castor-extractor 0.24.4__tar.gz → 0.24.7__tar.gz

{castor_extractor-0.24.4 → castor_extractor-0.24.7}/CHANGELOG.md

@@ -1,5 +1,17 @@
 # Changelog
 
+## 0.24.7 - 2025-04-07
+
+* Tableau - switch from `cursor` to `offset` pagination to mitigate timeout issues
+
+## 0.24.6 - 2025-04-03
+
+* Domo - extract cards metadata by batch to prevent from hitting URL max length
+
+## 0.24.5 - 2025-04-02
+
+* bump dependencies: google-cloud-storage
+
 ## 0.24.4 - 2025-03-19
 
 * Snowflake:

{castor_extractor-0.24.4 → castor_extractor-0.24.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: castor-extractor
-Version: 0.24.4
+Version: 0.24.7
 Summary: Extract your metadata assets.
 Home-page: https://www.castordoc.com/
 License: EULA
@@ -35,7 +35,7 @@ Requires-Dist: google-api-core (>=2.1.1,<3.0.0)
 Requires-Dist: google-api-python-client (>=2.121.0,<3.0.0) ; extra == "lookerstudio" or extra == "all"
 Requires-Dist: google-auth (>=2,<3)
 Requires-Dist: google-cloud-core (>=2.1.0,<3.0.0)
-Requires-Dist: google-cloud-storage (>=2,<3)
+Requires-Dist: google-cloud-storage (>=3.1.0,<4.0.0)
 Requires-Dist: google-resumable-media (>=2.0.3,<3.0.0)
 Requires-Dist: googleapis-common-protos (>=1.53.0,<2.0.0)
 Requires-Dist: looker-sdk (>=25.0.0,<26.0.0) ; extra == "looker" or extra == "all"
@@ -51,7 +51,7 @@ Requires-Dist: pymssql (>=2.2.11,<3.0.0) ; extra == "sqlserver" or extra == "all
 Requires-Dist: pymysql[rsa] (>=1.1.0,<2.0.0) ; extra == "mysql" or extra == "all"
 Requires-Dist: python-dateutil (>=2.0.0,<=3.0.0)
 Requires-Dist: requests (>=2.0.0,<3.0.0)
-Requires-Dist: setuptools (>=75.6)
+Requires-Dist: setuptools (>=78.1)
 Requires-Dist: snowflake-connector-python (>=3.4.0,<4.0.0) ; extra == "snowflake" or extra == "all"
 Requires-Dist: snowflake-sqlalchemy (!=1.2.5,<2.0.0) ; extra == "snowflake" or extra == "all"
 Requires-Dist: sqlalchemy (>=1.4,<1.5)
@@ -210,6 +210,18 @@ For any questions or bug report, contact us at [support@castordoc.com](mailto:su
 
 # Changelog
 
+## 0.24.7 - 2025-04-07
+
+* Tableau - switch from `cursor` to `offset` pagination to mitigate timeout issues
+
+## 0.24.6 - 2025-04-03
+
+* Domo - extract cards metadata by batch to prevent from hitting URL max length
+
+## 0.24.5 - 2025-04-02
+
+* bump dependencies: google-cloud-storage
+
 ## 0.24.4 - 2025-03-19
 
 * Snowflake:

{castor_extractor-0.24.4 → castor_extractor-0.24.7}/castor_extractor/utils/__init__.py

@@ -1,4 +1,5 @@
 from .argument_parser import parse_filled_arguments
+from .batch import batch_of_length
 from .client import (
     AbstractSourceClient,
     APIClient,

castor_extractor-0.24.7/castor_extractor/utils/batch.py

@@ -0,0 +1,16 @@
+from typing import Iterator, List, TypeVar
+
+T = TypeVar("T")
+
+
+def batch_of_length(
+    elements: List[T],
+    batch_size: int,
+) -> Iterator[List[T]]:
+    """
+    Split the given elements into smaller chunks
+    """
+    assert batch_size > 1, "batch size must be greater or equal to 1"
+    element_count = len(elements)
+    for index in range(0, element_count, batch_size):
+        yield elements[index : min((index + batch_size), element_count)]

