Script to extract a block of file using regex, like: ./extractBlock.sh "18, 2016 2:.*PM" "18, 2016 3:.*PM" c1vm081.out ras2.log

extractBlock.sh

#!/usr/bin/env bash
#############################################################
# Script to extract a block of text from big plain text files
# using regex to start and end lines
#
# How to use:
# ./extractBlock.sh [regex_start_line] [regex_end_line] [input_file] [output_file]
#
# Example of usage:
# ./extractBlock.sh \"18, 2016 2:.*PM\" \"18, 2016 3:.*PM\" c1vm081.out ras2.log
#
# Created by fgabbay
#############################################################


#############################################################
# HELPERS
#############################################################
is_empty() {
    local var=$1
    [[ -z $var ]]
}

#############################################################
# VALIDATE PARAMS
#############################################################
is_valid_params(){
	is_empty $regex_start && usage && exit 0;
	is_empty $regex_end && usage && exit 0;
	is_empty $input_file && usage && exit 0;
	is_empty $output_file && usage && exit 0;
	
	echo "The params is ok, keep going..."
}

#############################################################
# USAGE
#############################################################
usage(){
	cat <<- EOF 
    ===========================================================
    usage: $script_file_name
    ===========================================================

    Extract a block of text using regex to begining and end.

    $script_file_name [regex_start_line] [regex_end_line] [input_file] [output_file]

    Examples: 
    $script_file_name "18, 2016 2:.*PM" "18, 2016 3:.*PM" c1vm081.out ras2.log
	
    or
    $script_file_name "21, 2016 8:.*AM" "21, 2016 9:.*AM" logs-weblogic/c1vm081.out weblogic-travamento-8h27.log
	
	EOF
}

#############################################################
# MAIN EXECUTION
#############################################################
main() {
	# Check if params was sent
	is_valid_params

	echo "Looking for begin and end lines..."
	
	local begining_at=$(grep -n "$regex_start" $input_file | grep -Eo '^[^:]+' | awk 'NR==1 {print; exit}')
	
	# If begining was not found, abort script
	is_empty $begining_at \
		&& echo "The pattern \"${regex_start}\" for begining could not be found in file \"${input_file}\"." \
		&& exit 0;

	echo "Starting at line: $begining_at"

	local end_at=$(grep -n "$regex_end" $input_file | grep -Eo '^[^:]+' | awk 'NR==1 {print; exit}')
	echo "Ending block at line: $end_at"
	
	# If end_at is empty, get last line found from begining_at
	is_empty $end_at \
		&& echo "End of block was not found, getting last line from start param..." \
		&& end_at=$(grep -n "18, 2016 2:.*PM" ${input_file} | grep -Eo '^[^:]+' | awk 'END{print}') \
		&& echo "Ending block at line: $end_at"
	
	echo "Capturing between lines ${begining_at} and ${end_at}..."
	
	sed -n "${begining_at},${end_at}p" ${input_file} > ${output_file}
	
	echo "done!"
}

#############################################################
# GLOBAL VARIABLES
#############################################################
readonly regex_start=$1
readonly regex_end=$2
readonly input_file=$3
readonly output_file=$4
readonly script_file_name=$0

main