如何使用sed提取每第2行并将每两行存入指定新文件？

Got it, let's solve this step by step using sed (plus a few standard shell tools where needed—pure sed alone can't easily split output into multiple files without some tricky workarounds).

First: Extract Every 2nd Line from main.txt #

To pull all even-numbered lines (line 2, 4, 6, etc.) from your file, use this straightforward sed command:

sed -n '2~2p' main.txt

Here's the breakdown of what each part does:

• -n: Tells sed not to print every line by default (we only want to output specific lines).
• 2~2p: The pattern means "start at line 2, then every 2 lines after that, print the line".

Second: Split Those Extracted Lines into file1.txt, file2.txt, file3.txt #

There are two common ways to distribute the lines, depending on your exact needs:

Option 1: Cycle lines through the three files (e.g., line 1 → file1, line2→file2, line3→file3, line4→file1, etc.)

If you want to spread the extracted lines evenly across the three files, combine sed with bash process substitution to split the output into three separate sed processes:

sed -n '2~2p' main.txt | tee >(sed -n '1~3p' > file1.txt) >(sed -n '2~3p' > file2.txt) | sed -n '3~3p' > file3.txt

How this works:

• The first sed command extracts all even lines from main.txt.
• tee sends that output to three destinations:
  • >(sed -n '1~3p' > file1.txt): Captures every 3rd line starting at line 1 and saves it to file1.txt.
  • >(sed -n '2~3p' > file2.txt): Captures every 3rd line starting at line 2 and saves it to file2.txt.
  • The remaining output goes to the final sed command, which captures every 3rd line starting at line3 and saves it to file3.txt.

Option 2: Split into chunks of 2 lines per file (first 2 lines → file1, next 2→file2, next2→file3, repeat)

If you want each file to hold consecutive pairs of the extracted lines, use split (a standard shell tool) alongside sed:

# Extract even lines and split into 2-line chunks
sed -n '2~2p' main.txt | split -l 2 -d --additional-suffix=".txt" - temp_file

# Rename the split files to your desired names
mv temp_file00 file1.txt
mv temp_file01 file2.txt
mv temp_file02 file3.txt

Key flags for split:

• -l 2: Split the input into files with 2 lines each.
• -d: Use numeric suffixes instead of alphabetic ones.
• --additional-suffix=".txt": Add .txt to the end of each split file.
• - temp_file: Use "temp_file" as the prefix for the split files before renaming.

If you strictly need a pure sed solution (no external tools like split), let me know—I can share a more verbose version that handles the splitting entirely with sed.

内容的提问来源于stack exchange，提问作者Drsin