elasticsearch 9.2.1__py3-none-any.whl → 9.3.0__py3-none-any.whl

elasticsearch/dsl/field.py

@@ -1568,7 +1568,9 @@ class DenseVector(Field):
         self,
         *args: Any,
         dims: Union[int, "DefaultType"] = DEFAULT,
-        element_type: Union[Literal["bit", "byte", "float"], "DefaultType"] = DEFAULT,
+        element_type: Union[
+            Literal["bit", "byte", "float", "bfloat16"], "DefaultType"
+        ] = DEFAULT,
         index: Union[bool, "DefaultType"] = DEFAULT,
         index_options: Union[
             "types.DenseVectorIndexOptions", Dict[str, Any], "DefaultType"
@@ -1616,11 +1618,33 @@ class DenseVector(Field):
             kwargs["multi"] = True
         super().__init__(*args, **kwargs)
 
-    def _deserialize(self, data: Any) -> Any:
-        if self._element_type == "float":
-            return float(data)
-        elif self._element_type == "byte":
-            return int(data)
+
+class NumpyDenseVector(DenseVector):
+    """A dense vector field that uses numpy arrays.
+
+    Accepts the same arguments as class ``DenseVector`` plus:
+
+    :arg dtype: The numpy data type to use for the array. If not given, numpy will select the type based on the data.
+    """
+
+    def __init__(self, *args: Any, dtype: Optional[type] = None, **kwargs: Any):
+        super().__init__(*args, **kwargs)
+        self._dtype = dtype
+
+    def deserialize(self, data: Any) -> Any:
+        if isinstance(data, list):
+            import numpy as np
+
+            return np.array(data, dtype=self._dtype)
+        return super().deserialize(data)
+
+    def clean(self, data: Any) -> Any:
+        # this method does the same as the one in the parent classes, but it
+        # avoids comparisons that do not work for numpy arrays
+        if data is not None:
+            data = self.deserialize(data)
+        if (data is None or len(data) == 0) and self._required:
+            raise ValidationException("Value required for this field.")
         return data
 
 
@@ -1808,6 +1832,59 @@ class DoubleRange(RangeField):
         super().__init__(*args, **kwargs)
 
 
+class ExponentialHistogram(Field):
+    """
+    :arg time_series_metric:
+    :arg meta: Metadata about the field.
+    :arg properties:
+    :arg ignore_above:
+    :arg dynamic:
+    :arg fields:
+    :arg synthetic_source_keep:
+    """
+
+    name = "exponential_histogram"
+    _param_defs = {
+        "properties": {"type": "field", "hash": True},
+        "fields": {"type": "field", "hash": True},
+    }
+
+    def __init__(
+        self,
+        *args: Any,
+        time_series_metric: Union[
+            Literal["gauge", "counter", "summary", "histogram", "position"],
+            "DefaultType",
+        ] = DEFAULT,
+        meta: Union[Mapping[str, str], "DefaultType"] = DEFAULT,
+        properties: Union[Mapping[str, Field], "DefaultType"] = DEFAULT,
+        ignore_above: Union[int, "DefaultType"] = DEFAULT,
+        dynamic: Union[
+            Literal["strict", "runtime", "true", "false"], bool, "DefaultType"
+        ] = DEFAULT,
+        fields: Union[Mapping[str, Field], "DefaultType"] = DEFAULT,
+        synthetic_source_keep: Union[
+            Literal["none", "arrays", "all"], "DefaultType"
+        ] = DEFAULT,
+        **kwargs: Any,
+    ):
+        if time_series_metric is not DEFAULT:
+            kwargs["time_series_metric"] = time_series_metric
+        if meta is not DEFAULT:
+            kwargs["meta"] = meta
+        if properties is not DEFAULT:
+            kwargs["properties"] = properties
+        if ignore_above is not DEFAULT:
+            kwargs["ignore_above"] = ignore_above
+        if dynamic is not DEFAULT:
+            kwargs["dynamic"] = dynamic
+        if fields is not DEFAULT:
+            kwargs["fields"] = fields
+        if synthetic_source_keep is not DEFAULT:
+            kwargs["synthetic_source_keep"] = synthetic_source_keep
+        super().__init__(*args, **kwargs)
+
+
 class Flattened(Field):
     """
     :arg boost:
@@ -2277,6 +2354,7 @@ class HalfFloat(Float):
 class Histogram(Field):
     """
     :arg ignore_malformed:
+    :arg time_series_metric:
     :arg meta: Metadata about the field.
     :arg properties:
     :arg ignore_above:
@@ -2295,6 +2373,10 @@ class Histogram(Field):
         self,
         *args: Any,
         ignore_malformed: Union[bool, "DefaultType"] = DEFAULT,
+        time_series_metric: Union[
+            Literal["gauge", "counter", "summary", "histogram", "position"],
+            "DefaultType",
+        ] = DEFAULT,
         meta: Union[Mapping[str, str], "DefaultType"] = DEFAULT,
         properties: Union[Mapping[str, Field], "DefaultType"] = DEFAULT,
         ignore_above: Union[int, "DefaultType"] = DEFAULT,
@@ -2309,6 +2391,8 @@ class Histogram(Field):
     ):
         if ignore_malformed is not DEFAULT:
             kwargs["ignore_malformed"] = ignore_malformed
+        if time_series_metric is not DEFAULT:
+            kwargs["time_series_metric"] = time_series_metric
         if meta is not DEFAULT:
             kwargs["meta"] = meta
         if properties is not DEFAULT:
@@ -3892,7 +3976,7 @@ class SemanticText(Field):
             "types.SemanticTextIndexOptions", Dict[str, Any], "DefaultType"
         ] = DEFAULT,
         chunking_settings: Union[
-            "types.ChunkingSettings", Dict[str, Any], "DefaultType"
+            "types.ChunkingSettings", None, Dict[str, Any], "DefaultType"
         ] = DEFAULT,
         fields: Union[Mapping[str, Field], "DefaultType"] = DEFAULT,
         **kwargs: Any,

