add comments

Author: jan-elastic

x-pack/plugin/esql/compute/src/test/java/org/elasticsearch/compute/aggregation/blockhash/CategorizeBlockHashTests.java

@@ -290,6 +290,8 @@ public void testCategorize_withDriver() {
                 BytesRefVector.Builder textsBuilder = driverContext.blockFactory().newBytesRefVectorBuilder(10);
                 LongVector.Builder countsBuilder = driverContext.blockFactory().newLongVectorBuilder(10)
             ) {
+                // Note that just using "a" or "aaa" doesn't work, because the ml_standard
+                // tokenizer drops numbers, including hexadecimal ones.
                 textsBuilder.appendBytesRef(new BytesRef("aaazz"));
                 textsBuilder.appendBytesRef(new BytesRef("bbbzz"));
                 textsBuilder.appendBytesRef(new BytesRef("words words words goodbye jan"));

x-pack/plugin/esql/qa/testFixtures/src/main/resources/categorize.csv-spec

@@ -247,6 +247,9 @@ COUNT():long | category:keyword
            5 | .*?Kubernetes.+?cluster.*?
            2 | .*?servers.*?
            1 | null
+
+// Note: DB is removed from "DB servers", because the ml_standard
+// tokenizer drops numbers, including hexadecimal ones.
 ;
 
 on TO_UPPER