path: root/md.sh

#!/bin/sh
set -e

usage="usage:
	./md.sh <target dir>
	./md.sh <input md> <output html>"

# input: <target dir>/src.md ./md-header ./md-footer <site root>/nav.html
# output: <target dir>/index.html
# dependencies: lowdown, fzy, ./urlencode

# Copyright 2019-2021 DistressNetwork° <uplink@distress.network>
# 
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
# 
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
# 
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.


IN="$1/src.md"
OUT="$1/index.html"

[ "$2" ] && { IN="$1" ; OUT="$2" ; }

[ "$1" ] || { echo "$usage" ; exit 1 ; }

[ -f $IN ] ||		{ printf "%s \tFATAL: %s not found\n" "[->]" "$IN"; exit 1 ; }
[ -f md-header ] ||	{ printf "%s \tFATAL: md-header not found\n" "[->]" ; exit 1 ; }
[ -f md-footer ] ||	{ printf "%s \tFATAL: md-footer not found\n" "[->]" ; exit 1 ; }
[ $(which lowdown) ] ||	{ printf "%s \tFATAL: lowdown not found\n" "[->]" ; exit 1 ; }

cat md-header > $OUT
lowdown -Thtml --html-no-owasp $IN >> $OUT
cat md-footer >> $OUT
printf "[ok] \tinitial assembly\n"


sedesc() {
	sed -e 's;&;\\\&;g' -e 's/;/\\\;/g'
}

unesc() {
	sed -e 's;\\\&;\&;g' -e 's/\\\;/;/g'
}

encode() {
	./urlencode
}

decode() {
	/bin/printf '%b' "$(echo "$1" | sed 's;%;\\x;g')"
}


TITLE=$(grep -e '^title: ' -m 1 $IN | cut -f 2- -d ' ' | sedesc)
sed "s/<title>/&${TITLE} \&mdash\; DistressNetwork°/1" -i'' $OUT
printf "[ok] \ttitle: %s\n" "$(echo "$TITLE" | unesc)"

BOMBER=$(grep -e '^bomber: ' -m 1 $IN | cut -f 2- -d ' ' | sedesc)
sed "s/<div class=\"bomber\">/&${BOMBER}/1" -i'' $OUT
printf "[ok] \tbomber: %s\n" "$(echo "$BOMBER" | unesc)"

MDATE=$(stat -c '%y' $IN | sed -e 's/\.[0-9]* //' -e 's/\(..\)\(..\)$/\1:\2/')	# format: "yyyy-mm-dd hh:mm:ss±zz:zz" (ISO 8601)
DATE=$(echo $MDATE | cut -f 1 -d ' ' | sed -e 's/-//g' -e 's/^..//')		# format: "yymmdd" (for humans)
sed -e "s;<div class=\"ident\"><time datetime=\";&${MDATE};1" -e "s;</time>;${DATE}&;1" -i'' $OUT
sed "s;<meta http-equiv=\"last-modified\" content=\";&${MDATE};1" -i'' $OUT
printf "[ok] \tdate: %s (%s)\n" "$DATE" "$MDATE"

LEADING=$(grep -e '^leading: ' -m 1 $IN | cut -f 2- -d ' ' | sedesc)
sed "s;<div class=\"leading\">;&${LEADING};1" -i'' $OUT
printf "[ok] \tleading: %s\n" "$(echo "$LEADING" | unesc)"

METADESC="\&laquo\;${BOMBER}\&raquo\; \&mdash\; ${LEADING}"
sed "s;<meta name=\"description\" content=\";&${METADESC};1" -i'' $OUT
printf "[ok] \tmeta description\n"

if [ -f $(echo $IN | cut -f 1 -d /)/nav.html ] ; then
NAVLINE=$(grep -n '<nav>' $OUT | cut -f 1 -d :) ;
sed "${NAVLINE}r $(echo $OUT | cut -f 1 -d /)/nav.html" -i'' $OUT ;
sed '/^<!--title:/d' -i'' $OUT ;
printf "[ok] \tsitemap\n" ;
fi

[ "$(grep -n '<p><img' $OUT)" ] && {	# skip if no figures found
# escaping backslashes with '@', to be unescaped later:
grep -n '<p><img' $OUT | sed 's/\\/@/g' | \
while IFS='' read -r data ; do
	LINE=$(echo $data | cut -f 1 -d :) ;	# extract line number
	FILE=$(echo $data | cut -f 2 -d \") ;	# extract file path
	CAPT=$(echo $data | cut -f 4 -d \") ;	# extract image caption

	if [ "$FILE" = "%tex" ] ; then 		# for tex figures (to be caught and rendered later)
		sed "${LINE}s;.*;<figure><%tex \"${CAPT}\" /></figure>;" -i'' $OUT ;
	else
		sed "${LINE}s;.*;<figure><a href=\"${FILE}\"><img src=\"${FILE}\" alt=\"\"></a><figcaption><p>${CAPT}</p></figcaption></figure>;" -i'' $OUT ;
		printf "[ok] \timage at line %s\n" "$LINE" ;
	fi
done
}


toc() {

[ $(which fzy) ] || { printf "[warn] \tfzy not found, skipping toc\n" ; return ; }
[ -f ./urlencode ] || { printf "[warn] \turlencode not found, skipping toc\n" ; return ; }

tmp=$(mktemp -p /tmp) # using a tempfile, more convenient for storage and operations

