Cat一个.csv文件与目录中的所有其他文件

Question

I have multiple csv files that each needs header data placing in row 1, from a seperate csv file which is 1 line in length. The file is called Anon_student_ID, and every other file in the directory needs this data as line 1 while retaining it's file name.

I am trying to use the cat function from the command line.

My attempted to solution was - $ cat Anon_student_ID.csv *.csv > *.csv

However, I get bash: *.csv: ambiguous redirect ,

Could someone explain why it isn't taking the second filename and using it as the output file name? And what would be a better way of doing this.

Thanks

Answer 1

#!/bin/bash
shopt -s nullglob
shopt -s extglob
TEMP=$(mktemp)
IFS= read -r HEADER < Anon_student_ID.csv
for CSV in !(Anon_student_ID).csv; do
    { echo "$HEADER"; cat "$CSV"; } > "$TEMP" && cat "$TEMP" > "$CSV"
done
rm -f "$TEMP"

• nullglob prevents patterns from presenting themselves when no match is found.
• extglob enables extended patterns like !(...) .
• !(Anon_student_ID).csv matches all CSV files not having the filename Anon_student_ID .
• sed is good but it will look hacky on this. It would require repeated reading of the header file as well.

If files are not that big, using an array for a buffer instead of a temporary file would be good too:

#!/bin/bash
shopt -s nullglob
shopt -s extglob
[[ BASH_VERSINFO -ge 4 ]] || exit 1  ## Make sure we have `readarray`
IFS= read -r HEADER < Anon_student_ID.csv
for CSV in !(Anon_student_ID).csv; do
    readarray -t LINES < "$CSV"
    printf '%s\n' "$HEADER" "${LINES[@]}" > "$CSV"
done

Answer 2

As a one-liner:

mv Anon_student_ID.csv Anon_student_ID; for i in *.csv; do cat Anon_student_ID > tmpfile; cat ${i} >> tmpfile; mv tmpfile ${i}; done; mv Anon_student_ID Anon_student_ID.csv

In a readable format:

#!/bin/bash
mv Anon_student_ID.csv Anon_student_ID
for i in *.csv; do 
    cat Anon_student_ID > tmpfile
    cat ${i} >> tmpfile
    mv tmpfile ${i}
done
mv Anon_student_ID Anon_student_ID.csv

Anon_student_ID is renamed, so it will not be used for appending itself to itself, since it would match *.csv , too.