vandroucli: combine awk commands

vandroucli

@@ -22,41 +22,57 @@ declare -A URLS=(
 	["ru"]="https://vandrouki.ru"
 )
 
-rstc="\033[00m"
-yellow="\033[1;33m"
-blue="\033[1;34m"
-green="\033[1;32m"
-
 get_info() {
-	echo -e "$green>> $1 $rstc\n"
+	echo -e ">> $1 $rstc\n"
 
 	curl -s -q "$1" \
 	| sed 's/<html.*xmlns.*$//g' \
 	| xmllint --html --xpath "//html/body/div[@id='content']/div[@id='primary']/div/*[self::div/span[@class='published'] or self::h2/a]" - 2>&- \
-	| awk 'BEGIN { RS="" } {
+	| awk \
-		gsub(/\n+/, "@");
+	'BEGIN {
-		gsub(/<h2 class="entry-title">/, "\n");
+		RS="";
-		gsub(/(href=")|(" rel=[^>]*>)|(published.>)/, "@");
+		text="";
-		gsub(/(&#[0-9]+;)|(<[^@>]*>*)/, "");
+
-		gsub(/@\s*@*/,"@");
+		rstc="\033[00m";
-		gsub(/\r/,"");
+		yellow="\033[1;33m";
-
+		blue="\033[1;34m";
-		print
+		green="\033[1;32m";
-	}' \
+	}
-	| awk -F'@' -v blue="$blue" -v yellow="$yellow" -v green="$green" -v rstc="$rstc" '
+
-		/^\s*$/ {
+	function tokenize(html) {
-			next
+		gsub(/\n+/, "@", html);
-		} {
+		gsub(/<h2 class="entry-title">/, "\n", html);
-			if ($4 ~ /^\s*$/) {
+		gsub(/(href=")|(" rel=[^>]*>)|(published.>)/, "@", html);
-				date = $5
+		gsub(/(&#[0-9]+;)|(<[^@>]*>*)/, "", html);
+		gsub(/@\s*@*/, "@", html);
+		gsub(/\r/, "", html);
+
+		return html;
+	}
+
+	{
+		text = text"\n"tokenize($0);
+	}
+
+	END {
+		split(text, lines, "\n")
+
+		for (line in lines) {
+			split(lines[line], article, "@")
+
+			if (article[4] ~ /^\s*$/) {
+				date = article[5]
 			} else {
-				date = $4
+				date = article[4]
 			}
 
-			print blue " # " date "\t" yellow $3 "\n\t\t" green $2 rstc
+			if (article[3] !~ /^\s*$/) {
-		}'
+				print blue " # " date "\t" yellow article[3] "\n\t\t" green article[2] rstc
+			}
+		}
 
-	echo -e "\n--------------------------------\n"
+		print "\n--------------------------------\n"
+	}'
 }
 
 domains="${*:-.}"