TOCLIST=$(grep -e '^#' $IN | sed -e 's; ;- ;' -e 's;^#;;' -e 's;#;	;g') # headers from source as md list

echo "$TOCLIST" | while IFS='' read -r data ; do # encode headers as urls
	HEADER=$(echo "$data" | sed 's/^[[:space:]-]*//' | encode) ;
	echo "$data" | sed "s/^\([[:space:]-]*\)\(.*\)$/\1${HEADER}/" >> $tmp ;
done

buffer=$(cat $tmp) ; echo "$buffer" | lowdown -Thtml > $tmp # convert to html, write to tempfile (reset)

sed -e '1i <nav class="toc">' -e '/^$/d' -i'' $tmp # prepend starting tag, remove empty lines

cat $tmp | while IFS='' read -r data ; do
	# get header (both encoded and raw), write as link
	ID=$(echo $data | sed "s;^<li>\([^</>]*\)\(</li>\)*$;\1;") ;
	HEADER=$(decode "$ID" | sedesc) ;
	sed -e "s;^<li>\(${ID}\);<li><a href=\"#\1\">${HEADER};" -e "s;\([[:alnum:]]\)</li>$;\1</a></li>;" -i'' $tmp ;

	# find old id in output, replace with new id
	IDLINE=$(grep -ne "^<h[[:digit:]] id=" $OUT | fzy -e "$(echo $HEADER | unesc)" | head -n 1 | cut -f 1 -d :) ;
	if [ "$HEADER" = "<ul>" ] ; then
		IDLINE="" ;
	fi ;
	if [ "$HEADER" = "</ul></li>" ] ; then
		IDLINE="" ;
	fi ;
	if [ "$HEADER" = "</ul>" ] ; then
		IDLINE="" ;
	fi ;	# TODO use case expr instead
	if [ "$IDLINE" ] ; then
		sed "${IDLINE}s;id=\".*\">.*</h;id=\"${ID}\">${HEADER}</h;" -i'' $OUT ;
		printf "\tassembled header: %s\n" "$(echo $HEADER | unesc)" ;
 	fi ;
done

# close link tags on open lines, append closing tag

sed "s;[[:alnum:]]$;&</a>;" -i'' $tmp
echo '</nav>' >> $tmp

# get toc line number in output, insert data from tempfile

TOCLINE=$(wc -l md-header | cut -f 1 -d ' ')
sed "${TOCLINE}r $tmp" -i'' $OUT

rm $tmp # cleanup

printf "[ok] \ttable of contents\n"

}


math() {

# [ $(npx katex -V 2>&1 | grep '404 Not Found') ] && { printf "[warn] \tkatex not found, skipping math rendering\n" ; return ; }

sed '/main.css/a <link rel="stylesheet" href="/css/katex.css">' -i'' $OUT

# display / "block" style:
grep -n '<figure><%tex' $OUT | \
while IFS='' read -r data ; do
	LINE=$(echo "$data" | cut -f 1 -d :) ;
	EQNBODY=$(echo "$data" | cut -f 2 -d \" ) ;
	printf "\trendering tex figure at line %s ... " "$LINE" ;
	
	tmp=$(mktemp -p /tmp) ;
	echo "<figure>" > $tmp ;
	echo "$EQNBODY" | sed 's/@/\\/g' | npx katex -d 2>&1 >> $tmp ;
	echo "</figure>" >> $tmp ;
	sed ':a;N;$!ba;s/\n//g' -i'' $tmp ;

	if [ $(grep "KaTeX parse error" $tmp) ] ; then
		printf "\n[warn] \ttex expression at line %s contained errors, skipping\n" "$LINE" ;
	else
		sed "${LINE}d" -i'' $OUT ;
		sed "$(expr $LINE - 1)r $tmp" -i'' $OUT ;
		printf "complete\n" ;
	fi ;

	rm $tmp ;
done

# inline / "span" style:
sed 's;<img src="%itex" alt=".*" />;\n&\n;g' -i'' $OUT
grep -n '<img src="%itex"' $OUT | sed 's/\\/@/g' | \
while IFS='' read -r data ; do
	LINE=$(echo "$data" | cut -f 1 -d :) ;
	EQNBODY=$(echo "$data" | cut -f 4 -d \" ) ;
	printf "\trendering inline tex figure at line %s ... " "$LINE" ;
	
	tmp=$(mktemp -p /tmp) ;
	echo "$EQNBODY" | sed 's/@/\\/g' | npx katex 2>&1 > $tmp ;
	sed ':a;N;$!ba;s/\n//g' -i'' $tmp ;

	if [ $(grep "KaTeX parse error" $tmp) ] ; then
		printf "\n[warn] \tinline tex expression at line %s contained errors, skipping\n" "$LINE" ;
	else
		sed "${LINE}d" -i'' $OUT ;
		sed "$(expr $LINE - 1)r $tmp" -i'' $OUT ;
		printf "complete\n" ;
	fi ;

	rm $tmp ;
done

printf "[ok] \tmath typesetting\n"

}


opts=$(grep -e '^opts: ' -m 1 $IN | cut -f 2- -d ' ')
while [ "$opts" ]; do
case "$opts" in
	(*[Cc]*)
		toc
		opts=$(echo $opts | sed 's/[Cc]//g')
	;;

	(*[Ff]*)
		sed "/^<footer>/,/^<\/footer>/d" -i'' $OUT
		printf "[ok] \tfooterless\n"
		opts=$(echo $opts | sed 's/[Ff]//g')
	;;

	(*[Hh]*)
		sed -e "/^<header>/,/^<\/header>/d" -e '/^<meta name="description"/d' -i'' $OUT
		sed "/bg.js/d" -i'' $OUT
		printf "[ok] \theaderless\n"
		opts=$(echo $opts | sed 's/[Hh]//g')
	;;

	(*[Mm]*)
		math
		opts=$(echo $opts | sed 's/[Mm]//g')
	;;

	(*)
		printf "[warn] \tmetadata: unrecognized option(s) \"%s\"\n" "${opts}" 
	;;
esac
done


printf "%s \tbuild complete: %s\n" "[->]" "$OUT"