2 files changed, 60 insertions, 0 deletions
diff --git a/linguistics/src/main/java/com/yahoo/language/opennlp/LanguageDetectorFactory.java b/linguistics/src/main/java/com/yahoo/language/opennlp/LanguageDetectorFactory.java
new file mode 100644
index 00000000000..aa4387bcc45
--- /dev/null
+++ b/linguistics/src/main/java/com/yahoo/language/opennlp/LanguageDetectorFactory.java
@@ -0,0 +1,29 @@
+// Copyright Yahoo. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
+package com.yahoo.language.opennlp;
+
+import opennlp.tools.langdetect.DefaultLanguageDetectorContextGenerator;
+import opennlp.tools.langdetect.LanguageDetectorContextGenerator;
+import opennlp.tools.util.normalizer.EmojiCharSequenceNormalizer;
+import opennlp.tools.util.normalizer.NumberCharSequenceNormalizer;
+import opennlp.tools.util.normalizer.ShrinkCharSequenceNormalizer;
+import opennlp.tools.util.normalizer.TwitterCharSequenceNormalizer;
+
+/**
+ * Overrides the UrlCharSequenceNormalizer, which has a bad regex, until fixed: https://issues.apache.org/jira/browse/OPENNLP-1350
+ *
+ * @author jonmv
+ */
+@SuppressWarnings("unused") // Loaded by black magic.
+public class LanguageDetectorFactory extends opennlp.tools.langdetect.LanguageDetectorFactory {
+
+    @Override
+    public LanguageDetectorContextGenerator getContextGenerator() {
+        return new DefaultLanguageDetectorContextGenerator(1, 3,
+                                                           EmojiCharSequenceNormalizer.getInstance(),
+                                                           UrlCharSequenceNormalizer.getInstance(),
+                                                           TwitterCharSequenceNormalizer.getInstance(),
+                                                           NumberCharSequenceNormalizer.getInstance(),
+                                                           ShrinkCharSequenceNormalizer.getInstance());
+    }
+
+}
diff --git a/linguistics/src/main/java/com/yahoo/language/opennlp/UrlCharSequenceNormalizer.java b/linguistics/src/main/java/com/yahoo/language/opennlp/UrlCharSequenceNormalizer.java
new file mode 100644
index 00000000000..883319e2f8b
--- /dev/null
+++ b/linguistics/src/main/java/com/yahoo/language/opennlp/UrlCharSequenceNormalizer.java
@@ -0,0 +1,31 @@
+// Copyright Yahoo. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
+package com.yahoo.language.opennlp;
+
+import opennlp.tools.util.normalizer.CharSequenceNormalizer;
+
+import java.util.regex.Pattern;
+
+/**
+ * Modifies {@link opennlp.tools.util.normalizer.UrlCharSequenceNormalizer} to avoid the bad email regex.
+ *
+ * @author jonmv
+ */
+public class UrlCharSequenceNormalizer implements CharSequenceNormalizer {
+
+    private static final Pattern URL_REGEX =
+            Pattern.compile("https?://[-_.?&~;+=/#0-9A-Za-z]+");
+    private static final Pattern MAIL_REGEX =
+            Pattern.compile("(?<![-+_.0-9A-Za-z])[-+_.0-9A-Za-z]+@[-0-9A-Za-z]+[-.0-9A-Za-z]+");
+
+    private static final UrlCharSequenceNormalizer INSTANCE = new UrlCharSequenceNormalizer();
+
+    public static UrlCharSequenceNormalizer getInstance() {
+        return INSTANCE;
+    }
+
+    public CharSequence normalize(CharSequence text) {
+        String modified = URL_REGEX.matcher(text).replaceAll(" ");
+        return MAIL_REGEX.matcher(modified).replaceAll(" ");
+    }
+
+}