md.sh: refactored table of contents generation (removed fzy and urlencode dependencies)

3 files changed, 26 insertions, 43 deletions

diff --git a/md.sh b/md.sh
index a102e42..fbc5fec 100755
--- a/md.sh
+++ b/md.sh
@@ -7,7 +7,7 @@ usage="usage:
 
 # input: <target dir>/src.md ./md-header ./md-footer <site root>/nav.html
 # output: <target dir>/index.html
-# dependencies: lowdown, fzy, ./urlencode
+# dependencies: lowdown
 
 # Copyright 2019-2021 DistressNetwork° <uplink@distress.network>
 # 
@@ -32,10 +32,10 @@ OUT="$1/index.html"
 
 [ "$1" ] || { echo "$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 ; }
+[ -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 ; }
+[ -e $(which lowdown) ]	|| { printf "%s \tFATAL: lowdown not found\n" "[->]" ;		exit 1 ; }
 
 cat md-header > $OUT
 lowdown -Thtml --html-no-owasp $IN >> $OUT
@@ -51,14 +51,6 @@ 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
@@ -111,18 +103,10 @@ done
 
 toc() {
 
-[ $(which fzy) ] ||	{ printf "[warn] \tfzy not found, skipping toc\n" ;		return ; }
-[ -f ./urlencode ] ||	{ printf "[warn] \turlencode not found, skipping toc\n" ;	return ; }
-
-# tempfiles are more convenient for storage and operations
 tmp=$(mktemp -p /tmp) 
-# headers from source as md list
-TOCLIST=$(grep -e '^#' $IN | sed -e 's; ;- ;' -e 's;^#;;' -e 's;#;	;g')
 
-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
+# format headers from source as md list
+grep -e '^#' $IN | sed -e 's; ;- ;' -e 's;^#;;' -e 's;#;	;g' > $tmp
 
 # convert to html, write to tempfile (reset)
 buffer=$(cat $tmp) ; echo "$buffer" | lowdown -Thtml > $tmp 
@@ -130,24 +114,23 @@ buffer=$(cat $tmp) ; echo "$buffer" | lowdown -Thtml > $tmp
 # prepend starting tag, remove empty lines
 sed -e '1i <nav class="toc">' -e '/^$/d' -i'' $tmp 
 
-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 $HEADER | unesc)" | head -n 1 | cut -f 1 -d :) ;
-	case "$HEADER" in	# filter out bad tag matches
-		("<nav class=\"toc\">")	;;
-		("<ul>")		;;
-		("</ul></li>")		;;
-		("</ul>")		;;
-		("</nav>")		;;
-		(*)	
-			sed "${IDLINE}s;id=\".*\">.*</h;id=\"${ID}\">${HEADER}</h;" -i'' $OUT ;
-			printf "\tassembled header: %s\n" "$(echo $HEADER | unesc)" ;;
-	esac ;
+# get queue of headers from output
+HEADERS=$(grep -e "^<h[[:digit:]] id=" $OUT)
+
+# for each toc list item:
+cat $tmp | grep -e "^<li>" | while IFS='' read -r data ; do
+	# get next header in queue, extract id
+	HEADER=$(echo "$HEADERS" | head -n 1) ;
+	ID=$(echo "$HEADER" | cut -f 2 -d \" | sedesc) ;
+
+	# rewrite toc item as link with id
+	ITEM=$(echo "$data" | sed "s;^<li>\([^</>]*\)\(</li>\)*$;\1;" | sedesc) ;
+	sed -e "s;^<li>\(${ITEM}\);<li><a href=\"#${ID}\">\1;" -e "s;\([[:alnum:]]\)</li>$;\1</a></li>;" -i'' $tmp ;
+
+	# remove processed header from queue
+	HEADERS=$(echo "$HEADERS" | sed '1d') ;
+
+	printf "\tassembled header: %s\n" "$(echo $ITEM | unesc)" ;
 done
 
 # close link tags on open lines, append closing tag
@@ -158,7 +141,7 @@ echo '</nav>' >> $tmp
 TOCLINE=$(wc -l md-header | cut -f 1 -d ' ')
 sed "${TOCLINE}r $tmp" -i'' $OUT
 
-rm $tmp # cleanup
+rm $tmp
 
 printf "[ok] \ttable of contents\n"
 
diff --git a/urlencode b/urlencode
index 517813d..e0a1e0c 100755
--- a/urlencode
+++ b/urlencode
diff --git a/urlencode.c b/urlencode.c
index 693b98c..0255454 100755
--- a/urlencode.c
+++ b/urlencode.c
@@ -18,7 +18,7 @@
 #include <string.h>
 
 int main() {
-	char str[2047];
+	char str[2048];
 	char safe[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz1234567890-._~";
 	char *ret;
 	int i;