DJviolin · August 31, 2024 07:08
diff --git a/videos.awk b/videos.awk
 #!/usr/bin/awk -f

 # This AWK script outputs a valid json file. You can convert it to ndjson stripping off some extra work from this script
 # How to run this script
 # $ awk -f videos.awk videos.csv > videos.json

 BEGIN {
    print "["
    first = 1
 }

 {
    line = $0;
    video_id = $1; # store columns in variables

    # strip off url parts from video id
    gsub(/.*\?v=|&.*/, "", video_id)

    # if line is blank, then remove it
    # remove non-alphanumeric lines
    if (video_id ~ /^$/ || video_id !~ /[[:alnum:]]/) {
        next;
    }

    # print JSON output with comma handling
    if (!first) {
        printf(",\n");
    }
    # print JSON output, because for some reason you want your DB to really handle all of those quotation marks and curly braces instead of raw data
    printf("  { \"video_id\": \"%s\" }", video_id);
    first = 0;
 }

 END {
    print "\n]"
 }
	#!/usr/bin/awk -f

	# This AWK script outputs a valid json file. You can convert it to ndjson stripping off some extra work from this script
	# How to run this script
	# $ awk -f videos.awk videos.csv > videos.json

	BEGIN {
	print "["
	first = 1
	}

	{
	line = $0;
	video_id = $1; # store columns in variables

	# strip off url parts from video id
	gsub(/.\?v=\|&./, "", video_id)

	# if line is blank, then remove it
	# remove non-alphanumeric lines
	if (video_id ~ /^$/ \|\| video_id !~ /[[:alnum:]]/) {
	next;
	}

	# print JSON output with comma handling
	if (!first) {
	printf(",\n");
	}
	# print JSON output, because for some reason you want your DB to really handle all of those quotation marks and curly braces instead of raw data
	printf(" { \"video_id\": \"%s\" }", video_id);
	first = 0;
	}

	END {
	print "\n]"
	}