castor-extractor 0.24.2__tar.gz → 0.24.4__tar.gz

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/CHANGELOG.md

@@ -1,5 +1,17 @@
 # Changelog
 
+## 0.24.4 - 2025-03-19
+
+* Snowflake:
+  * improve the list of ignored queries in the query history extraction
+    * ignore the following query types : CALL, COMMENT, EXPLAIN, REFRESH_DYNAMIC_TABLE_AT_REFRESH_VERSION, REVOKE, TRUNCATE_TABLE, UNDROP
+    * ignore queries with empty text
+  * filter out schemas with empty names
+
+## 0.24.3 - 2025-03-18
+
+* Replace ThoughtSpot endpoint `/api/rest/2.0/report/liveboard` with `/api/rest/2.0/metadata/liveboard/data` following the deprecation of the CSV option
+
 ## 0.24.2 - 2025-03-17
 
 * Rename Revamped Tableau Connector classes

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: castor-extractor
-Version: 0.24.2
+Version: 0.24.4
 Summary: Extract your metadata assets.
 Home-page: https://www.castordoc.com/
 License: EULA
@@ -210,6 +210,18 @@ For any questions or bug report, contact us at [support@castordoc.com](mailto:su
 
 # Changelog
 
+## 0.24.4 - 2025-03-19
+
+* Snowflake:
+  * improve the list of ignored queries in the query history extraction
+    * ignore the following query types : CALL, COMMENT, EXPLAIN, REFRESH_DYNAMIC_TABLE_AT_REFRESH_VERSION, REVOKE, TRUNCATE_TABLE, UNDROP
+    * ignore queries with empty text
+  * filter out schemas with empty names
+
+## 0.24.3 - 2025-03-18
+
+* Replace ThoughtSpot endpoint `/api/rest/2.0/report/liveboard` with `/api/rest/2.0/metadata/liveboard/data` following the deprecation of the CSV option
+
 ## 0.24.2 - 2025-03-17
 
 * Rename Revamped Tableau Connector classes

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/castor_extractor/visualization/thoughtspot/client/client.py

@@ -1,13 +1,17 @@
+import logging
 from collections.abc import Iterator
-from typing import Optional
+from functools import partial
+from typing import Iterable, Optional
 
 import requests
+from requests import Response
 
 from ....utils import (
     APIClient,
     BearerAuth,
     RequestSafeMode,
     build_url,
+    fetch_all_pages,
     handle_response,
 )
 from ..assets import (
@@ -19,9 +23,7 @@ from .credentials import (
 from .endpoints import (
     ThoughtspotEndpointFactory,
 )
-from .utils import (
-    usage_liveboard_reader,
-)
+from .pagination import METADATA_BATCH_SIZE, ThoughtSpotPagination
 
 _AUTH_TIMEOUT_S = 60
 _THOUGHTSPOT_HEADERS = {
@@ -29,7 +31,6 @@ _THOUGHTSPOT_HEADERS = {
     "Accept": "application/json",
     "Content-Type": "application/json",
 }
-_METADATA_BATCH_SIZE = 100
 # https://docs.thoughtspot.com/cloud/latest/object-usage-liveboard
 _OBJECT_USAGE_LIVEBOARD = "Object Usage"
 _ANSWER_USAGE_VIZ = "Answer Usage, by User"
@@ -40,6 +41,9 @@ _LIVEBOARD_USAGE_VIZ = "Popular Liveboards Last 30 Days"
 THOUGHTSPOT_SAFE_MODE = RequestSafeMode()
 
 
+logger = logging.getLogger(__name__)
+
+
 class ThoughtspotBearerAuth(BearerAuth):
     def __init__(self, host: str, token_payload: dict[str, str]):
         auth_endpoint = ThoughtspotEndpointFactory.authentication()
@@ -86,7 +90,7 @@ class ThoughtspotClient(APIClient):
             search_filters = {
                 "metadata": [{"type": metadata_type}],
                 "include_details": True,
-                "record_size": _METADATA_BATCH_SIZE,
+                "record_size": METADATA_BATCH_SIZE,
                 "record_offset": offset,
             }
             if identifier:
@@ -100,9 +104,9 @@ class ThoughtspotClient(APIClient):
                 data=search_filters,
             )
             yield from metadata
-            if len(metadata) < _METADATA_BATCH_SIZE:
+            if len(metadata) < METADATA_BATCH_SIZE:
                 break
-            offset = offset + _METADATA_BATCH_SIZE
+            offset = offset + METADATA_BATCH_SIZE
 