castor_extractor-0.24.7/castor_extractor/utils/batch_test.py

@@ -0,0 +1,27 @@
+import pytest
+
+from .batch import batch_of_length
+
+
+def test_batch_of_length():
+    elements = ["a", "b", "c", "d", "e", "f", "g", "h"]
+    result = list(batch_of_length(elements, 3))
+    assert result == [
+        ["a", "b", "c"],
+        ["d", "e", "f"],
+        ["g", "h"],
+    ]
+
+    result = list(batch_of_length(elements, 1000))
+    assert result == [
+        elements,
+    ]
+
+    result = list(batch_of_length(elements, 7))
+    assert result == [
+        ["a", "b", "c", "d", "e", "f", "g"],
+        ["h"],
+    ]
+
+    with pytest.raises(AssertionError):
+        list(batch_of_length(elements, -12))

{castor_extractor-0.24.4 → castor_extractor-0.24.7}/castor_extractor/visualization/domo/client/client.py

@@ -9,6 +9,7 @@ import requests
 from ....utils import (
     RequestSafeMode,
     at_midnight,
+    batch_of_length,
     current_date,
     empty_iterator,
     handle_response,
@@ -48,6 +49,8 @@ _RETRY_BASE_MS = 10 * 60 * 1000  # 10 minutes
 
 _PARENT_FOLDER = "/Dashboards"
 
+_CARDS_BATCH_SIZE = 100
+
 logger = logging.getLogger(__name__)
 
 
@@ -156,16 +159,19 @@ class DomoClient:
 
         return all_results
 
+    def _cards_metadata(self, card_ids: list[int]) -> Iterator[dict]:
+        # batch to avoid hitting the URL max length
+        for batch_card_ids in batch_of_length(card_ids, _CARDS_BATCH_SIZE):
+            endpoint = self._endpoint_factory.cards_metadata(batch_card_ids)
+            yield from self._get_element(endpoint)
+
     def _datasources(self, card_ids: list[int]) -> RawData:
         """Yields all distinct datasources associated to the given cards"""
         if not card_ids:
             return empty_iterator()
 
-        endpoint = self._endpoint_factory.cards_metadata(card_ids)
-        cards_metadata = self._get_element(endpoint)
-
         processed: set[str] = set()
-        for card in cards_metadata:
+        for card in self._cards_metadata(card_ids):
             for datasource in card["datasources"]:
                 id_ = datasource["dataSourceId"]
                 if id_ in processed:

{castor_extractor-0.24.4 → castor_extractor-0.24.7}/castor_extractor/visualization/tableau/client/client_metadata_api.py

@@ -1,3 +1,4 @@
+import logging
 from collections.abc import Iterator
 from typing import Optional
 
@@ -9,15 +10,14 @@ from ..constants import DEFAULT_PAGE_SIZE
 from .errors import TableauApiError, TableauApiTimeout
 from .gql_queries import FIELDS_QUERIES, GQL_QUERIES, QUERY_TEMPLATE
 
+logger = logging.getLogger(__name__)
+
 # increase the value when extraction is too slow
 # decrease the value when timeouts arise
 _CUSTOM_PAGE_SIZE: dict[TableauAsset, int] = {
-    # for some clients, extraction of columns tend to hit the node limit
-    # https://community.tableau.com/s/question/0D54T00000YuK60SAF/metadata-query-nodelimitexceeded-error
-    # the workaround is to reduce pagination
-    TableauAsset.COLUMN: 50,
     # fields are light but volumes are bigger
     TableauAsset.FIELD: 1000,
+    # tables are sometimes heavy
     TableauAsset.TABLE: 50,
 }
 
@@ -51,8 +51,9 @@ def _check_errors(answer: dict) -> None:
 
 def gql_query_scroll(
     server,
-    query: str,
     resource: str,
+    fields: str,
+    page_size: int,
 ) -> Iterator[SerializedAsset]:
     """
     Iterate over GQL query results, handling pagination and cursor
@@ -67,23 +68,27 @@ def gql_query_scroll(
         max_retries=_RETRY_COUNT,
         base_ms=_RETRY_BASE_MS,
     )
-    def _call(cursor: Optional[str]) -> dict:
-        # If cursor is defined it must be quoted else use null token
-        token = "null" if cursor is None else f'"{cursor}"'
-        query_ = query.replace("AFTER_TOKEN_SIGNAL", token)
-        answer = server.metadata.query(query_)
+    def _call(first: int, offset: int) -> dict:
+        query = QUERY_TEMPLATE.format(
+            resource=resource,
+            fields=fields,
+            first=first,
+            offset=offset,
+        )
+        answer = server.metadata.query(query)
         _check_errors(answer)
         return answer["data"][f"{resource}Connection"]
 
