Bump to release 2.3.0

HISTORY.md

@@ -1,3 +1,21 @@
+## 2.3.0 (June 19, 2024)
+
+### Breaking changes
+
+- Introduced a new pagination approach in `AssetClient.search()` and `FluentSearch.execute()` called **bulk search** (disabled by default). It minimizes system impact when handling large result sets. The SDK switches to this search operation automatically if results exceed a predefined threshold (i.e: `100,000` results). Alternatively, users can enable bulk search explicitly by setting `bulk=True` in `AssetClient.search()` or `FluentSearch.execute()`. One side effect of this search operation is that any requested sorting of the results will be overridden. If you need to sort such a large number of results, you will have to do so after retrieving them.
+
+- The `AssetClient.search()` and `FluentSearch.execute()` methods will now raise an exception (`InvalidRequestError`) in the following scenarios:
+
+  - when bulk search is enabled (`bulk=True`) and any user-specified sorting options are found in the search request.
+
+  - when bulk search is disabled (`bulk=False`), the number of results exceeds the predefined threshold (i.e: `100,000` assets), and any user-specified sorting options are found in the search request.
+
+  _This is because the bulk search approach ignores user-specified sorting and instead reorders the results based on the creation timestamps of assets to handle large numbers of assets efficiently._
+
+### QOL improvements
+
+- Pinned `urllib3>=1.26.0,<3` and moved `networkx` to the dev requirements to avoid potential version mismatches.
+
 ## 2.2.4 (June 11, 2024)
 
 ### New features

pyatlan/client/asset.py

@@ -158,15 +158,26 @@ def _prepare_sorts_for_bulk_search(sorts: List[SortItem]):
     def search(self, criteria: IndexSearchRequest, bulk=False) -> IndexSearchResults:
         """
         Search for assets using the provided criteria.
+        `Note:` if the number of results exceeds the predefined threshold
+        (100,000 assets) this will be automatically converted into a `bulk` search.
 
         :param criteria: detailing the search query, parameters, and so on to run
         :param bulk: whether to run the search to retrieve assets that match the supplied criteria,
         for large numbers of results (> `100,000`), defaults to `False`. Note: this will reorder the results
         (based on creation timestamp) in order to iterate through a large number (more than `100,000`) results.
-        :returns: the results of the search
+        :raises InvalidRequestError:
+
+            - if bulk search is enabled (`bulk=True`) and any
+              user-specified sorting options are found in the search request.
+            - if bulk search is disabled (`bulk=False`) and the number of results
+              exceeds the predefined threshold (i.e: `100,000` assets)
+              and any user-specified sorting options are found in the search request.
+
         :raises AtlanError: on any API communication issue
+        :returns: the results of the search
         """
         if bulk:
+            # If there is any user-specified sorting present in the search request
             if criteria.dsl.sort and len(criteria.dsl.sort) > 1:
                 raise ErrorCode.UNABLE_TO_RUN_BULK_WITH_SORTS.exception_with_parameters()
             criteria.dsl.sort = self._prepare_sorts_for_bulk_search(criteria.dsl.sort)
@@ -190,10 +201,17 @@ def search(self, criteria: IndexSearchRequest, bulk=False) -> IndexSearchResults
         aggregations = self._get_aggregations(raw_json)
         count = raw_json.get("approximateCount", 0)
 
-        if not bulk and count > IndexSearchResults._MASS_EXTRACT_THRESHOLD:
-            raise ErrorCode.ENABLE_BULK_FOR_MASS_EXTRACTION.exception_with_parameters(
-                IndexSearchResults._MASS_EXTRACT_THRESHOLD
-            )
+        if (
+            count > IndexSearchResults._MASS_EXTRACT_THRESHOLD
+            and not IndexSearchResults.presorted_by_timestamp(criteria.dsl.sort)
+        ):
+            # If there is any user-specified sorting present in the search request
+            if criteria.dsl.sort and len(criteria.dsl.sort) > 1:
+                raise ErrorCode.UNABLE_TO_RUN_BULK_WITH_SORTS.exception_with_parameters()
+            # Re-fetch the first page results with updated timestamp sorting
+            # for bulk search if count > _MASS_EXTRACT_THRESHOLD (100,000 assets)
+            criteria.dsl.sort = self._prepare_sorts_for_bulk_search(criteria.dsl.sort)
+            return self.search(criteria)
 