     def _get_all_answers(self) -> Iterator[dict]:
         yield from self._metadata_search(metadata_type="ANSWER")
@@ -120,7 +124,7 @@ class ThoughtspotClient(APIClient):
         self,
         liveboard_name: str,
         visualization_name: str,
-    ) -> Iterator[dict]:
+    ) -> Iterator[list[list]]:
         """
         Yields the data of a given visualization in the given liveboard.
         ThoughtSpot maintains two system liveboards with stats about data usage,
@@ -133,29 +137,62 @@ class ThoughtspotClient(APIClient):
         )
         liveboard_id = usage_liveboard["metadata_id"]
 
-        data = self._post(
-            endpoint=ThoughtspotEndpointFactory.liveboard(),
-            headers={"Accept": "application/octet-stream"},
+        def handler(response: Response) -> dict:
+            response_dict = response.json()
+            contents = response_dict.get("contents", [])
+            if not contents:
+                logger.warning("No data found in response")
+                return dict()
+            return contents[0]
+
+        request = partial(
+            self._post,
+            endpoint=ThoughtspotEndpointFactory.liveboard_data(),
             data={
                 "metadata_identifier": liveboard_id,
-                "file_format": "CSV",
                 "visualization_identifiers": [visualization_name],
+                "record_offset": 0,
+                "record_size": METADATA_BATCH_SIZE,
             },
-            handler=lambda x: x.text,
+            handler=handler,
         )
-        yield from usage_liveboard_reader(data)
+        yield from fetch_all_pages(request, ThoughtSpotPagination)
 
     def _get_answer_usages(self) -> Iterator[dict]:
-        return self._get_usages(
+        """
+        Returns the usage data of saved Answers, which is found in a visualization
+        of the "Object Usage" liveboard.
+        Each data row returned by the API is transformed from a list into a dictionary.
+        The columns are explicitly listed here because in the API response,
+        there is a mismatch between the number of column names and the number
+        of values per data row.
+        """
+        data: Iterable[list[list]] = self._get_usages(
             liveboard_name=_OBJECT_USAGE_LIVEBOARD,
             visualization_name=_ANSWER_USAGE_VIZ,
         )
+        columns = (
+            "Answer name",
+            "Number of unique users",
+            "Count of object interactions",
+        )
+        for row in data:
+            yield dict(zip(columns, row))
 
     def _get_liveboards_usages(self) -> Iterator[dict]:
-        return self._get_usages(
+        """
+        Returns the usage data of Liveboards, which is found in a visualization
+        of the "User Adoption" liveboard.
+        Each data row returned by the API is transformed from a list into a dictionary.
+        See `_get_answer_usages` regarding the columns list.
+        """
+        data: Iterable[list[list]] = self._get_usages(
             liveboard_name=_USER_ADOPTION_LIVEBOARD,
             visualization_name=_LIVEBOARD_USAGE_VIZ,
         )
+        columns = ("Pinboard", "Unique Number of User", "Pinboard Views")
+        for row in data:
+            yield dict(zip(columns, row))
 
     def fetch(self, asset: ThoughtspotAsset) -> Iterator[dict]:
         if asset == ThoughtspotAsset.ANSWERS:

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/castor_extractor/visualization/thoughtspot/client/endpoints.py

@@ -8,5 +8,5 @@ class ThoughtspotEndpointFactory:
         return "api/rest/2.0/metadata/search"
 
     @classmethod
-    def liveboard(cls) -> str:
-        return "api/rest/2.0/report/liveboard"
+    def liveboard_data(cls) -> str:
+        return "api/rest/2.0/metadata/liveboard/data"

castor_extractor-0.24.4/castor_extractor/visualization/thoughtspot/client/pagination.py

