Changeset 4016

Timestamp:

11/13/13 14:24:31 (11 years ago)

Author:

teckart

Message:

Added a simple blacklisting approach based on filtering XPath substrings. This needs <blacklistPattern> elements in the facetConcept configuration; all extracted XPath expression that contain one of these blacklisted substrings get removed

Location:

vlo/trunk/vlo_importer/src/main/java/eu/clarin/cmdi/vlo/importer

Files:

• FacetConceptMapping.java (modified) (2 diffs)
• FacetMappingFactory.java (modified) (1 diff)

vlo/trunk/vlo_importer/src/main/java/eu/clarin/cmdi/vlo/importer/FacetConceptMapping.java

@@ -97 +97 @@
         private List<String> patterns = new ArrayList<String>();
         
+        @XmlElement(name = "blacklistPattern")
+        private List<String> blacklistPatterns = new ArrayList<String>();
+        
         public void setConcepts(List<String> concepts) {
             this.concepts = concepts;
@@ -171 +174 @@
         public List<String> getPatterns() {
             return patterns;
+        }
+        
+        public void setBlacklistPatterns(List<String> blacklistPatterns) {
+            this.blacklistPatterns = blacklistPatterns;
+        }
+
+        public List<String> getBlacklistPatterns() {
+            return blacklistPatterns;
         }
 

vlo/trunk/vlo_importer/src/main/java/eu/clarin/cmdi/vlo/importer/FacetMappingFactory.java

@@ -131 +131 @@
                     }
                 }
+                
+                //add hardcoded patterns only when there is no xpath generated from conceptlink
                 if (xpaths.isEmpty()) {
-                    //add hardcoded patterns only when there is no xpath generated from conceptlink
                     xpaths.addAll(facetConcept.getPatterns());
                 }
+                
+                // pattern-based blacklisting: remove all XPath expressions that contain a blacklisted substring;
+                // this is basically a hack to enhance the quality of the visualised information in the VLO;
+                // should be replaced by a more intelligent approach in the future
+                for(String blacklistPattern : facetConcept.getBlacklistPatterns()) {
+                        Iterator<String> xpathIterator = xpaths.iterator();
+                        while(xpathIterator.hasNext()) {
+                                String xpath = xpathIterator.next();
+                                if(xpath.contains(blacklistPattern)) {
+                                        LOG.debug("Rejecting "+xpath+" because of blacklisted substring "+blacklistPattern);
+                                        xpathIterator.remove();
+                                }
+                        }
+                }                
+                
                 config.setCaseInsensitive(facetConcept.isCaseInsensitive());
                 config.setAllowMultipleValues(facetConcept.isAllowMultipleValues());