         return IndexSearchResults(
             client=self._client,
@@ -1839,6 +1857,16 @@ def _prepare_query_for_timestamp_paging(self, query: Query):
             self._criteria.dsl.from_ = 0  # type: ignore[attr-defined]
             self._criteria.dsl.query = rewritten_query  # type: ignore[attr-defined]
 
+    def _get_bulk_search_log_message(self):
+        return (
+            (
+                "Bulk search option is enabled. "
+                if self._bulk
+                else "Result size (%s) exceeds threshold (%s). "
+            )
+            + "Ignoring requests for offset-based paging and using timestamp-based paging instead."
+        )
+
     def _get_next_page(self):
         """
         Fetches the next page of results.
@@ -1848,14 +1876,18 @@ def _get_next_page(self):
         query = self._criteria.dsl.query
         self._criteria.dsl.size = self._size
         self._criteria.dsl.from_ = self._start
+        is_bulk_search = (
+            self._bulk or self._approximate_count > self._MASS_EXTRACT_THRESHOLD
+        )
 
-        if self._bulk:
+        if is_bulk_search:
             LOGGER.debug(
-                "Bulk search option is enabled. Ignoring requests for default offset-based "
-                "paging and switching to a creation timestamp-based paging approach."
+                self._get_bulk_search_log_message(),
+                self._approximate_count,
+                self._MASS_EXTRACT_THRESHOLD,
             )
             self._prepare_query_for_timestamp_paging(query)
-        if raw_json := super()._get_next_page_json(self._bulk):
+        if raw_json := super()._get_next_page_json(is_bulk_search):
             self._count = raw_json.get("approximateCount", 0)
             return True
         return False

pyatlan/errors.py

@@ -544,13 +544,6 @@ class ErrorCode(Enum):
         "Please ensure that no sorting options are included in your search request when performing a bulk search.",
         InvalidRequestError,
     )
-    ENABLE_BULK_FOR_MASS_EXTRACTION = (
-        400,
-        "ATLAN-PYTHON-400-063",
-        "Number of results exceeds the predefined threshold {}. Please execute the search again with `bulk=True`.",
-        "Please note that this will reorder the results based on creation timestamp to efficiently handle a large number of results.",  # noqa
-        InvalidRequestError,
-    )
     AUTHENTICATION_PASSTHROUGH = (
         401,
         "ATLAN-PYTHON-401-000",

pyatlan/model/fluent_search.py

@@ -402,11 +402,22 @@ def count(self, client: AtlanClient) -> int:
     def execute(self, client: AtlanClient, bulk: bool = False) -> IndexSearchResults:
         """
         Run the fluent search to retrieve assets that match the supplied criteria.
+        `Note:` if the number of results exceeds the predefined threshold
+        (100,000 assets) this will be automatically converted into a `bulk` search.
 
         :param client: client through which to retrieve the assets.
         :param bulk: whether to run the search to retrieve assets that match the supplied criteria,
         for large numbers of results (> `100,000`), defaults to `False`. Note: this will reorder the results
         (based on creation timestamp) in order to iterate through a large number (more than `100,000`) results.
+        :raises InvalidRequestError:
+
+            - if bulk search is enabled (`bulk=True`) and any
+              user-specified sorting options are found in the search request.
+            - if bulk search is disabled (`bulk=False`) and the number of results
+              exceeds the predefined threshold (i.e: `100,000` assets)
+              and any user-specified sorting options are found in the search request.
+
+        :raises AtlanError: on any API communication issue
         :returns: an iterable list of assets that match the supplied criteria, lazily-fetched
         """
         return client.asset.search(criteria=self.to_request(), bulk=bulk)

pyatlan/model/search.py

@@ -1798,7 +1798,7 @@ def validate_order(cls, v, values):
 