@@ -0,0 +1,25 @@
+from pydantic import ConfigDict, Field
+
+from ....utils import PaginationModel
+
+METADATA_BATCH_SIZE = 100
+
+
+class ThoughtSpotPagination(PaginationModel):
+    data_rows: list = Field(default_factory=list)
+    record_offset: int
+    record_size: int
+
+    model_config = ConfigDict(
+        populate_by_name=True,
+        from_attributes=True,
+    )
+
+    def is_last(self) -> bool:
+        return len(self.data_rows) < METADATA_BATCH_SIZE
+
+    def next_page_payload(self) -> dict:
+        return {"record_offset": self.record_offset + METADATA_BATCH_SIZE}
+
+    def page_results(self) -> list:
+        return self.data_rows

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/castor_extractor/warehouse/snowflake/queries/column.sql

@@ -47,7 +47,9 @@ FROM snowflake.account_usage.columns AS c
     JOIN snowflake.account_usage.tables AS t ON t.table_id = c.table_id
     JOIN tags_agg_columns ta ON c.column_id = ta.column_id
 WHERE TRUE
-    AND COALESCE(c.column_name, '') != ''
+    AND TRIM(COALESCE(c.column_name, '')) != ''
+    AND TRIM(COALESCE(t.table_name, '')) != ''
+    AND TRIM(COALESCE(s.schema_name, '')) != ''
     AND UPPER(c.table_catalog) NOT IN ('SNOWFLAKE', 'UTIL_DB')
     AND (
         c.deleted IS NULL

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/castor_extractor/warehouse/snowflake/queries/query.sql

@@ -51,20 +51,28 @@ WHERE TRUE
     AND HOUR(CONVERT_TIMEZONE('UTC', start_time)) BETWEEN :hour_min AND :hour_max
     AND execution_status = 'SUCCESS'
     AND query_text != 'SELECT 1'
+    AND TRIM(COALESCE(query_text, '')) != ''
     AND query_type NOT IN (
-        'SHOW',
-        'USE',
-        'ROLLBACK',
-        'DESCRIBE',
         'ALTER_SESSION',
-        'PUT_FILES',
+        'BEGIN_TRANSACTION',
+        'CALL',
+        'COMMENT',
+        'COMMIT',
         'CREATE', -- create objects: stage|function|schema|procedure|file|storage|pipe|notification integration
-        'SET',
+        'DESCRIBE',
+        'DROP',
+        'EXPLAIN',
+        'GET_FILES',
         'GRANT',
-        'COMMIT',
+        'PUT_FILES',
+        'REFRESH_DYNAMIC_TABLE_AT_REFRESH_VERSION',
+        'REMOVE_FILES',
+        'REVOKE',
+        'ROLLBACK',
+        'SET',
+        'SHOW',
+        'TRUNCATE_TABLE',
+        'UNDROP',
         'UNLOAD',
-        'GET_FILES',
-        'DROP',
-        'BEGIN_TRANSACTION',
-        'REMOVE_FILES'
+        'USE'
     )

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/castor_extractor/warehouse/snowflake/queries/schema.sql

@@ -16,6 +16,7 @@ WHERE TRUE
         deleted IS NULL
         OR deleted > CURRENT_TIMESTAMP - INTERVAL '1 day'
     )
+    AND TRIM(COALESCE(schema_name, '')) != ''
     {database_allowed}
     {database_blocked}
     AND CASE {has_fetch_transient} WHEN FALSE THEN NOT s.is_transient::BOOLEAN ELSE TRUE END

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/castor_extractor/warehouse/snowflake/queries/table.sql

@@ -41,8 +41,8 @@ FROM snowflake.account_usage.tables AS t
     JOIN snowflake.account_usage.schemata AS s ON s.schema_id = t.table_schema_id
     JOIN tags_agg_tables ta ON t.table_id = ta.table_id
 WHERE TRUE
-    AND t.table_name IS NOT NULL
-    AND t.table_name != ''
+    AND TRIM(COALESCE(t.table_name, '')) != ''
+    AND TRIM(COALESCE(s.schema_name, '')) != ''
     AND UPPER(t.table_catalog) NOT IN ('SNOWFLAKE', 'UTIL_DB')
     AND (
         t.deleted IS NULL

{castor_extractor-0.24.2 → castor_extractor-0.24.4}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools>=61.2"]
 
 [tool.poetry]
 name = "castor-extractor"
-version = "0.24.2"
+version = "0.24.4"
 description = "Extract your metadata assets."
 authors = ["Castor <support@castordoc.com>"]
 license = "EULA"

castor_extractor-0.24.2/castor_extractor/visualization/thoughtspot/client/utils.py

