Add many missing queries and aggs

Author: sethmlarson

elasticsearch_dsl/aggs.py

@@ -247,6 +247,13 @@ class Range(Bucket):
     name = "range"
 
 
+class RareTerms(Bucket):
+    name = "rare_terms"
+
+    def result(self, search, data):
+        return FieldBucketData(self, search, data)
+
+
 class ReverseNested(Bucket):
     name = "reverse_nested"
 
@@ -282,6 +289,13 @@ class Composite(Bucket):
     }
 
 
+class VariableWidthHistogram(Bucket):
+    name = "variable_width_histogram"
+
+    def result(self, search, data):
+        return FieldBucketData(self, search, data)
+
+
 # metric aggregations
 class TopHits(Agg):
     name = "top_hits"
@@ -318,6 +332,10 @@ class Max(Agg):
     name = "max"
 
 
+class MedianAbsoluteDeviation(Agg):
+    name = "median_absolute_deviation"
+
+
 class Min(Agg):
     name = "min"
 
@@ -342,6 +360,10 @@ class Sum(Agg):
     name = "sum"
 
 
+class TTest(Agg):
+    name = "t_test"
+
+
 class ValueCount(Agg):
     name = "value_count"
 
@@ -363,6 +385,10 @@ class CumulativeSum(Pipeline):
     name = "cumulative_sum"
 
 
+class CumulativeCardinality(Pipeline):
+    name = "cumulative_cardinality"
+
+
 class Derivative(Pipeline):
     name = "derivative"
 
@@ -371,6 +397,10 @@ class ExtendedStatsBucket(Pipeline):
     name = "extended_stats_bucket"
 
 
+class Inference(Pipeline):
+    name = "inference"
+
+
 class MaxBucket(Pipeline):
     name = "max_bucket"
 
@@ -387,6 +417,14 @@ class MovingAvg(Pipeline):
     name = "moving_avg"
 
 
+class MovingPercentiles(Pipeline):
+    name = "moving_percentiles"
+
+
+class Normalize(Pipeline):
+    name = "normalize"
+
+
 class PercentilesBucket(Pipeline):
     name = "percentiles_bucket"
 

elasticsearch_dsl/query.py

@@ -335,11 +335,16 @@ class FieldMaskingSpan(Query):
     _param_defs = {"query": {"type": "query"}}
 
 
-class SpanContainining(Query):
+class SpanContaining(Query):
     name = "span_containing"
     _param_defs = {"little": {"type": "query"}, "big": {"type": "query"}}
 
 
+# Original implementation contained
+# a typo: remove in v8.0.
+SpanContainining = SpanContaining
+
+
 class SpanWithin(Query):
     name = "span_within"
     _param_defs = {"little": {"type": "query"}, "big": {"type": "query"}}
@@ -398,6 +403,10 @@ class Ids(Query):
     name = "ids"
 
 
+class Intervals(Query):
+    name = "intervals"
+
+
 class Limit(Query):
     name = "limit"
 
@@ -450,6 +459,10 @@ class Regexp(Query):
     name = "regexp"
 
 
+class Shape(Query):
+    name = "shape"
+
+
 class SimpleQueryString(Query):
     name = "simple_query_string"
 
@@ -488,3 +501,7 @@ class Type(Query):
 
 class ParentId(Query):
     name = "parent_id"
+
+
+class Wrapper(Query):
+    name = "wrapper"

test_elasticsearch_dsl/test_aggs.py

@@ -272,3 +272,85 @@ def test_boxplot_aggregation():
     a = aggs.Boxplot(field="load_time")
 
     assert {"boxplot": {"field": "load_time"}} == a.to_dict()
+
+
+def test_rare_terms_aggregation():
+    a = aggs.RareTerms(field="the-field")
+    a.bucket("total_sales", "sum", field="price")
+    a.bucket(
+        "sales_bucket_sort",
+        "bucket_sort",
+        sort=[{"total_sales": {"order": "desc"}}],
+        size=3,
+    )
+
+    assert {
+        "aggs": {
+            "sales_bucket_sort": {
+                "bucket_sort": {"size": 3, "sort": [{"total_sales": {"order": "desc"}}]}
+            },
+            "total_sales": {"sum": {"field": "price"}},
+        },
+        "rare_terms": {"field": "the-field"},
+    } == a.to_dict()
+
+
+def test_variable_width_histogram_aggregation():
+    a = aggs.VariableWidthHistogram(field="price", buckets=2)
+    assert {"variable_width_histogram": {"buckets": 2, "field": "price"}} == a.to_dict()
+
+
+def test_median_absolute_deviation_aggregation():
+    a = aggs.MedianAbsoluteDeviation(field="rating")
+
+    assert {"median_absolute_deviation": {"field": "rating"}} == a.to_dict()
+
+
+def test_t_test_aggregation():
+    a = aggs.TTest(
+        a={"field": "startup_time_before"},
+        b={"field": "startup_time_after"},
+        type="paired",
+    )
+
+    assert {
+        "t_test": {
+            "a": {"field": "startup_time_before"},
+            "b": {"field": "startup_time_after"},
+            "type": "paired",
+        }
+    } == a.to_dict()
+
+
+def test_inference_aggregation():
+    a = aggs.Inference(model_id="model-id", buckets_path={"agg_name": "agg_name"})
+    assert {
+        "inference": {"buckets_path": {"agg_name": "agg_name"}, "model_id": "model-id"}
+    } == a.to_dict()
+
+
+def test_moving_percentiles_aggregation():
+    a = aggs.DateHistogram()
+    a.bucket("the_percentile", "percentiles", field="price", percents=[1.0, 99.0])
+    a.pipeline(
+        "the_movperc", "moving_percentiles", buckets_path="the_percentile", window=10
+    )
+
+    assert {
+        "aggs": {
+            "the_movperc": {
+                "moving_percentiles": {"buckets_path": "the_percentile", "window": 10}
+            },
+            "the_percentile": {
+                "percentiles": {"field": "price", "percents": [1.0, 99.0]}
+            },
+        },
+        "date_histogram": {},
+    } == a.to_dict()
+
+
+def test_normalize_aggregation():
+    a = aggs.Normalize(buckets_path="normalized", method="percent_of_sum")
+    assert {
+        "normalize": {"buckets_path": "normalized", "method": "percent_of_sum"}
+    } == a.to_dict()