Regression in PDF keywords extraction since 1.23

Hey team.

I have a unit test which is not passing anymore with Tika 1.23. Code is https://github.com/dadoonet/fscrawler/blob/master/tika/src/test/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaDocParserTest.java#L430-L496.

Basically, this line is now failing:

assertThat(doc.getMeta().getKeywords(), containsInAnyOrder("keyword1", " keyword2"));

But also this line which seems related:

assertThat(raw, hasEntry("dc:subject", "keyword1, keyword2"));

This line is failing as well although not related I think:

assertThat(raw, hasEntry("created", "2016-07-07T08:37:42Z"));

I'm not sure if it's a bug or expected. If the later, I'll simply change my test suite.