castor-extractor 0.24.22__py3-none-any.whl → 0.24.27__py3-none-any.whl

CHANGELOG.md

@@ -1,5 +1,25 @@
 # Changelog
 
+## 0.24.27 - 2025-06-20
+
+* Strategy: extract logical tables
+
+## 0.24.26 - 2025-06-16
+
+* Coalesce: increase _MAX_ERRORS client parameter
+
+## 0.24.25 - 2025-06-12
+
+* DBT: Fix API base url
+
+## 0.24.24 - 2025-06-06
+
+* Power BI: handle rate limit issues when extracting pages
+
+## 0.24.23 - 2025-06-05
+
+* Salesforce: print response's error message when authentication fails
+
 ## 0.24.22 - 2025-05-27
 
 * Add retry for `Request.Timeout` on **ApiClient**

castor_extractor/transformation/coalesce/client/client.py

@@ -1,6 +1,9 @@
+import logging
 from http import HTTPStatus
 from typing import Iterator, Optional
 
+from requests import ConnectionError
+
 from ....utils import (
     APIClient,
     BearerAuth,
@@ -16,7 +19,9 @@ from .type import NodeIDToNamesMapping
 from .utils import column_names_per_node, is_test, test_names_per_node
 
 _LIMIT_MAX = 1_000
-_MAX_ERRORS = 50
+_MAX_ERRORS = 200
+
+logger = logging.getLogger(__name__)
 
 
 def _run_result_payload(result: dict, query_result: dict) -> dict:
@@ -80,8 +85,14 @@ class CoalesceClient(APIClient):
         result = self._get(endpoint=endpoint)
         nodes: list[dict] = []
         for node in result["data"]:
-            details = self._node_details(environment_id, node["id"])
-            nodes.append({**node, **details})
+            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
 
     def _fetch_all_nodes(self) -> SerializedAsset:

castor_extractor/transformation/dbt/client.py

@@ -8,12 +8,13 @@ from typing import Literal, Optional
 import requests
 from dateutil.parser import parse
 
+from ...utils.url import add_path
 from .credentials import DbtCredentials
 
 logger = logging.getLogger(__name__)
 
 
-_URL_SUFFIX = "/api/v2/accounts"
+_URL_SUFFIX = "/api/v2/accounts/"
 
 _DATA_KEY = "data"
 _SUCCESSFUL_RUN_STATUS = 10
@@ -52,7 +53,7 @@ class DbtClient:
         self._credentials = credentials
         self._account_url = _account_url(self._credentials.host)
         self._session = requests.Session()
-        self._account_id: int = self._infer_account_id()
+        self._account_id: str = self._infer_account_id()
 
     def _headers(self, content_type: ContentType) -> dict:
         return {
@@ -88,16 +89,16 @@ class DbtClient:
                 return result[_DATA_KEY]
             return result
 
-    def _infer_account_id(self) -> int:
+    def _infer_account_id(self) -> str:
         result = self._call(url=self._account_url)
-        return result[0]["id"]
+        return str(result[0]["id"])
 
     def list_job_identifiers(self) -> set[int]:
         """
         Return the IDs of all non-deleted jobs for this account
         https://docs.getdbt.com/dbt-cloud/api-v2-legacy#tag/Jobs/operation/listJobsForAccount
         """
-        url = f"{self._account_url}/{self._account_id}/jobs/"
+        url = add_path(self._account_url, self._account_id, "jobs", "/")
         jobs = self._call(url)
         return {job["id"] for job in jobs if not _is_deleted(job)}
 
