check mime prior to ext for metadata modification for images

source/net/yacy/cora/document/id/MultiProtocolURL.java

@@ -1178,6 +1178,10 @@ public class MultiProtocolURL implements Serializable, Comparable<MultiProtocolU
         return extension != null && extension.length() > 0 && "cgi.exe".indexOf(extension.toLowerCase()) >= 0;
     }
 
+    /**
+     * @deprecated use a mimetype considering methode (e.g. Document.getContentDomain() == ContentDomain.IMAGE)
+     */
+    @Deprecated
     public static final boolean isImage(final String extension) {
         return extension != null && extension.length() > 0 && Response.docTypeExt(extension.toLowerCase()) == Response.DT_IMAGE;
     }

source/net/yacy/search/schema/CollectionConfiguration.java

@@ -48,6 +48,7 @@ import java.util.concurrent.atomic.AtomicInteger;
 import java.util.regex.Pattern;
 
 import net.yacy.cora.document.analysis.Classification;
+import net.yacy.cora.document.analysis.Classification.ContentDomain;
 import net.yacy.cora.document.analysis.EnhancedTextProfileSignature;
 import net.yacy.cora.document.encoding.ASCII;
 import net.yacy.cora.document.id.AnchorURL;
@@ -537,14 +538,13 @@ public class CollectionConfiguration extends SchemaConfiguration implements Seri
         LinkedHashMap<DigestURL,String> outboundLinks = document.outboundLinks();
 
         Subgraph subgraph = new Subgraph(inboundLinks.size(), outboundLinks.size());
-        List<ImageEntry> images = new ArrayList<ImageEntry>();
         int c = 0;
         final Object parser = document.getParserObject();
         boolean containsCanonical = false;
         DigestURL canonical = null;
         if (parser instanceof ContentScraper) {
             final ContentScraper html = (ContentScraper) parser;
-            images = html.getImages();
+            List<ImageEntry> images = html.getImages();
 
             // header tags
             int h = 0;
@@ -913,7 +913,8 @@ public class CollectionConfiguration extends SchemaConfiguration implements Seri
             }
         }
 
-        if ((allAttr || contains(CollectionSchema.images_text_t)) && MultiProtocolURL.isImage(MultiProtocolURL.getFileExtension(digestURL.getFileName()))) {
+        // handle image source meta data
+        if ((allAttr || contains(CollectionSchema.images_text_t)) && (document.getContentDomain() == ContentDomain.IMAGE)) {
             add(doc, CollectionSchema.images_text_t, content); // the content may contain the exif data from the image parser
             content = digestURL.toTokens(); // remove all other entry but the url tokens
         }