 class DSL(AtlanObject):
     from_: int = Field(default=0, alias="from")
-    size: int = Field(default=100)
+    size: int = Field(default=300)
     aggregations: Dict[str, Aggregation] = Field(default_factory=dict)
     track_total_hits: Optional[bool] = Field(default=True, alias="track_total_hits")
     post_filter: Optional[Query] = Field(default=None, alias="post_filter")

pyatlan/version.txt

@@ -1 +1 @@
-2.2.4
+2.3.0

tests/integration/test_index_search.py

@@ -8,7 +8,7 @@
 
 import pytest
 
-from pyatlan.client.asset import LOGGER
+from pyatlan.client.asset import LOGGER, IndexSearchResults
 from pyatlan.client.atlan import AtlanClient
 from pyatlan.model.assets import Asset, Table
 from pyatlan.model.assets.atlas_glossary_term import AtlasGlossaryTerm
@@ -249,6 +249,28 @@ def test_search_pagination(mock_logger, client: AtlanClient):
     assert "Bulk search option is enabled." in mock_logger.call_args_list[0][0][0]
     mock_logger.reset_mock()
 
+    # Test search(): when the number of results exceeds the predefined threshold,
+    # the SDK automatically switches to a `bulk` search option using timestamp-based pagination.
+    with patch.object(IndexSearchResults, "_MASS_EXTRACT_THRESHOLD", 1):
+        request = (
+            FluentSearch()
+            .where(CompoundQuery.active_assets())
+            .where(CompoundQuery.asset_type(AtlasGlossaryTerm))
+            .page_size(2)
+        ).to_request()
+        results = client.asset.search(criteria=request)
+        expected_sorts = [
+            Asset.CREATE_TIME.order(SortOrder.ASCENDING),
+            Asset.GUID.order(SortOrder.ASCENDING),
+        ]
+        _assert_search_results(results, expected_sorts, size, TOTAL_ASSETS)
+        assert mock_logger.call_count < TOTAL_ASSETS
+        assert (
+            "Result size (%s) exceeds threshold (%s)."
+            in mock_logger.call_args_list[0][0][0]
+        )
+        mock_logger.reset_mock()
+
 
 def test_search_iter(client: AtlanClient):
     size = 15

tests/unit/test_client.py

@@ -1460,8 +1460,8 @@ def test_index_search_pagination(
     mock_logger.reset_mock()
     mock_api_caller.reset_mock()
 
-    # Test search(): Raise an exception suggesting the user switch to bulk search
-    # when the number of results exceeds the predefined threshold
+    # Test search(): when the number of results exceeds the predefined threshold
+    # it will automatically convert to a `bulk` search.
     TEST_THRESHOLD = 1
     with patch.object(IndexSearchResults, "_MASS_EXTRACT_THRESHOLD", TEST_THRESHOLD):
         mock_api_caller._call_api.side_effect = [
@@ -1477,17 +1477,52 @@ def test_index_search_pagination(
             .where(CompoundQuery.asset_type(AtlasGlossaryTerm))
             .page_size(2)
         ).to_request()
+        results = client.search(criteria=request)
+        expected_sorts = [
+            Asset.CREATE_TIME.order(SortOrder.ASCENDING),
+            Asset.GUID.order(SortOrder.ASCENDING),
+        ]
+        _assert_search_results(results, index_search_paging_json, expected_sorts)
+        assert mock_api_caller._call_api.call_count == 3
+        assert mock_logger.call_count == 1
+        assert (
+            "Result size (%s) exceeds threshold (%s)"
+            in mock_logger.call_args_list[0][0][0]
+        )
+    mock_logger.reset_mock()
+    mock_api_caller.reset_mock()
+
+    # Test search(bulk=False): Raise an exception when the number of results exceeds
+    # the predefined threshold and there are any user-defined sorting options present
+    with patch.object(IndexSearchResults, "_MASS_EXTRACT_THRESHOLD", TEST_THRESHOLD):
+        mock_api_caller._call_api.side_effect = [
+            index_search_paging_json,
+        ]
+        request = (
+            FluentSearch()
+            .where(CompoundQuery.active_assets())
+            .where(CompoundQuery.asset_type(AtlasGlossaryTerm))
+            .page_size(2)
+            # With some sort options
+            .sort(Asset.NAME.order(SortOrder.ASCENDING))
+        ).to_request()
+
         with pytest.raises(
             InvalidRequestError,
             match=(
-                "ATLAN-PYTHON-400-063 Number of results exceeds the predefined threshold "
-                f"{TEST_THRESHOLD}. Please execute the search again with `bulk=True`."
+                "ATLAN-PYTHON-400-063 Unable to execute "
+                "bulk search with user-defined sorting options. "
+                "Suggestion: Please ensure that no sorting options are "
+                "included in your search request when performing a bulk search."
             ),
         ):
-            results = client.search(criteria=request)
+            client.search(criteria=request)
+            assert mock_api_caller._call_api.call_count == 1
+    mock_api_caller.reset_mock()
+    mock_api_caller.reset_mock()
 