@@ -110,7 +111,7 @@ class DbtClient:
         Extract the last successful run id, optionally filtered on a given datetime range
         https://docs.getdbt.com/dbt-cloud/api-v2#tag/Runs/operation/listRunsForAccount
         """
-        url = f"{self._account_url}/{self._account_id}/runs/"
+        url = add_path(self._account_url, self._account_id, "runs", "/")
 
         params = {
             "job_definition_id": job_id or self._credentials.job_id,
@@ -142,7 +143,14 @@ class DbtClient:
         Fetch dbt manifest or run results
         https://docs.getdbt.com/dbt-cloud/api-v2-legacy#tag/Runs/operation/getArtifactsByRunId
         """
-        url = f"{self._account_url}/{self._account_id}/runs/{run_id}/artifacts/{artifact}"
+        url = add_path(
+            self._account_url,
+            self._account_id,
+            "runs",
+            str(run_id),
+            "artifacts",
+            artifact,
+        )
         logger.info(
             f"Extracting {artifact} from run id {run_id} with url {url}"
         )

castor_extractor/transformation/dbt/client_test.py

@@ -46,7 +46,7 @@ def test_DbtClient_last_run():
     mock_response_default_job = [{"id": 1, "finished_at": _OLD_DATE_STR}]
     mock_response_job_42 = [{"id": 2, "finished_at": _RECENT_DATE_STR}]
 
-    with patch(infer_path, return_value=40), patch(call_path) as mocked_call:
+    with patch(infer_path, return_value="40"), patch(call_path) as mocked_call:
         credentials = DbtCredentials(token="some-token", job_id=default_job_id)
 
         dbt_client = DbtClient(credentials=credentials)
@@ -85,7 +85,7 @@ def test_DbtClient_list_job_identifiers():
         {"id": 395, "state": 1},
     ]
 
-    with patch(infer_path, return_value=40), patch(call_path) as mocked_call:
+    with patch(infer_path, return_value="40"), patch(call_path) as mocked_call:
         mocked_call.return_value = jobs
         credentials = DbtCredentials(token="some-token", job_id="1")
         dbt_client = DbtClient(credentials=credentials)
@@ -100,7 +100,7 @@ def test_DbtClient_fetch_artifacts():
     run_id = 12345
     url = "https://cloud.getdbt.com/api/v2/accounts/40/runs/{}/artifacts/{}"
 
-    with patch(infer_path, return_value=40), patch(call_path) as mocked_call:
+    with patch(infer_path, return_value="40"), patch(call_path) as mocked_call:
         credentials = DbtCredentials(token="some-token", job_id="1")
         dbt_client = DbtClient(credentials=credentials)
 
@@ -123,7 +123,7 @@ def test_DbtClient_fetch_artifacts():
 
 def test___account_url():
     base_url = "https://cloud.getdbt.com"
-    assert _account_url(base_url) == "https://cloud.getdbt.com/api/v2/accounts"
+    assert _account_url(base_url) == "https://cloud.getdbt.com/api/v2/accounts/"
 
     base_url = "https://emea.dbt.com/"
-    assert _account_url(base_url) == "https://emea.dbt.com/api/v2/accounts"
+    assert _account_url(base_url) == "https://emea.dbt.com/api/v2/accounts/"

castor_extractor/utils/__init__.py

@@ -52,6 +52,7 @@ from .time import (
     yesterday,
 )
 from .type import Callback, Getter, JsonType, SerializedAsset
+from .url import add_path as add_path_to_url, url_from
 from .validation import clean_path, validate_baseurl
 from .write import (
     get_output_filename,

castor_extractor/utils/client/api/client.py

@@ -76,11 +76,6 @@ class APIClient:
         self._auth = auth
         self._safe_mode = safe_mode
 
-    @retry(
-        exceptions=_TIMEOUT_RETRY_EXCEPTIONS,
-        max_retries=_TIMEOUT_RETRY_COUNT,
-        base_ms=_TIMEOUT_RETRY_BASE_MS,
-    )
     def _call(
         self,
         method: HttpMethod,
@@ -90,6 +85,7 @@ class APIClient:
         params: Optional[dict] = None,
         data: Optional[dict] = None,
         pagination_params: Optional[dict] = None,
+        retry_on_timeout: bool = True,
     ) -> Response:
         headers = headers or {}
 
@@ -102,7 +98,17 @@ class APIClient:
 
         url = build_url(self._host, endpoint)
 
