remove specific lines of file using bash

Question

I have to file suppost 1.txt and 2.txt 1.txt has a number in each line indicate the itemID. for example

43
345
65

the second file is csv and has the following pattern userID,itemID,time

I want to remove from second file all lines which their itemID is in the first file for this purpose I do the following

#!/bin/bash

while IFS= read -r var 
do
 paste -sd '|' | xargs -I{} grep -v -E {} 2.txt
done < "1.txt"

I read the first file and create a regular expression, but don't know the argument of egrep to get the second filed(itemID)

Answer 1

You can use sed to add commas to the beginning and end of each id in 1.txt, then just use grep to filter out the resulting strings:

sed 's/^\|$/,/g' 1.txt | grep -vFf- 2.txt

• -F interprets the patterns as fixed strings, ie not regular expressions
• -f tells grep to read the patterns from the given file, in this case - , ie standard input

Answer 2

With awk

awk -F "," 'FNR==NR{a[$0]=1;next} a[$2]!=1{print}' 1.txt 2.txt

• a[$0] = 1 records item id that exists in 1.txt into an array a
• a[$2] != 1 prints out lines from 2.txt that does not have the item id in array a