castor-extractor 0.24.33__tar.gz → 0.24.35__tar.gz

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 0.24.35 - 2025-07-29
+
+* Coalesce - Fix pagination issue
+
+## 0.24.34 - 2025-07-02
+
+* SQLServer: multiple databases
+
 ## 0.24.33 - 2025-07-10
 
 * Tableau - Add an option to skip fields ingestion
@@ -10,7 +18,7 @@
 
 ## 0.24.31 - 2025-07-02
 
-* Looker Studio: add option to list users via a provided JSON file
+* Looker Studio: add an option to list users via a provided JSON file
 
 ## 0.24.30 - 2025-06-26
 

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: castor-extractor
-Version: 0.24.33
+Version: 0.24.35
 Summary: Extract your metadata assets.
 Home-page: https://www.castordoc.com/
 License: EULA
@@ -215,6 +215,14 @@ For any questions or bug report, contact us at [support@coalesce.io](mailto:supp
 
 # Changelog
 
+## 0.24.35 - 2025-07-29
+
+* Coalesce - Fix pagination issue
+
+## 0.24.34 - 2025-07-02
+
+* SQLServer: multiple databases
+
 ## 0.24.33 - 2025-07-10
 
 * Tableau - Add an option to skip fields ingestion
@@ -225,7 +233,7 @@ For any questions or bug report, contact us at [support@coalesce.io](mailto:supp
 
 ## 0.24.31 - 2025-07-02
 
-* Looker Studio: add option to list users via a provided JSON file
+* Looker Studio: add an option to list users via a provided JSON file
 
 ## 0.24.30 - 2025-06-26
 

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/commands/extract_sqlserver.py

@@ -23,6 +23,16 @@ def main():
         action="store_true",
         help="Skips files already extracted instead of replacing them",
     )
+    parser.add_argument(
+        "--db-allowed",
+        nargs="*",
+        help="List of databases that should be extracted",
+    )
+    parser.add_argument(
+        "--db-blocked",
+        nargs="*",
+        help="List of databases that should not be extracted",
+    )
     parser.set_defaults(skip_existing=False)
 
     args = parser.parse_args()
@@ -35,4 +45,6 @@ def main():
         password=args.password,
         output_directory=args.output,
         skip_existing=args.skip_existing,
+        db_allowed=args.db_allowed,
+        db_blocked=args.db_blocked,
     )

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/transformation/coalesce/client/client.py

@@ -1,31 +1,47 @@
 import logging
+from functools import partial
 from http import HTTPStatus
-from typing import Iterator, Optional
+from typing import Callable, Optional
 
-from requests import ConnectionError
+from pydantic import ValidationError
 
 from ....utils import (
     APIClient,
     BearerAuth,
     RequestSafeMode,
     SerializedAsset,
+    fetch_all_pages,
 )
 from ..assets import CoalesceAsset, CoalesceQualityAsset
 from .credentials import CoalesceCredentials
 from .endpoint import (
     CoalesceEndpointFactory,
 )
-from .type import NodeIDToNamesMapping
-from .utils import column_names_per_node, is_test, test_names_per_node
+from .pagination import CoalescePagination
+
+logger = logging.getLogger(__name__)
+
+
+COALESCE_PAGE_SIZE = 300
+COALESCE_PAGE_SIZE_RUN_RESULTS = 1_000
+
+COALESCE_TIMEOUT_SECONDS = 90
 
-_LIMIT_MAX = 1_000
 _MAX_ERRORS = 200
 
-logger = logging.getLogger(__name__)
+COALESCE_SAFE_MODE = RequestSafeMode(
+    status_codes=(HTTPStatus.INTERNAL_SERVER_ERROR,),
+    max_errors=_MAX_ERRORS,
+)
 
 
-def _run_result_payload(result: dict, query_result: dict) -> dict:
+def _run_result_payload(
+    environment_id: str,
+    result: dict,
+    query_result: dict,
+) -> dict:
     return {
+        "environment_id": environment_id,
         "node_id": result["nodeID"],
         "node_name": result["name"],
         "test_name": query_result["name"],
@@ -37,13 +53,6 @@ def _run_result_payload(result: dict, query_result: dict) -> dict:
     }
 