@@ -1,31 +0,0 @@
-import csv
-import re
-from collections.abc import Iterator
-from io import StringIO
-
-_END_OF_GENERATED_TEXT = r'^""$'
-
-
-def usage_liveboard_reader(usage_liveboard_csv: str) -> Iterator[dict]:
-    """
-    Converts a CSV string into an iterator of dictionaries after
-    ignoring the generated text that preceeds the actual CSV header row.
-    The generated block ends with a row containing only two double quotes.
-    Here is an example:
-
-        "Data extract produced by Castor on 09/19/2024 06:54"
-        "Filters applied on data :"
-        "User Action IN [pinboard_embed_view,pinboard_tspublic_no_runtime_filter,pinboard_tspublic_runtime_filter,pinboard_view]"
-        "Pinboard NOT IN [mlm - availability pinboard,null]"
-        "Timestamp >= 20240820 00:00:00 < 20240919 00:00:00"
-        "Timestamp >= 20240919 00:00:00 < 20240920 00:00:00"
-        ""
-
-    """
-    csv_file = StringIO(usage_liveboard_csv)
-
-    line = next(csv_file)
-    while not re.match(_END_OF_GENERATED_TEXT, line.strip()):
-        line = next(csv_file)
-
-    yield from csv.DictReader(csv_file)

castor_extractor-0.24.2/castor_extractor/visualization/thoughtspot/client/utils_test.py

@@ -1,75 +0,0 @@
-from .utils import (
-    usage_liveboard_reader,
-)
-
-VALID_CSV_1 = '''"Data extract produced by Castor on 09/19/2024 06:54"
-"Filters applied on data :"
-"User Action IN [pinboard_embed_view,pinboard_tspublic_no_runtime_filter,pinboard_tspublic_runtime_filter,pinboard_view]"
-"Pinboard NOT IN [mlm - availability pinboard,null]"
-"Timestamp >= 20240820 00:00:00 < 20240919 00:00:00"
-"Timestamp >= 20240919 00:00:00 < 20240920 00:00:00"
-""
-"Pinboard","Pinboard Views","Unique Number of User"
-"Market Report","559","19"
-"Retailer report","204","14"
-"Second-hand market","72","6"
-"September test","25","2"'''
-
-
-VALID_CSV_2 = '''"Data extract produced by Castor on 01/07/2025 16:07"
-"Filters applied on data :"
-"Timestamp >= 20241208 00:00:00 < 20250107 00:00:00"
-""
-"Answer name","User name","Number of unique users","Count of object interactions"
-"toto","tata","1","666"'''
-
-# Invalid CSV input (missing data rows)
-INVALID_CSV = '''"Data extract produced by Castor on 09/19/2024 06:54"
-"Filters applied on data :"
-"User Action IN [pinboard_embed_view,pinboard_tspublic_no_runtime_filter,pinboard_tspublic_runtime_filter,pinboard_view]"
-"Pinboard NOT IN [mlm - availability pinboard,null]"
-"Timestamp >= 20240820 00:00:00 < 20240919 00:00:00"
-"Timestamp >= 20240919 00:00:00 < 20240920 00:00:00"
-""'''
-
-
-def test_usage_liveboard_reader():
-    expected_output_1 = [
-        {
-            "Pinboard": "Market Report",
-            "Pinboard Views": "559",
-            "Unique Number of User": "19",
-        },
-        {
-            "Pinboard": "Retailer report",
-            "Pinboard Views": "204",
-            "Unique Number of User": "14",
-        },
-        {
-            "Pinboard": "Second-hand market",
-            "Pinboard Views": "72",
-            "Unique Number of User": "6",
-        },
-        {
-            "Pinboard": "September test",
-            "Pinboard Views": "25",
-            "Unique Number of User": "2",
-        },
-    ]
-    expected_output_2 = [
-        {
-            "Answer name": "toto",
-            "User name": "tata",
-            "Number of unique users": "1",
-            "Count of object interactions": "666",
-        }
-    ]
-
-    result = list(usage_liveboard_reader(VALID_CSV_1))
-    assert result == expected_output_1
-
-    result = list(usage_liveboard_reader(VALID_CSV_2))
-    assert result == expected_output_2
-
-    result = list(usage_liveboard_reader(INVALID_CSV))
-    assert result == []  # Expect an empty result since there is no data