castor-extractor 0.24.32__tar.gz → 0.24.33__tar.gz

{castor_extractor-0.24.32 → castor_extractor-0.24.33}/CHANGELOG.md

@@ -1,9 +1,13 @@
 # Changelog
 
+## 0.24.33 - 2025-07-10
+
+* Tableau - Add an option to skip fields ingestion
+
 ## 0.24.32 - 2025-07-02
 
 * Salesforce reporting - extract report's metadata
-*
+
 ## 0.24.31 - 2025-07-02
 
 * Looker Studio: add option to list users via a provided JSON file

{castor_extractor-0.24.32 → castor_extractor-0.24.33}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: castor-extractor
-Version: 0.24.32
+Version: 0.24.33
 Summary: Extract your metadata assets.
 Home-page: https://www.castordoc.com/
 License: EULA
@@ -215,10 +215,14 @@ For any questions or bug report, contact us at [support@coalesce.io](mailto:supp
 
 # Changelog
 
+## 0.24.33 - 2025-07-10
+
+* Tableau - Add an option to skip fields ingestion
+
 ## 0.24.32 - 2025-07-02
 
 * Salesforce reporting - extract report's metadata
-*
+
 ## 0.24.31 - 2025-07-02
 
 * Looker Studio: add option to list users via a provided JSON file

{castor_extractor-0.24.32 → castor_extractor-0.24.33}/castor_extractor/commands/extract_tableau.py

@@ -28,6 +28,13 @@ def main():
         help="Option to avoid extracting Tableau columns, default to False",
     )
 
+    parser.add_argument(
+        "--skip-fields",
+        dest="skip_fields",
+        action="store_true",
+        help="Option to avoid extracting Tableau fields, default to False",
+    )
+
     parser.add_argument(
         "--with-pulse",
         dest="with_pulse",
@@ -41,6 +48,14 @@ def main():
         required=False,
     )
 
+    parser.add_argument(
+        "-ie",
+        "--ignore-errors",
+        action="store_true",
+        dest="ignore_errors",
+        help="Allow partial extraction of Fields and Columns: skip batch in case of Timeout errors",
+    )
+
     parser.add_argument("-o", "--output", help="Directory to write to")
 
     tableau.extract_all(**parse_filled_arguments(parser))

{castor_extractor-0.24.32 → castor_extractor-0.24.33}/castor_extractor/visualization/tableau/client/client.py

@@ -122,13 +122,17 @@ class TableauClient:
         credentials: TableauCredentials,
         timeout_sec: int = DEFAULT_TIMEOUT_SECONDS,
         with_columns: bool = True,
+        with_fields: bool = True,
         with_pulse: bool = False,
         override_page_size: Optional[int] = None,
+        ignore_errors: bool = False,
     ):
         self._credentials = credentials
         self._server = _server(credentials.server_url, timeout_sec)
         self._with_columns = with_columns
+        self._with_fields = with_fields
         self._with_pulse = with_pulse
+        self._ignore_errors = ignore_errors
 
         self._client_metadata = TableauClientMetadataApi(
             server=self._server,
@@ -221,6 +225,10 @@ class TableauClient:
             logger.info(f"Skipping asset {asset} - deactivated columns")
             return []
 
+        if asset == TableauAsset.FIELD and not self._with_fields:
+            logger.info(f"Skipping asset {asset} - deactivated fields")
+            return []
+
         logger.info(f"Extracting {asset.name}...")
 
         if asset == TableauAsset.DATASOURCE:
@@ -240,4 +248,4 @@ class TableauClient:
             return self._client_rest.fetch(asset)
 
         # other assets can be extracted via Metadata API
-        return self._client_metadata.fetch(asset)
+        return self._client_metadata.fetch(asset, self._ignore_errors)

{castor_extractor-0.24.32 → castor_extractor-0.24.33}/castor_extractor/visualization/tableau/client/client_metadata_api.py

@@ -2,6 +2,7 @@ import logging
 from collections.abc import Iterator
 from typing import Optional
 
+import requests
 import tableauserverclient as TSC  # type: ignore
 
 from ....utils import SerializedAsset, retry
@@ -12,6 +13,13 @@ from .gql_queries import FIELDS_QUERIES, GQL_QUERIES, QUERY_TEMPLATE
 
 logger = logging.getLogger(__name__)
 