 
-COALESCE_SAFE_MODE = RequestSafeMode(
-    status_codes=(HTTPStatus.INTERNAL_SERVER_ERROR,),
-    max_errors=_MAX_ERRORS,
-)
-COALESCE_TIMEOUT_SECONDS = 90
-
-
 class CoalesceBearerAuth(BearerAuth):
     """Bearer Authentication for Coalesce"""
 
@@ -69,65 +78,74 @@ class CoalesceClient(APIClient):
             timeout=COALESCE_TIMEOUT_SECONDS,
         )
 
-    def _fetch_environments(self) -> Iterator[dict]:
-        endpoint = CoalesceEndpointFactory.environments()
-        result = self._get(endpoint=endpoint)
-        return result["data"]
-
-    def _node_details(self, environment_id: int, node_id: str) -> dict:
-        endpoint = CoalesceEndpointFactory.nodes(
-            environment_id=environment_id, node_id=node_id
+    def _get_paginated(
+        self,
+        endpoint: str,
+        limit: int = COALESCE_PAGE_SIZE,
+        params: Optional[dict] = None,
+    ) -> Callable:
+        return partial(
+            self._get,
+            retry_on_timeout=False,  # explained in the docstring
+            endpoint=endpoint,
+            params={
+                "limit": limit,
+                **(params or dict()),
+            },
         )
-        return self._get(endpoint=endpoint)
+
+    def _fetch_environments(self) -> SerializedAsset:
+        endpoint = CoalesceEndpointFactory.environments()
+        request = self._get_paginated(endpoint=endpoint)
+        result = fetch_all_pages(request, CoalescePagination)
+        return list(result)
 
     def _fetch_env_nodes(self, environment_id: int) -> SerializedAsset:
         endpoint = CoalesceEndpointFactory.nodes(environment_id=environment_id)
-        result = self._get(endpoint=endpoint)
-        nodes: list[dict] = []
-        for node in result["data"]:
-            try:
-                details = self._node_details(environment_id, node["id"])
-                nodes.append({**node, **details})
-            except ConnectionError as e:
-                node_id = node["id"]
-                message = f"ConnectionError, environment: {environment_id}, node: {node_id}"
-                logger.warning(message)
-                raise e
-        return nodes
+        request = self._get_paginated(
+            endpoint=endpoint,
+            params={"detail": "true"},
+        )
+        result = fetch_all_pages(request, CoalescePagination)
+        return [
+            {
+                **node,
+                "environment_id": environment_id,
+            }
+            for node in result
+        ]
 
     def _fetch_all_nodes(self) -> SerializedAsset:
+        environments = self._fetch_environments()
+        total = len(environments)
         nodes: list[dict] = []
-        for environment in self._fetch_environments():
-            environment_id = environment["id"]
-            nodes.extend(self._fetch_env_nodes(environment_id))
+
+        for index, env in enumerate(environments):
+            env_id = env["id"]
+            logger.info(f"Fetching nodes for env #{env_id} - {index}/{total}")
+            try:
+                nodes.extend(self._fetch_env_nodes(env_id))
+            except ValidationError as e:
+                # 500 Server Error: Internal Server Error on Coalesce API
+                logger.warning(
+                    f"Skipping nodes for {env_id} due to the following Error: {e}"
+                )
+            logger.info(f"{len(nodes)} nodes extracted so far")
         return nodes
 
     def _fetch_runs(self, starting_from: str) -> SerializedAsset:
-        """
-        fetch runs, per environment;
-        we break per environment to lower the chance of exceeding the 1k limit
-        """
-        runs: list[dict] = []
-        for environment in self._fetch_environments():
-            environment_id = environment["id"]
-            runs.extend(
-                self._fetch_recent_runs_per_env(environment_id, starting_from)
-            )
-        return runs
-
-    def _fetch_recent_runs_per_env(
-        self, environment_id: int, starting_from: str
-    ) -> SerializedAsset:
         endpoint = CoalesceEndpointFactory.runs()
         params = {
-            "environmentID": environment_id,
-            "limit": _LIMIT_MAX,
             "orderBy": "runEndTime",
             "orderByDirection": "asc",
             "startingFrom": starting_from,
         }
-        result = self._get(endpoint=endpoint, params=params)
-        return result["data"]
+        request = self._get_paginated(
+            endpoint=endpoint,
+            params=params,
+            limit=COALESCE_PAGE_SIZE_RUN_RESULTS,
+        )
+        return list(fetch_all_pages(request, CoalescePagination))
 
