#!/bin/bash -l
http_get https://nginx.org/en/docs/varindex.html \
| xmllint --html --xpath "/html/body//div[@id=\"content\"]//p" - 2>/dev/null \
-| sed -E -e "s,<a href=\"([^\"]+)\">([^<]+)</a> *(\(([^\)]+)\))?,https://nginx.org/en/docs/\1\t\2\t\4\n,g" -e "s,<[^>]+>,,g" \
+| sed -E \
+ -e "s, *<a href=\" *([^\"]+)\">([^<]+)</a> *(\(([^\)]+)\))?,https://nginx.org/en/docs/\1\t\2\t\4\n,g" \
+ -e "s,<[^>]+>,,g" \
| grep -v "^ *$" | sed "s,(\([^(]\+\))$,\1,g" \
| awk 'BEGIN{FS="\t"}{print $2"\t"$3"\t"$1}' \