elasticsearch/dsl/query.py

@@ -1397,7 +1397,7 @@ class MoreLikeThis(Query):
         minimum_should_match: Union[int, str, "DefaultType"] = DEFAULT,
         min_term_freq: Union[int, "DefaultType"] = DEFAULT,
         min_word_length: Union[int, "DefaultType"] = DEFAULT,
-        routing: Union[str, "DefaultType"] = DEFAULT,
+        routing: Union[str, Sequence[str], "DefaultType"] = DEFAULT,
         stop_words: Union[
             Literal[
                 "_arabic_",
@@ -1719,7 +1719,7 @@ class Percolate(Query):
         index: Union[str, "DefaultType"] = DEFAULT,
         name: Union[str, "DefaultType"] = DEFAULT,
         preference: Union[str, "DefaultType"] = DEFAULT,
-        routing: Union[str, "DefaultType"] = DEFAULT,
+        routing: Union[str, Sequence[str], "DefaultType"] = DEFAULT,
         version: Union[int, "DefaultType"] = DEFAULT,
         boost: Union[float, "DefaultType"] = DEFAULT,
         _name: Union[str, "DefaultType"] = DEFAULT,

elasticsearch/dsl/response/__init__.py

@@ -338,6 +338,7 @@ class UpdateByQueryResponse(AttrDict[Any], Generic[_R]):
     :arg retries: The number of retries attempted by update by query.
         `bulk` is the number of bulk actions retried. `search` is the
         number of search actions retried.
+    :arg slices: Status of each slice if the update by query was sliced
     :arg task:
     :arg timed_out: If true, some requests timed out during the update by
         query.
@@ -366,6 +367,7 @@ class UpdateByQueryResponse(AttrDict[Any], Generic[_R]):
     deleted: int
     requests_per_second: float
     retries: "types.Retries"
+    slices: Sequence["types.ReindexStatus"]
     task: str
     timed_out: bool
     took: Any

elasticsearch/dsl/types.py

@@ -400,6 +400,9 @@ class DenseVectorIndexOptions(AttrDict[Any]):
     :arg rescore_vector: The rescore vector options. This is only
         applicable to `bbq_disk`, `bbq_hnsw`, `int4_hnsw`, `int8_hnsw`,
         `bbq_flat`, `int4_flat`, and `int8_flat` index types.
+    :arg on_disk_rescore: `true` if vector rescoring should be done on-
+        disk  Only applicable to `bbq_disk`, `bbq_hnsw`, `int4_hnsw`,
+        `int8_hnsw`
     """
 
     type: Union[
@@ -422,6 +425,7 @@ class DenseVectorIndexOptions(AttrDict[Any]):
     rescore_vector: Union[
         "DenseVectorIndexOptionsRescoreVector", Dict[str, Any], DefaultType
     ]
+    on_disk_rescore: Union[bool, DefaultType]
 
     def __init__(
         self,
@@ -446,6 +450,7 @@ class DenseVectorIndexOptions(AttrDict[Any]):
         rescore_vector: Union[
             "DenseVectorIndexOptionsRescoreVector", Dict[str, Any], DefaultType
         ] = DEFAULT,
+        on_disk_rescore: Union[bool, DefaultType] = DEFAULT,
         **kwargs: Any,
     ):
         if type is not DEFAULT:
@@ -458,6 +463,8 @@ class DenseVectorIndexOptions(AttrDict[Any]):
             kwargs["m"] = m
         if rescore_vector is not DEFAULT:
             kwargs["rescore_vector"] = rescore_vector
+        if on_disk_rescore is not DEFAULT:
+            kwargs["on_disk_rescore"] = on_disk_rescore
         super().__init__(kwargs)
 
