#! /bin/bash

google() (
	_is_sheets() {
		echo "$@" | grep -q 'docs.google.com.spreadsheets'
	}

	_is_doc() {
		echo "$@" | grep -q 'docs.google.com.document'
	}

   is() {
		_is_sheets "$@" || _is_doc "$@"
   }

	human_url() {
		echo "$1"
	}

   get() {
		local url="$1"
		local id="${url%/*}"
		id="${id##*/}"
		local downloaded="$(rclone get_google "$id")"
		echo "# ${downloaded##*/}"
		echo ""
		if [ "${downloaded##*.}" == "csv" ]; then
			_csv_to_md "$downloaded"
		elif [ "${downloaded##*.}" == "html" ]; then
			_html_to_md "$downloaded"
		else
			cat "$downloaded"
		fi
   }

	_html_to_md() {
		which pandoc &> /dev/null
		local f="$1"
		#log f=$f
		cat "$f" \
			| sed 's/.*<body/<body/' \
			| sed 's/<\/body>.*/<\/body>/' \
			| sed 's/<[\/]*span[^>]*>//g' \
			| perl -pe 's|<div class="c[0-9][0-9]*">.*?<\/div>||g' \
			| sed 's/<\([a-z][a-z]*\)[^>]*/<\1/g' \
			| pandoc - -f html -t commonmark -s -o - \
			| sed 's/^<[\/]*div>$//g'
	}

	_csv_to_md() {
		local f="$1"
		(
			head -n 1 "$f"
			head -n 1 "$f" \
				| sed 's/^[^,][^,]*/--- /' \
				| sed 's/[^,][^,]*$/ ---/' \
				| sed 's/,[^,][^,]*/, --- /g' \
				| sed 's/[^|]$/|/'
			tail -n +2 "$f"
		) \
			| grep . \
			| sed 's/,/ | /g' \
			| sed 's/^/| /'
	}

   expand() {
		get "$@" | head -n 1 | sed 's/^[#]* //' | base64
   }

   "$@"
)