URL encoding now uses its own C program; removed Python dependency

4 files changed, 32 insertions, 23 deletions

diff --git a/md.sh b/md.sh
index 3441a65..b8d48c1 100755
--- a/md.sh
+++ b/md.sh
@@ -5,8 +5,9 @@ set -e
 #	./md.sh <target dir>
 #	./md.sh <input md> <output html>
 #
-# target dir: input `src.md`, output `index.html`
-# dependencies: lowdown, fzf, python (for now)
+# input: <target dir>/src.md ; ./md-header ; ./md-footer
+# output: <target dir>/index.html
+# dependencies: lowdown, fzf, ./urlencode
 
 # Copyright 2019-2020 DistressNetwork° <uplink@distress.network>
 # 
@@ -57,8 +58,8 @@ unesc() {
 	sed -e 's;\\\&;\&;g' -e 's/\\\;/;/g'
 }
 
-encode() ( # TODO: python't
-	python -c "import urllib, sys; print urllib.quote(sys.argv[1])" "$1"
+encode() (
+	./urlencode
 )
 
 decode() {
@@ -118,7 +119,7 @@ tmp=$(mktemp -p /tmp) # using a tempfile, more convenient for storage and operat
 TOCLIST=$(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 "$data" | sed 's/^[[:space:]-]*//' | encode "$(cat /dev/stdin)") ;
+	HEADER=$(echo "$data" | sed 's/^[[:space:]-]*//' | encode) ;
 	echo "$data" | sed "s/^\([[:space:]-]*\)\(.*\)$/\1${HEADER}/" >> $tmp ;
 done
 
diff --git a/sitemap.sh b/sitemap.sh
index c6cfc38..d7c7b37 100755
--- a/sitemap.sh
+++ b/sitemap.sh
@@ -19,23 +19,7 @@ set -e
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 urlencode() (
-	# urlencode <string>
-
-	old_lc_collate=$LC_COLLATE
-	LC_COLLATE=C
-
-	length="${#1}"
-	i=1
-	while [ $i -le $length ] ; do
-		c=$(expr substr "$1" $i 1) ;
-		case $c in
-			([a-zA-Z0-9./~_-])	printf "$c" ;;
-			(*)			printf '%%%02X' "'$c" ;;
-		esac ;
-		true $((i=i+1)) ;
-	done
-
-	LC_COLLATE=$old_lc_collate
+	./urlencode | sed 's;%2F;/;g'
 )
 
 echo '<?xml version="1.0" encoding="UTF-8"?>
@@ -46,7 +30,7 @@ xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">' > $2
 echo "$(find -L $1 -type d | sort | sed -e 's;^\./;;' -e '\;/\(.git\|css\|error\|favicon\|fonts\|js\|media\);d' -e 's;$;/;')" | \
 while IFS='' read -r data ; do
 	DATE=$(grep "last-modified" ${data}/index.html | cut -f 4 -d \" | sed 's/ /T/' ) ;
-	urlencode "$data" | sed -e 's;^;<url><loc>https://;' -e "s;$;</loc><lastmod>${DATE}</lastmod></url>;" ;
+	echo "$data" | urlencode | sed -e 's;^;<url><loc>https://;' -e "s;$;</loc><lastmod>${DATE}</lastmod></url>;" ;
 	printf "\n" ;
 done >> $2
 
diff --git a/urlencode b/urlencode
new file mode 100755
index 0000000..bd0df16
--- /dev/null
+++ b/urlencode
diff --git a/urlencode.c b/urlencode.c
new file mode 100644
index 0000000..5a418f2
--- /dev/null
+++ b/urlencode.c
@@ -0,0 +1,24 @@
+#include <stdio.h>
+#include <string.h>
+
+int main() {
+	char str[2047];
+	char safe[] = "ABCDEFGHIJKLMNOPQRSTUVXYZabcdefghijklmnopqrstuvwxyz1234567890-._~";
+	char *ret;
+	int i;
+
+	scanf("%[^\n]", str);
+
+	for(i = 0; i < (strlen(str)); i++) {
+
+		ret = strchr(safe, str[i]);
+		if(ret == NULL) {
+			printf("%%%02X", (unsigned)str[i] & 0xffU);
+		} else {
+			printf("%c", str[i]);
+		}
+	}
+
+	printf("\n");
+	return 0;
+}