     def _fetch_run_results(self, run_id: str) -> SerializedAsset:
         endpoint = CoalesceEndpointFactory.run_results(run_id)
@@ -136,51 +154,37 @@ class CoalesceClient(APIClient):
 
     def _run_results_by_run(
         self,
+        environment_id: str,
         run_id: str,
-        test_names: NodeIDToNamesMapping,
-        column_names: NodeIDToNamesMapping,
     ) -> SerializedAsset:
         run_results: list[dict] = []
         for result in self._fetch_run_results(run_id):
-            node_id = result["nodeID"]
             for query_result in result["queryResults"]:
-                _is_test = is_test(
+                if query_result["type"] != "sqlTest":
+                    continue
+                run_result = _run_result_payload(
+                    environment_id,
+                    result,
                     query_result,
-                    node_id,
-                    test_names,
-                    column_names,
                 )
-                if not _is_test:
-                    continue
-                run_result = _run_result_payload(result, query_result)
                 run_results.append(run_result)
         return run_results
 
-    def _run_results_by_env(
-        self, environment_id: int, starting_from: str
+    def _fetch_all_run_results(
+        self,
+        starting_from: str,
     ) -> SerializedAsset:
         run_results: list[dict] = []
-        nodes = self._fetch_env_nodes(environment_id)
-        test_names = test_names_per_node(nodes)
-        column_names = column_names_per_node(nodes)
-        runs = self._fetch_recent_runs_per_env(environment_id, starting_from)
 
-        for run in runs:
-            run_id = run["id"]
-            _results = self._run_results_by_run(
-                run_id, test_names, column_names
-            )
-            run_results.extend(_results)
-        return run_results
-
-    def _fetch_all_run_results(self, starting_from: str) -> SerializedAsset:
-        run_results: list[dict] = []
-
-        for environment in self._fetch_environments():
-            environment_id = environment["id"]
-            _results = self._run_results_by_env(environment_id, starting_from)
-            run_results.extend(_results)
+        runs = self._fetch_runs(starting_from)
+        total = len(runs)
 
+        for index, run in enumerate(runs):
+            logger.info(f"Extracting run results ({index}/{total})")
+            run_id = run["id"]
+            environment_id = run["environmentID"]
+            current_results = self._run_results_by_run(environment_id, run_id)
+            run_results.extend(current_results)
         return run_results
 
     def fetch(

castor_extractor-0.24.35/castor_extractor/transformation/coalesce/client/pagination.py

@@ -0,0 +1,26 @@
+from typing import Optional, Union
+
+from ....utils import PaginationModel
+
+
+class CoalescePagination(PaginationModel):
+    """
+    Class to handle paginated results for Coalesce
+    See their documentation here
+    https://docs.coalesce.io/docs/api
+    """
+
+    data: list
+    next: Union[Optional[str], Optional[int]] = None
+
+    def is_last(self) -> bool:
+        """Stopping condition for the pagination"""
+        return self.next is None
+
+    def next_page_payload(self):
+        """Payload enabling to generate the request for the next page"""
+        return {"startingFrom": self.next}
+
+    def page_results(self) -> list:
+        """List of results of the current page"""
+        return self.data

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/utils/__init__.py

@@ -19,7 +19,13 @@ from .client import (
     handle_response,
     uri_encode,
 )
-from .collection import deduplicate, empty_iterator, group_by, mapping_from_rows
+from .collection import (
+    deduplicate,
+    empty_iterator,
+    filter_items,
+    group_by,
+    mapping_from_rows,
+)
 from .constants import OUTPUT_DIR
 from .deprecate import deprecate_python
 from .env import from_env

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/utils/client/api/pagination.py

@@ -66,16 +66,19 @@ def fetch_all_pages(
     """
     page_number = 1
     response_payload = request()
+
     paginated_response = pagination_model(**response_payload)
+
     while not paginated_response.is_last():
         logger.debug(f"Fetching page number {page_number}")
         yield from paginated_response.page_results()
         next_page_parameters = paginated_response.next_page_parameters()
-        new_request = partial(request, **next_page_parameters)
+        request_with_pagination = partial(request, **next_page_parameters)
         if rate_limit:
             sleep(rate_limit)
         paginated_response = pagination_model(
-            current_page_payload=next_page_parameters, **new_request()
+            current_page_payload=next_page_parameters,
+            **request_with_pagination(),
         )
         page_number += 1
 