+# These assets are known to be error-prone, so it's acceptable if a few are missed.
+# If errors occur, skip the current batch.
+_SAFE_MODE_ASSETS = (
+    TableauAsset.COLUMN,
+    TableauAsset.FIELD,
+)
+
 # increase the value when extraction is too slow
 # decrease the value when timeouts arise
 _CUSTOM_PAGE_SIZE: dict[TableauAsset, int] = {
@@ -92,6 +100,7 @@ def gql_query_scroll(
     resource: str,
     fields: str,
     page_size: int,
+    skip_batch: bool,
 ) -> Iterator[SerializedAsset]:
     """
     Iterate over GQL query results, handling pagination and cursor
@@ -119,15 +128,22 @@ def gql_query_scroll(
 
     current_offset = 0
     while True:
-        payload = _call(first=page_size, offset=current_offset)
-        yield payload["nodes"]
+        try:
+            payload = _call(first=page_size, offset=current_offset)
+            yield payload["nodes"]
+
+            current_offset += len(payload["nodes"])
+            total = payload["totalCount"]
+            logger.info(f"Extracted {current_offset}/{total} {resource}")
 
-        current_offset += len(payload["nodes"])
-        total = payload["totalCount"]
-        logger.info(f"Extracted {current_offset}/{total} {resource}")
+            if not payload["pageInfo"]["hasNextPage"]:
+                break
+        except requests.exceptions.ReadTimeout:
+            if not skip_batch:
+                raise
 
-        if not payload["pageInfo"]["hasNextPage"]:
-            break
+            logger.warning("Skipping batch because of TableauServer Timeout")
+            current_offset += page_size
 
 
 def _deduplicate(result_pages: Iterator[SerializedAsset]) -> SerializedAsset:
@@ -177,12 +193,14 @@ class TableauClientMetadataApi:
         resource: str,
         fields: str,
         page_size: int = DEFAULT_PAGE_SIZE,
+        skip_batch: bool = False,
     ) -> SerializedAsset:
         result_pages = gql_query_scroll(
             self._server,
             resource=resource,
             fields=fields,
             page_size=page_size,
+            skip_batch=skip_batch,
         )
         return _deduplicate(result_pages)
 
@@ -193,21 +211,41 @@ class TableauClientMetadataApi:
             or DEFAULT_PAGE_SIZE
         )
 
-    def _fetch_fields(self) -> SerializedAsset:
+    def _fetch_fields(self, skip_batch: bool = False) -> SerializedAsset:
         result: SerializedAsset = []
         page_size = self._page_size(TableauAsset.FIELD)
         for resource, fields in FIELDS_QUERIES:
-            current = self._call(resource, fields, page_size)
+            current = self._call(
+                resource,
+                fields,
+                page_size,
+                skip_batch=skip_batch,
+            )
             result.extend(current)
         return result
 
+    @staticmethod
+    def _should_skip_batch_with_timeout(
+        asset: TableauAsset,
+        ignore_metadata_errors: bool = False,
+    ) -> bool:
+        return asset in _SAFE_MODE_ASSETS and ignore_metadata_errors
+
     def fetch(
         self,
         asset: TableauAsset,
+        ignore_errors: bool = False,
     ) -> SerializedAsset:
+        skip_batch = self._should_skip_batch_with_timeout(asset, ignore_errors)
+
         if asset == TableauAsset.FIELD:
-            return self._fetch_fields()
+            return self._fetch_fields(skip_batch=skip_batch)
 
         page_size = self._page_size(asset)
         resource, fields = GQL_QUERIES[asset]
-        return self._call(resource, fields, page_size)
+        return self._call(
+            resource=resource,
+            fields=fields,
+            page_size=page_size,
+            skip_batch=skip_batch,
+        )

{castor_extractor-0.24.32 → castor_extractor-0.24.33}/castor_extractor/visualization/tableau/extract.py

@@ -33,16 +33,20 @@ def extract_all(**kwargs) -> None:
     """
     output_directory = kwargs.get("output") or from_env(OUTPUT_DIR)
     with_columns = not kwargs.get("skip_columns")
+    with_fields = not kwargs.get("skip_fields")
     with_pulse = kwargs.get("with_pulse") or False
     page_size = kwargs.get("page_size")
+    ignore_errors = kwargs.get("ignore_errors") or False
     timestamp = current_timestamp()
 
     credentials = TableauCredentials(**kwargs)
     client = TableauClient(
         credentials,
         with_columns=with_columns,
+        with_fields=with_fields,
         with_pulse=with_pulse,
         override_page_size=page_size,
+        ignore_errors=ignore_errors,
     )
     client.login()
 

{castor_extractor-0.24.32 → castor_extractor-0.24.33}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools>=61.2"]
 
 [tool.poetry]
 name = "castor-extractor"
-version = "0.24.32"
+version = "0.24.33"
 description = "Extract your metadata assets."
 authors = ["Castor <support@castordoc.com>"]
 license = "EULA"