starting cleanup..."
    # Locate 7-day-old uncompressed logs, output to terminal + temp file, then compress
    find /var/log -name "*.log" ! -name "*.gz" -mtime +7 | tee /tmp/cleanup_files.txt | xargs -I {} gzip {}
    echo "Cleanup complete. Processed file list saved to /tmp/cleanup_files.txt"
else
    echo "Disk usage is ${USAGE}%, no cleanup needed."
fi

Key details in this script:

• Uses df and text processing to get the numeric usage percentage
• find excludes .gz files to target pre-compression originals
• tee handles dual output to terminal and temp file
• xargs -I {} safely replaces each filename in the gzip command, even for files with special characters

内容的提问来源于stack exchange，提问作者capser