-    cursor = None
+    current_offset = 0
     while True:
-        payload = _call(cursor)
+        payload = _call(first=page_size, offset=current_offset)
         yield payload["nodes"]
 
-        page_info = payload["pageInfo"]
-        if page_info["hasNextPage"]:
-            cursor = page_info["endCursor"]
-        else:
+        current_offset += len(payload["nodes"])
+        total = payload["totalCount"]
+        logger.info(f"Extracted {current_offset}/{total} {resource}")
+
+        if not payload["pageInfo"]["hasNextPage"]:
             break
 
 
@@ -107,12 +112,12 @@ class TableauClientMetadataApi:
         fields: str,
         page_size: int = DEFAULT_PAGE_SIZE,
     ) -> SerializedAsset:
-        query = QUERY_TEMPLATE.format(
+        result_pages = gql_query_scroll(
+            self._server,
             resource=resource,
             fields=fields,
             page_size=page_size,
         )
-        result_pages = gql_query_scroll(self._server, query, resource)
         return [asset for page in result_pages for asset in page]
 
     def _page_size(self, asset: TableauAsset) -> int:

{castor_extractor-0.24.4 → castor_extractor-0.24.7}/castor_extractor/visualization/tableau/client/gql_queries.py

@@ -2,7 +2,7 @@ from ..assets import TableauAsset
 
 QUERY_TEMPLATE = """
 {{
-  {resource}Connection(first: {page_size}, after: AFTER_TOKEN_SIGNAL) {{
+  {resource}Connection(first: {first}, offset: {offset}) {{
     nodes {{ {fields}
     }}
     pageInfo {{

{castor_extractor-0.24.4 → castor_extractor-0.24.7}/pyproject.toml

@@ -1,9 +1,9 @@
-[build-system]
+    [build-system]
 requires = ["setuptools>=61.2"]
 
 [tool.poetry]
 name = "castor-extractor"
-version = "0.24.4"
+version = "0.24.7"
 description = "Extract your metadata assets."
 authors = ["Castor <support@castordoc.com>"]
 license = "EULA"
@@ -32,7 +32,7 @@ google-api-core = ">=2.1.1, <3.0.0"
 google-api-python-client = {version = "^2.121.0", optional = true}
 google-auth = "^2"
 google-cloud-core = ">=2.1.0, <3.0.0"
-google-cloud-storage = "^2"
+google-cloud-storage = "^3.1.0"
 google-resumable-media = "^2.0.3"
 googleapis-common-protos = "^1.53.0"
 looker-sdk = { version = "^25.0.0", optional = true }
@@ -50,7 +50,7 @@ python-dateutil = ">=2.0.0, <=3.0.0"
 requests = ">=2.0.0, <3.0.0"
 snowflake-connector-python = { version = "^3.4.0", optional = true }
 snowflake-sqlalchemy = { version = "!=1.2.5, <2.0.0", optional = true }
-setuptools = ">=75.6"
+setuptools = ">=78.1"
 sqlalchemy = "~1.4"
 sqlalchemy-bigquery = { version = ">=1.0.0, <= 2.0.0", optional = true, extras = [
   "bqstorage",
@@ -66,7 +66,7 @@ pymssql = { version = "^2.2.11", optional = true }
 pymysql = { extras = ["rsa"], version = "^1.1.0", optional = true }
 
 # 3.7 has issue with urrllib dependency
-# https://castorglobal.slack.com/archives/C049PRU1GK1/p1738328811319309
+# https://coalescesoftware.slack.com/archives/C08HEQHS487/p1738328811319309
 databricks-sql-connector = { version = "3.6.0", optional = true }
 
 [tool.poetry.extras]