基于CSV第三列批量编辑第二列并处理_f_行前缀

Here's the modified awk command that combines both your requirements: removing the prefix from the second column using the third column's value, and adding the "f. " prefix when the first column contains "f":

BEGIN{FS=OFS=","}
NR==1 && NF>2 {
    # Capture the X value from the third column of the first row
    s = $3
    # Escape regex special characters in X to ensure literal matching
    gsub(/[.*+?^$()|[\]\\]/, "\\&", s)
}
{
    # Remove leading X (and any following whitespace) from the second column
    sub("^" s "[[:space:]]+", "", $2)
    # Add "f. " prefix to the second column if first column has "_f_"
    if (index($1, "_f_") != 0) {
        $2 = "f. " $2
    }
    # Print the modified line
    print
}

How it works: #

1. Set CSV field separators: FS=OFS="," ensures we correctly split and rejoin columns using commas, preserving the CSV structure.
2. Capture the shared X value: On the first row (NR==1), we store the third column's value in variable s (this is your consistent X value). We escape regex special characters in s to avoid unintended matches if X contains dots, asterisks, or other regex-specific symbols.
3. Clean up the second column: We remove the leading X value (and any trailing whitespace) from the second column using sub().
4. Add the "f. " prefix: If the first column (filename) contains "f", we prepend "f. " to the cleaned-up second column.
5. Output the modified row: The final print statement writes the updated line to the output.

To use this command, replace $pathToCSV with your actual file path (without the .csv suffix):

# Run with a separate output file (safer for testing)
awk 'BEGIN{FS=OFS=","} NR==1 && NF>2 {s=$3; gsub(/[.*+?^$()|[\]\\]/, "\\&", s)} {sub("^" s "[[:space:]]+", "", $2); if (index($1, "_f_") !=0) $2="f. "$2} 1' "$pathToCSV.csv" > output.csv

If you want to edit the file in-place (like sed -i), you can use GNU awk's in-place editing (version 4.1+):

awk -i inplace 'BEGIN{FS=OFS=","} NR==1 && NF>2 {s=$3; gsub(/[.*+?^$()|[\]\\]/, "\\&", s)} {sub("^" s "[[:space:]]+", "", $2); if (index($1, "_f_") !=0) $2="f. "$2} 1' "$pathToCSV.csv"

内容的提问来源于stack exchange，提问作者Kay Gee