 
@@ -602,7 +609,7 @@ class FieldLookup(AttrDict[Any]):
     id: Union[str, DefaultType]
     index: Union[str, DefaultType]
     path: Union[str, InstrumentedField, DefaultType]
-    routing: Union[str, DefaultType]
+    routing: Union[str, Sequence[str], DefaultType]
 
     def __init__(
         self,
@@ -610,7 +617,7 @@ class FieldLookup(AttrDict[Any]):
         id: Union[str, DefaultType] = DEFAULT,
         index: Union[str, DefaultType] = DEFAULT,
         path: Union[str, InstrumentedField, DefaultType] = DEFAULT,
-        routing: Union[str, DefaultType] = DEFAULT,
+        routing: Union[str, Sequence[str], DefaultType] = DEFAULT,
         **kwargs: Any,
     ):
         if id is not DEFAULT:
@@ -2328,7 +2335,7 @@ class LikeDocument(AttrDict[Any]):
     _id: Union[str, DefaultType]
     _index: Union[str, DefaultType]
     per_field_analyzer: Union[Mapping[Union[str, InstrumentedField], str], DefaultType]
-    routing: Union[str, DefaultType]
+    routing: Union[str, Sequence[str], DefaultType]
     version: Union[int, DefaultType]
     version_type: Union[Literal["internal", "external", "external_gte"], DefaultType]
 
@@ -2342,7 +2349,7 @@ class LikeDocument(AttrDict[Any]):
         per_field_analyzer: Union[
             Mapping[Union[str, InstrumentedField], str], DefaultType
         ] = DEFAULT,
-        routing: Union[str, DefaultType] = DEFAULT,
+        routing: Union[str, Sequence[str], DefaultType] = DEFAULT,
         version: Union[int, DefaultType] = DEFAULT,
         version_type: Union[
             Literal["internal", "external", "external_gte"], DefaultType
@@ -3931,7 +3938,7 @@ class TermsLookup(AttrDict[Any]):
     index: Union[str, DefaultType]
     id: Union[str, DefaultType]
     path: Union[str, InstrumentedField, DefaultType]
-    routing: Union[str, DefaultType]
+    routing: Union[str, Sequence[str], DefaultType]
 
     def __init__(
         self,
@@ -3939,7 +3946,7 @@ class TermsLookup(AttrDict[Any]):
         index: Union[str, DefaultType] = DEFAULT,
         id: Union[str, DefaultType] = DEFAULT,
         path: Union[str, InstrumentedField, DefaultType] = DEFAULT,
-        routing: Union[str, DefaultType] = DEFAULT,
+        routing: Union[str, Sequence[str], DefaultType] = DEFAULT,
         **kwargs: Any,
     ):
         if index is not DEFAULT:
@@ -4894,7 +4901,7 @@ class CompletionSuggestOption(AttrDict[Any]):
     fields: Mapping[str, Any]
     _id: str
     _index: str
-    _routing: str
+    _routing: Union[str, Sequence[str]]
     _score: float
     _source: Any
     score: float
@@ -6235,6 +6242,58 @@ class RateAggregate(AttrDict[Any]):
     meta: Mapping[str, Any]
 
 
+class ReindexStatus(AttrDict[Any]):
+    """
+    :arg batches: (required) The number of scroll responses pulled back by
+        the reindex.
+    :arg deleted: (required) The number of documents that were
+        successfully deleted.
+    :arg noops: (required) The number of documents that were ignored
+        because the script used for the reindex returned a `noop` value
+        for `ctx.op`.
+    :arg requests_per_second: (required) The number of requests per second
+        effectively executed during the reindex.
+    :arg retries: (required) The number of retries attempted by reindex.
+        `bulk` is the number of bulk actions retried and `search` is the
+        number of search actions retried.
+    :arg throttled_millis: (required) Number of milliseconds the request
+        slept to conform to `requests_per_second`.
+    :arg throttled_until_millis: (required) This field should always be
+        equal to zero in a `_reindex` response. It only has meaning when
+        using the Task API, where it indicates the next time (in
+        milliseconds since epoch) a throttled request will be executed
+        again in order to conform to `requests_per_second`.
+    :arg total: (required) The number of documents that were successfully
+        processed.
+    :arg version_conflicts: (required) The number of version conflicts
+        that reindex hits.
+    :arg slice_id: The slice ID
+    :arg created: The number of documents that were successfully created.
+    :arg throttled:
+    :arg throttled_until:
+    :arg updated: The number of documents that were successfully updated,
+        for example, a document with same ID already existed prior to
+        reindex updating it.
+    :arg cancelled: The reason for cancellation if the slice was canceled
+    """
+
+    batches: int
+    deleted: int
+    noops: int
+    requests_per_second: float
+    retries: "Retries"
+    throttled_millis: Any
+    throttled_until_millis: Any
+    total: int
+    version_conflicts: int
+    slice_id: int
+    created: int
+    throttled: Any
+    throttled_until: Any
+    updated: int
+    cancelled: str
+
+
 class Retries(AttrDict[Any]):
     """
     :arg bulk: (required) The number of bulk actions retried.

elasticsearch/dsl/utils.py

@@ -611,8 +611,17 @@ class ObjectBase(AttrDict[Any]):
             if skip_empty:
                 # don't serialize empty values
                 # careful not to include numeric zeros
-                if v in ([], {}, None):
-                    continue
+                try:
+                    if v in ([], {}, None):
+                        continue
+                except ValueError:
+                    # the above fails when v is a numpy array
+                    # try using len() instead
+                    try:
+                        if len(v) == 0:
+                            continue
+                    except TypeError:
+                        pass
 
             out[k] = v
         return out