private <T> void assertExpectedTokens(LuceneAnalyzer analyzer, List<T> testData) { JavaRDD<T> rdd = jsc.parallelize(testData); Row[] pairs = analyzer .transform(jsql.createDataFrame(rdd, testData.get(0).getClass())) .select("wantedTokens", "tokens") .collect(); for (Row r : pairs) { Assert.assertEquals(r.get(0), r.get(1)); } }