setting up support for validation during harvest using SHACL (CLARIAH…

…/clariah-plus#50)

Dockerfile

@@ -21,6 +21,8 @@ ARG nginx_pass='Gue33mypa33!'
 RUN htpasswd -c -b /etc/nginx/.htpasswd $nginx_user $nginx_pass
 #copy additional static resources
 COPY static/* /var/www/static/
+#SHACL validation schema (opt-in with CODEMETA_VALIDATE=true)
+COPY schemas/shacl/software.ttl /etc/software.ttl
 
 ADD etc /etc
 ADD bin /usr/bin/

bin/harvest.sh

@@ -18,6 +18,12 @@ else
     CODEMETAPY_OPTS="--toolstore --css $CSS"
 fi
 
+if [ "$CODEMETA_VALIDATE" = "true" ]; then
+    HARVEST_OPTS="$HARVEST_OPTS --validate /etc/software.ttl"
+    if [ -n "$VALIDATION_TEXT" ]; then
+        HARVEST_OPTS="$HARVEST_OPTS --validatetext \"$VALIDATION_TEXT\""
+    fi
+fi
 
 echo "Starting Harvester at $(date)">&2
 if [ -n "$GITHUB_TOKEN" ]; then

my-env.env

@@ -20,3 +20,4 @@ CODEMETA_INPUTLOGDIR=/tool-store-data/
 CODEMETA_GRAPH=/tool-store-data/data.json
 #enable uploader for codemeta.json
 UPLOADER=true
+CODEMETA_VALIDATE=false