cush

scrape.sh

@@ -222,6 +222,7 @@ html_to_markdown() {
    local f="$(mktemp)"
    log url=$1
    cat > "$f"
+   grep '<title>' "$f" | sed 's/.*>\([^<]*\)<.title>.*/\1/g' >&2
    cat "$f" \
    | readability "file://$(realpath $f)" \
    | pandoc -f html -t markdown --ascii --toc --wrap=none --strip-comments -o - \