l-ma commented on code in PR #559:
URL: https://github.com/apache/opennlp/pull/559#discussion_r1421685959
##########
opennlp-tools/src/test/java/opennlp/tools/tokenize/TokenizerFactoryTest.java:
##########
@@ -163,6 +163,106 @@ void testCustomPatternAndAlphaOpt() throws IOException {
Assertions.assertTrue(factory.isUseAlphaNumericOptimization());
}
+ void checkCustomPatternForTokenizerME(String lang, String pattern, String
sentence,
+ int expectedNumTokens) throws IOException {
+
+ TokenizerModel model = train(new TokenizerFactory(lang, null, true,
+ Pattern.compile(pattern)));
+
+ TokenizerME tokenizer = new TokenizerME(model);
+ String[] tokens = tokenizer.tokenize(sentence);
+
+ Assertions.assertEquals(expectedNumTokens, tokens.length);
+ String[] sentSplit = sentence.replaceAll("\\.", " .").split(" ");
+ for (int i = 0; i < sentSplit.length; i++) {
+ Assertions.assertEquals(sentSplit[i], tokens[i]);
+ }
+ }
+
+ @Test
+ void testCustomPatternForTokenizerMEDeu() throws IOException {
+ String lang = "deu";
+ String pattern = "^[A-Za-z0-9äéöüÄÉÖÜß]+$";
+ String sentence = "Ich wähle den auf S. 183 ff. mitgeteilten Traum von der
botanischen Monographie.";
+ checkCustomPatternForTokenizerME(lang, pattern, sentence, 16);
Review Comment:
Thanks for accepting my PR! I'll be happy to help with this new issue or
other issues. I should have more time after December 19th.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
