如何利用csplit拆分文本文件，并按指定字符串命名输出的多个文件？

Since csplit doesn't support custom output filenames directly, here are two straightforward solutions using shell scripting and awk that will get the job done:

Option 1: Bash Script #

Create a script (e.g., split_entries.sh) with the following content:

#!/bin/bash

# Replace with your input file name
INPUT_FILE="original_text_file.txt"

current_output=""

while IFS= read -r line; do
    # Check if we're starting a new entry
    if [[ "$line" == ">"* ]]; then
        # Extract the filename from the line (remove leading ">")
        current_output="${line#>}"
        # Write the header line to the new file (overwrites if exists)
        echo "$line" > "$current_output"
    # Check if we've reached the end of an entry
    elif [[ "$line" == "--" ]]; then
        # Reset the current output file
        current_output=""
    # If we're in the middle of an entry, append the line
    elif [[ -n "$current_output" ]]; then
        echo "$line" >> "$current_output"
    fi
done < "$INPUT_FILE"

How to run: #

1. Save the script to a file (like split_entries.sh)
2. Make it executable: chmod +x split_entries.sh
3. Run it: ./split_entries.sh

This script reads your input file line by line:

• When it hits a line starting with >, it creates a new file named after the text following > and writes the header line to it.
• It appends subsequent lines to that file until it encounters a -- separator.
• Existing files with the same names will be overwritten, so be cautious if you have files with those names already.

Option 2: Awk One-Liner #

If you prefer a more concise approach, use this awk command directly in your terminal:

awk '
/^>/ {
    if (outfile != "") close(outfile);
    outfile = substr($0, 2);
    print > outfile;
    next;
}
/^--$/ {
    close(outfile);
    outfile = "";
    next;
}
outfile != "" {
    print >> outfile;
}
' original_text_file.txt

How it works: #

• When a line starts with >, it closes any open output file, sets the output filename to the text after >, and writes the header line to that file.
• When it sees --, it closes the current output file and stops writing to it.
• For all other lines, if an output file is active, it appends the line to that file.

Both methods will generate the 5 files you need, each named exactly after the header string and containing the full entry including the header line.

内容的提问来源于stack exchange，提问作者Zez