-        return requests.request(
+        if retry_on_timeout:
+            retry_wrapper = retry(
+                exceptions=_TIMEOUT_RETRY_EXCEPTIONS,
+                max_retries=_TIMEOUT_RETRY_COUNT,
+                base_ms=_TIMEOUT_RETRY_BASE_MS,
+            )
+            request_fn = retry_wrapper(requests.request)
+        else:
+            request_fn = requests.request
+
+        return request_fn(
             method=method,
             url=url,
             auth=self._auth,
@@ -128,6 +134,7 @@ class APIClient:
         params: Optional[dict] = None,
         data: Optional[dict] = None,
         pagination_params: Optional[dict] = None,
+        retry_on_timeout: bool = True,
     ):
         response = self._call(
             method="GET",
@@ -136,6 +143,7 @@ class APIClient:
             data=data,
             pagination_params=pagination_params,
             headers=headers,
+            retry_on_timeout=retry_on_timeout,
         )
         if response.status_code == HTTPStatus.UNAUTHORIZED:
             self._auth.refresh_token()

castor_extractor/utils/salesforce/client.py

@@ -1,9 +1,11 @@
 import logging
 from collections.abc import Iterator
 from functools import partial
+from http import HTTPStatus
 from typing import Optional
 
 import requests
+from requests import HTTPError, Response
 
 from ...utils import (
     APIClient,
@@ -21,6 +23,21 @@ logger = logging.getLogger(__name__)
 SALESFORCE_TIMEOUT_S = 120
 
 
+class SalesforceBadRequestError(HTTPError):
+    """
+    Custom Exception to print the response's text when an error occurs
+    during Salesforce's authentication.
+    """
+
+    def __init__(self, response: Response):
+        text = response.text
+        message = (
+            f"{response.status_code} Client Error: {response.reason} for url: {response.url}"
+            f"\nResponse text: {text}"
+        )
+        super().__init__(message, response=response)
+
+
 class SalesforceAuth(BearerAuth):
     _AUTH_ENDPOINT = "services/oauth2/token"
 
@@ -29,8 +46,16 @@ class SalesforceAuth(BearerAuth):
         self._token_payload = credentials.token_request_payload()
 
     def fetch_token(self) -> Optional[str]:
+        """
+        Fetches the access token from Salesforce using the provided credentials.
+        A custom Exception is raised if the request fails with a 400 status code.
+        """
         url = build_url(self._host, self._AUTH_ENDPOINT)
         response = requests.post(url, "POST", params=self._token_payload)
+
+        if response.status_code == HTTPStatus.BAD_REQUEST:
+            raise SalesforceBadRequestError(response)
+
         handled_response = handle_response(response)
         return handled_response["access_token"]
 

castor_extractor/utils/url.py

@@ -0,0 +1,48 @@
+from urllib.parse import urlsplit, urlunsplit
+
+
+def url_from(
+    scheme: str = "",
+    netloc: str = "",
+    path: str = "",
+    query: str = "",
+    fragment: str = "",
+) -> str:
+    """Constructs an url from part"""
+    return urlunsplit((scheme, netloc, path, query, fragment))
+
+
+def add_path(base_url: str, *paths: str) -> str:
+    """Adds a path from a base_url."""
+
+    if not is_valid(base_url):
+        raise ValueError(f"Invalid base_url: {base_url}")
+    base_url = _format_base_url(base_url)
+    split = urlsplit(base_url)
+
+    return url_from(
+        split.scheme,
+        split.netloc,
+        "/".join([split.path] + [p.strip("/") for p in paths]),
+        split.query,
+        split.fragment,
+    )
+
+
+def _format_base_url(url: str) -> str:
+    """Remove trailing slash in base url, if applicable."""
+    if url.endswith("/"):
+        return url[:-1]
+    return url
+
+
+def is_valid(
+    url: str,
+    valid_schemes: tuple[str, ...] = ("http", "https"),
+) -> bool:
+    """
+    Simple url validation that ensures the scheme and that there is an hostname.
+    Malformatted url can pass this check such as http://http://toto.com
+    """
+    split = urlsplit(url)
+    return split.scheme in valid_schemes and bool(split.netloc)