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/utils/collection.py

@@ -2,6 +2,8 @@ from collections import defaultdict
 from collections.abc import Iterable, Sequence
 from typing import (
     Any,
+    List,
+    Optional,
     TypeVar,
 )
 
@@ -80,3 +82,27 @@ def deduplicate(
         deduplicated.append(element)
 
     return deduplicated
+
+
+def filter_items(
+    items: Iterable[T],
+    allowed: Optional[Iterable[T]] = None,
+    blocked: Optional[Iterable[T]] = None,
+) -> List[T]:
+    """
+    Filters `items` by excluding any in `blocked` or including only those in `allowed`.
+    If both `allowed` and `blocked` are None, returns all items.
+    If both are provided, raise an error.
+    """
+    items = list(items)
+
+    if allowed and blocked:
+        raise AttributeError(
+            "Only one of `allowed` and `blocked` can be provided"
+        )
+    if blocked:
+        return [item for item in items if item not in blocked]
+    if allowed:
+        return [item for item in items if item in allowed]
+
+    return items

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/utils/collection_test.py

@@ -1,4 +1,6 @@
-from .collection import deduplicate, mapping_from_rows
+import pytest
+
+from .collection import deduplicate, filter_items, mapping_from_rows
 
 
 def test__mapping_from_rows__basic_mapping():
@@ -72,3 +74,31 @@ def test_deduplicate():
         {"id": "2", "name": "duplicate"},
     ]
     assert deduplicate("id", elements) == [e1, e2, e3]
+
+
+def test_sqlserver_databases():
+    databases = [
+        "prod",
+        "staging",
+        "test",
+    ]
+
+    # 1. No allowed or blocked: should return all
+    result1 = filter_items(databases)
+    assert result1 == [
+        "prod",
+        "staging",
+        "test",
+    ]
+
+    # 2. Block "prod": only staging and test should remain
+    result2 = filter_items(databases, blocked=["prod"])
+    assert result2 == ["staging", "test"]
+
+    # 3. Only allow "staging" and "test"
+    result3 = filter_items(databases, allowed=["staging", "test"])
+    assert result3 == ["staging", "test"]
+
+    # 4. allowed and blocked, should raise
+    with pytest.raises(AttributeError):
+        filter_items(databases, blocked=["prod"], allowed=["staging", "test"])

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/visualization/looker_studio/client/queries/query.sql

@@ -29,6 +29,7 @@ WITH ranked_by_datasource AS (
         `{project}.region-{region}.INFORMATION_SCHEMA.JOBS_BY_PROJECT`
     WHERE
         job_type = 'QUERY'
+        AND ARRAY_LENGTH(referenced_tables) > 0
         AND EXISTS (
             SELECT
                 1

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/warehouse/sqlserver/client.py

@@ -1,15 +1,20 @@
+import logging
 from collections.abc import Iterator
 
 from sqlalchemy import text
 
 from ...utils import ExtractionQuery, SqlalchemyClient, uri_encode
 
+logger = logging.getLogger(__name__)
+
 SERVER_URI = "{user}:{password}@{host}:{port}/{database}"
 MSSQL_URI = f"mssql+pymssql://{SERVER_URI}"
 DEFAULT_PORT = 1433
 
 _KEYS = ("user", "password", "host", "port", "database")
 
+_SYSTEM_DATABASES = ("master", "model", "msdb", "tempdb", "DBAdmin")
+
 
 def _check_key(credentials: dict) -> None:
     for key in _KEYS:
@@ -51,3 +56,13 @@ class MSSQLClient(SqlalchemyClient):
             yield from results
         finally:
             self.close()
+
+    def get_databases(self) -> list[str]:
+        result = self.execute(
+            ExtractionQuery("SELECT name FROM sys.databases", {})
+        )
+        return [
+            row["name"]
+            for row in result
+            if row["name"] not in _SYSTEM_DATABASES
+        ]

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/warehouse/sqlserver/extract.py

@@ -1,6 +1,6 @@
 import logging
 
-from ...utils import LocalStorage, from_env, write_summary
+from ...utils import LocalStorage, filter_items, from_env, write_summary
 from ..abstract import (
     CATALOG_ASSETS,
     EXTERNAL_LINEAGE_ASSETS,
@@ -51,7 +51,13 @@ def extract_all(**kwargs) -> None:
 
     client = MSSQLClient(credentials=_credentials(kwargs))
 
