Say I want to search for '123456789' and I want it to be near 'firstname'. Is there a way to do it?
I've 0 logic to do this. Maybe piping greps could work?
What kind of algorithm is best suited for this purpose?
Input:
search_string1='firstname' search_string2='123456789'
proximity_#_of_lines=10
Output:
Find search_string1 and search_string2 instance if they are 10 lines near each other. i.e if these two strings are 10 lines near each other then display it to a file.
A standard approach could be:
what_we_want='123456789'
context='firstname'
distance=10
grep -E -e "${context}" -C "${distance}" file_to_look_into | grep -E -e "${what_we_want}" -C "${distance}"
The first grep ensures that we only look $distance lines around $context matching lines. The 2nd then sees if it find $what_we_want in those 2*$distance+1 lines.
If you just want the matching line as a result, drop the -C "${distance}" from the 2nd grep.
Assuming the strings can only appear once each in your input then using any awk (untested):
#!/usr/bin/env bash
awk -v str1='string1' -v str2='string2' -v prox=10 '
index($0,str1) { nr1 = NR }
index($0,str2) { nr2 = NR }
nr1 && nr2 {
delta = (nr1 > nr2 ? nr1 - nr2 : nr2 - nr1)
exit ( delta > prox )
}
' file
if (( $? == 0 )); then
printf 'success\n'
else
printf 'failure\n'
fi
123456789if it's within 10 lines of a line containing the substringfirstname, or the other way around? Is your data in some common format, like JSON, YAML, CSV, XML or other?