castor_extractor/utils/url_test.py

@@ -0,0 +1,55 @@
+from pytest import raises
+
+from ..utils.url import (
+    add_path,
+    is_valid,
+    url_from,
+)
+
+
+def test_add_path():
+    base = "https://test.com"
+
+    # simple
+    assert add_path(base, "toto") == f"{base}/toto"
+
+    # multiple parts
+    assert add_path(base, "to", "ta") == f"{base}/to/ta"
+
+    # multiple parts with slash
+    assert add_path(base, "a/b", "/c/d") == f"{base}/a/b/c/d"
+
+    # base with path
+    assert add_path(f"{base}/my/path", "/1/2/", "3") == f"{base}/my/path/1/2/3"
+
+    # base with query string and fragment
+    assert add_path(f"{base}?q=2#frag", "1/2") == f"{base}/1/2?q=2#frag"
+
+    # bad base url
+    with raises(ValueError):
+        add_path("toto", "toto")
+
+    # trailing slash
+    base = "https://test.com/"
+
+    # multiple parts with slash
+    assert add_path(base, "a/b", "/c/d") == "https://test.com/a/b/c/d"
+
+
+def test_url_is_valid():
+    # valid
+    assert is_valid("https://google.com")
+    assert is_valid("http://user:pass@test.com:444/my/path?my=query#fragment")
+    assert is_valid("ftp://hello.com", valid_schemes=("ftp",))
+
+    # invalid
+    assert not is_valid("hello.com")
+    assert not is_valid("ftp://hello.com")
+    assert not is_valid("http://")
+
+
+def test_url_from():
+    assert url_from() == ""
+    assert url_from("http") == "http://"
+    assert url_from("https", "google.com") == "https://google.com"
+    assert url_from(netloc="te.st", query="q=3") == "//te.st?q=3"

castor_extractor/visualization/powerbi/client/client.py

@@ -2,6 +2,7 @@ import logging
 from collections.abc import Iterator
 from datetime import date
 from functools import partial
+from http import HTTPStatus
 from time import sleep
 from typing import Optional, Union
 
@@ -11,6 +12,7 @@ from requests import HTTPError
 from ....utils import (
     APIClient,
     fetch_all_pages,
+    retry_request,
 )
 from ..assets import PowerBiAsset
 from .authentication import PowerBiBearerAuth
@@ -27,6 +29,9 @@ METADATA_BATCH_SIZE = 100
 POWERBI_SCAN_STATUS_DONE = "Succeeded"
 POWERBI_SCAN_SLEEP_S = 1
 
+MAX_RETRY_PAGES = 1
+RETRY_PAGES_TIMEOUT_MS = 35 * 1000  # 35 seconds
+
 logger = logging.getLogger(__name__)
 
 
@@ -71,6 +76,23 @@ class PowerbiClient(APIClient):
         """
         yield from self._get(self.endpoint_factory.dashboards())[Keys.VALUE]
 
+    @retry_request(
+        status_codes=(HTTPStatus.TOO_MANY_REQUESTS,),
+        max_retries=MAX_RETRY_PAGES,
+        base_ms=RETRY_PAGES_TIMEOUT_MS,
+    )
+    def _pages(self, report_id: str) -> Iterator[dict]:
+        """
+        Extracts the pages of a report.
+        This endpoint is very flaky and frequently returns 400 and 404 errors.
+        After around 50 requests, it hits the rate limit and returns 429 Too Many Requests,
+        which is why we retry it after a short delay.
+        Timeouts are also common; we must skip them because the extraction task
+        might take too long otherwise.
+        """
+        pages_endpoint = self.endpoint_factory.pages(report_id)
+        return self._get(pages_endpoint, retry_on_timeout=False)[Keys.VALUE]
+
     def _reports(self) -> Iterator[dict]:
         """
         Returns a list of reports for the organization.
@@ -83,8 +105,7 @@ class PowerbiClient(APIClient):
             report_id = report.get(Keys.ID)
 