-    query_builder = MSSQLQueryBuilder()
+    databases = filter_items(
+        client.get_databases(), kwargs.get("allowed"), kwargs.get("blocked")
+    )
+
+    query_builder = MSSQLQueryBuilder(
+        databases=databases,
+    )
 
     storage = LocalStorage(directory=output_directory)
 

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/warehouse/sqlserver/queries/column.sql

@@ -11,7 +11,7 @@ WITH extended_tables AS (
         table_owner_id = principal_id,
         schema_id
     FROM
-        sys.tables
+        {database}.sys.tables
 
     UNION
 
@@ -21,7 +21,7 @@ WITH extended_tables AS (
         table_owner_id = principal_id,
         schema_id
     FROM
-        sys.views
+        {database}.sys.views
 
     UNION
 
@@ -31,7 +31,7 @@ WITH extended_tables AS (
         table_owner_id = principal_id,
         schema_id
     FROM
-        sys.external_tables
+        {database}.sys.external_tables
 ),
 /*
 `sys.columns` contains, among others:
@@ -54,11 +54,11 @@ column_ids AS (
         schema_name = ss.name,
         schema_id = ss.schema_id,
         comment = CONVERT(varchar(1024), ep.value)
-    FROM sys.columns AS sc
+    FROM {database}.sys.columns AS sc
     LEFT JOIN extended_tables AS et ON sc.object_id = et.table_id
-    LEFT JOIN sys.schemas AS ss ON et.schema_id = ss.schema_id
-    LEFT JOIN sys.databases AS sd ON sd.name = DB_NAME()
-    LEFT JOIN sys.extended_properties AS ep
+    LEFT JOIN {database}.sys.schemas AS ss ON et.schema_id = ss.schema_id
+    LEFT JOIN {database}.sys.databases AS sd ON sd.name = '{database}'
+    LEFT JOIN {database}.sys.extended_properties AS ep
         ON
             sc.object_id = ep.major_id
             AND sc.column_id = ep.minor_id
@@ -70,9 +70,9 @@ columns AS (
         i.database_name,
         i.database_id,
         schema_name = c.table_schema,
-        i.schema_id,
+        schema_id = CAST(i.database_id AS VARCHAR(10)) + '_' + CAST(i.schema_id AS VARCHAR(10)),
         table_name = c.table_name,
-        i.table_id,
+        table_id = CAST(i.database_id AS VARCHAR(10)) + '_' + CAST(i.schema_id AS VARCHAR(10)) + '_' + CAST(i.table_id AS VARCHAR(10)),
         c.column_name,
         c.data_type,
         c.ordinal_position,
@@ -87,7 +87,7 @@ columns AS (
         i.comment,
         column_id = CONCAT(i.table_id, '.', c.column_name)
     FROM
-        information_schema.columns AS c
+        {database}.information_schema.columns AS c
     LEFT JOIN column_ids AS i
         ON
             (

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/warehouse/sqlserver/queries/database.sql

@@ -2,4 +2,4 @@ SELECT
     db.database_id,
     database_name = db.name
 FROM sys.databases AS db
-WHERE db.name = DB_NAME()
+WHERE db.name NOT IN ('master', 'model', 'msdb', 'tempdb', 'DBAdmin');

{castor_extractor-0.24.33 → castor_extractor-0.24.35}/castor_extractor/warehouse/sqlserver/queries/schema.sql

@@ -3,21 +3,20 @@ WITH ids AS (
     SELECT DISTINCT
         table_catalog,
         table_schema
-    FROM information_schema.tables
-    WHERE table_catalog = DB_NAME()
+    FROM {database}.information_schema.tables
 )
 
 SELECT
     d.database_id,
     database_name = i.table_catalog,
     schema_name = s.name,
-    s.schema_id,
+    schema_id = CAST(d.database_id AS VARCHAR(10)) + '_' + CAST(s.schema_id AS VARCHAR(10)),
     schema_owner = u.name,
     schema_owner_id = u.uid
-FROM sys.schemas AS s
+FROM {database}.sys.schemas AS s
 INNER JOIN ids AS i
     ON s.name = i.table_schema
-LEFT JOIN sys.sysusers AS u
+LEFT JOIN {database}.sys.sysusers AS u
     ON s.principal_id = u.uid
-LEFT JOIN sys.databases AS d
+LEFT JOIN {database}.sys.databases AS d
     ON i.table_catalog = d.name