-    # Test search(): Raise an exception when
-    # bulk search is attempted with any user-defined sorting options
+    # Test search(bulk=True): Raise an exception when bulk search is enabled
+    # and there are any user-defined sorting options present
     request = (
         FluentSearch()
         .where(CompoundQuery.active_assets())

tests/unit/test_search_model.py

@@ -267,7 +267,7 @@ def test_dsl():
     )
     assert (
         dsl.json(by_alias=True, exclude_none=True)
-        == '{"from": 0, "size": 100, "aggregations": {}, "track_total_hits": true, '
+        == '{"from": 0, "size": 300, "aggregations": {}, "track_total_hits": true, '
         '"post_filter": {"term": {"databaseName.keyword": '
         '{"value": "ATLAN_SAMPLE_DATA"}}}, "query": {"term": '
         '{"__typeName.keyword": {"value": "Schema"}}}, "sort": []}'
@@ -283,7 +283,7 @@ def test_index_search_request():
     assert (
         request.json(by_alias=True, exclude_none=True)
         == '{"attributes": ["schemaName", "databaseName"],'
-        ' "dsl": {"from": 0, "size": 100, "aggregations": {}, "track_total_hits": true, '
+        ' "dsl": {"from": 0, "size": 300, "aggregations": {}, "track_total_hits": true, '
         '"post_filter": {"term": {"databaseName.keyword": '
         '{"value": "ATLAN_SAMPLE_DATA"}}}, "query": {"term": {"__typeName.keyword": {"value": "Schema"}}}, '
         '"sort": [{"__guid": {"order": "asc"}}]}, "relationAttributes": [], '
@@ -300,7 +300,7 @@ def test_audit_search_request():
     assert (
         request.json(by_alias=True, exclude_none=True)
         == '{"attributes": ["schemaName", "databaseName"],'
-        ' "dsl": {"from": 0, "size": 100, "aggregations": {}, "track_total_hits": true, '
+        ' "dsl": {"from": 0, "size": 300, "aggregations": {}, "track_total_hits": true, '
         '"post_filter": {"term": {"databaseName.keyword": '
         '{"value": "ATLAN_SAMPLE_DATA"}}}, "query": {"term": {"__typeName.keyword": {"value": "Schema"}}}, '
         '"sort": [{"entityId": {"order": "asc"}}]}}'
@@ -316,7 +316,7 @@ def test_search_log_request():
     assert (
         request.json(by_alias=True, exclude_none=True)
         == '{"attributes": ["schemaName", "databaseName"],'
-        ' "dsl": {"from": 0, "size": 100, "aggregations": {}, "track_total_hits": true, '
+        ' "dsl": {"from": 0, "size": 300, "aggregations": {}, "track_total_hits": true, '
         '"post_filter": {"term": {"databaseName.keyword": '
         '{"value": "ATLAN_SAMPLE_DATA"}}}, "query": {"term": {"__typeName.keyword": {"value": "Schema"}}}, '
         '"sort": [{"entityGuidsAll": {"order": "asc"}}]}}'