REST tests for variable width historam (elastic#84836)

This expands on the super basic REST tests for variable width histogram
with some ever so slightly less basic tests. They are still pretty
basic, but they do a good job of making sure that we don't break
backwards compatibility with the API.

Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>

rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.aggregation/350_variable_width_histogram.yml

@@ -24,9 +24,47 @@ setup:
           - '{"number": 4}'
           - '{"index": {}}'
           - '{"number": 5}'
+          - '{"index": {}}'
+          - '{"number": 6}'
+          - '{"index": {}}'
+          - '{"number": 7}'
+          - '{"index": {}}'
+          - '{"number": 8}'
+          - '{"index": {}}'
+          - '{"number": 9}'
+          - '{"index": {}}'
+          - '{"number": 10}'
+          - '{"index": {}}'
+          - '{"number": 11}'
+
+---
+defaults:
+  - skip:
+      version: " - 7.8.99"
+      reason: added in 7.9.0
+  - do:
+      search:
+        body:
+          size: 0
+          aggs:
+            histo:
+              variable_width_histogram:
+                field: number
+  - match: { hits.total.value: 11 }
+  - length: { aggregations.histo.buckets: 10 }
+  - match: { aggregations.histo.buckets.0: {"min": -3.0, "key": -2.5, "max": -2.0, "doc_count": 2 } }
+  - match: { aggregations.histo.buckets.1: {"min":  1.0, "key":  1.0, "max":  1.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.2: {"min":  4.0, "key":  4.0, "max":  4.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.3: {"min":  5.0, "key":  5.0, "max":  5.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.4: {"min":  6.0, "key":  6.0, "max":  6.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.5: {"min":  7.0, "key":  7.0, "max":  7.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.6: {"min":  8.0, "key":  8.0, "max":  8.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.7: {"min":  9.0, "key":  9.0, "max":  9.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.8: {"min": 10.0, "key": 10.0, "max": 10.0, "doc_count": 1 } }
+  - match: { aggregations.histo.buckets.9: {"min": 11.0, "key": 11.0, "max": 11.0, "doc_count": 1 } }
 
 ---
-"basic":
+num_buckets:
   - skip:
       version: " - 7.8.99"
       reason: added in 7.9.0
@@ -39,12 +77,48 @@ setup:
               variable_width_histogram:
                 field: number
                 buckets: 3
-  - match: { hits.total.value: 5 }
+  - match: { hits.total.value: 11 }
   - length: { aggregations.histo.buckets: 3 }
-  - match: { aggregations.histo.buckets.0.key: -2.5 }
-  - match: { aggregations.histo.buckets.0.doc_count: 2 }
-  - match: { aggregations.histo.buckets.1.key: 1.0 }
-  - match: { aggregations.histo.buckets.1.doc_count: 1 }
-  - match: { aggregations.histo.buckets.2.key: 4.5 }
-  - match: { aggregations.histo.buckets.2.doc_count: 2 }
+  - match: { aggregations.histo.buckets.0: {"min": -3.0, "key": -1.3333333333333333, "max": 1.0, "doc_count": 3 } }
+  - match: { aggregations.histo.buckets.1: {"min":  4.0, "key":  5.5, "max":   7.0, "doc_count": 4 } }
+  - match: { aggregations.histo.buckets.2: {"min":  8.0, "key":  9.5, "max":  11.0, "doc_count": 4 } }
+
+---
+shard_size:
+  - skip:
+      version: " - 7.8.99"
+      reason: added in 7.9.0
+  - do:
+      search:
+        body:
+          size: 0
+          aggs:
+            histo:
+              variable_width_histogram:
+                field: number
+                buckets: 2
+                shard_size: 10000  # This is an accuracy tuning parameter that the tests can't do much with. This just makes sure you can set it.
+  - match: { hits.total.value: 11 }
+  - length: { aggregations.histo.buckets: 2 }
+  - match: { aggregations.histo.buckets.0: {"min": -3.0, "key": -1.3333333333333333, "max": 1.0, "doc_count": 3 } }
+  - match: { aggregations.histo.buckets.1: {"min":  4.0, "key":  7.5, "max": 11.0, "doc_count": 8 } }
 
+---
+initial_buffer:
+  - skip:
+      version: " - 7.8.99"
+      reason: added in 7.9.0
+  - do:
+      search:
+        body:
+          size: 0
+          aggs:
+            histo:
+              variable_width_histogram:
+                field: number
+                buckets: 2
+                initial_buffer: 10000  # This is an accuracy tuning parameter that the tests can't do much with. This just makes sure you can set it.
+  - match: { hits.total.value: 11 }
+  - length: { aggregations.histo.buckets: 2 }
+  - match: { aggregations.histo.buckets.0: {"min": -3.0, "key": -1.3333333333333333, "max": 1.0, "doc_count": 3 } }
+  - match: { aggregations.histo.buckets.1: {"min":  4.0, "key":  7.5, "max": 11.0, "doc_count": 8 } }