Changeset 33469

Timestamp:

2019-09-13T21:46:09+12:00 (5 years ago)

Author:

ak19

Message:

Don't want URLs with the word product(s) in them (but production should be acceptable). As products URLs tend to be autotranslated and poorly so into Maori.

File:

• gs3-extensions/maori-lang-detection/src/org/greenstone/atea/WETProcessor.java (modified) (1 diff)

gs3-extensions/maori-lang-detection/src/org/greenstone/atea/WETProcessor.java

@@ -200 +200 @@
     
     File parentFolder = null;
-    
-    if(lineCount >= MIN_LINE_COUNT && contentLength >= MIN_CONTENT_LENGTH) {
+
+    // want to match "product(s)" but not "production"
+    
+    //if(recordURI.matches(".*/?product[^a-rt-z].*")) {//if(recordURI.matches(".*/?products?/?.*")) {
+    if(recordURI.contains("product") && !recordURI.contains("production")) {
+
+        // don't want a "translated" product site/online store
+        // These curiously often tend to have "product(s)" in the URL
+        parentFolder = WETProcessor.discardFolder;
+    }   
+    else if(lineCount >= MIN_LINE_COUNT && contentLength >= MIN_CONTENT_LENGTH) {
         parentFolder = WETProcessor.keepFolder;
         System.err.println("@@@KEEPING");