             try:
-                pages_endpoint = self.endpoint_factory.pages(report_id)
-                pages = self._get(pages_endpoint)[Keys.VALUE]
+                pages = self._pages(report_id)
                 report["pages"] = pages
             except (requests.HTTPError, requests.exceptions.Timeout) as e:
                 logger.debug(e)

castor_extractor/visualization/powerbi/client/client_test.py

@@ -85,7 +85,7 @@ def test__reports(power_bi_client):
         reports = list(power_bi_client._reports())
         calls = [
             call(ENDPOINT_FACTORY.reports()),
-            call(ENDPOINT_FACTORY.pages("1")),
+            call(ENDPOINT_FACTORY.pages("1"), retry_on_timeout=False),
         ]
         mocked_get.assert_has_calls(calls)
         assert reports == [

castor_extractor/visualization/strategy/assets.py

@@ -1,14 +1,20 @@
-from ...types import ExternalAsset
+from ...types import ExternalAsset, classproperty
 
 
 class StrategyAsset(ExternalAsset):
     """Strategy assets that can be extracted"""
 
     ATTRIBUTE = "attribute"
+    COLUMN = "column"
     CUBE = "cube"
     DASHBOARD = "dashboard"
     DOCUMENT = "document"
     FACT = "fact"
+    LOGICAL_TABLE = "logical_table"
     METRIC = "metric"
     REPORT = "report"
     USER = "user"
+
+    @classproperty
+    def optional(cls) -> set["StrategyAsset"]:
+        return {StrategyAsset.COLUMN}

castor_extractor/visualization/strategy/client/client.py

@@ -1,14 +1,18 @@
 import logging
 from collections.abc import Iterator
-from enum import Enum
 from typing import Any, Callable, Optional
 from urllib.parse import urlparse
 
 from mstrio.connection import Connection  # type: ignore
 from mstrio.helpers import IServerError  # type: ignore
 from mstrio.modeling import (  # type: ignore
+    Attribute,
+    LogicalTable,
+    PhysicalTable,
+    PhysicalTableType,
     list_attributes,
     list_facts,
+    list_logical_tables,
     list_metrics,
 )
 from mstrio.project_objects import (  # type: ignore
@@ -19,93 +23,21 @@ from mstrio.project_objects import (  # type: ignore
     list_reports,
 )
 from mstrio.server import Environment  # type: ignore
-from mstrio.types import ObjectSubTypes, ObjectTypes  # type: ignore
 from mstrio.users_and_groups import User, list_users  # type: ignore
 from mstrio.utils.entity import Entity  # type: ignore
-from mstrio.utils.helper import is_dashboard  # type: ignore
-from pydantic import BaseModel, ConfigDict
 
 from ..assets import StrategyAsset
 from .credentials import StrategyCredentials
+from .properties import (
+    column_properties,
+    format_url,
+    list_dependencies,
+    lookup_table_id,
+    safe_get_property,
+)
 
 logger = logging.getLogger(__name__)
 
-_BATCH_SIZE: int = 100
-
-
-class URLTemplates(Enum):
-    DASHBOARD = "https://{hostname}/MicroStrategyLibrary/app/{project_id}/{id_}"
-    DOCUMENT = "https://{hostname}/MicroStrategy/servlet/mstrWeb?documentID={id_}&projectID={project_id}"
-    REPORT = "https://{hostname}/MicroStrategy/servlet/mstrWeb?reportID={id_}&projectID={project_id}"
-    FOLDER = "https://{hostname}/MicroStrategy/servlet/mstrWeb?folderID={id_}&projectID={project_id}"
-
-
-def _is_dashboard(entity: Entity) -> bool:
-    """
-    Returns True if the entity is a Dashboard. They can only be distinguished
-    from Documents by checking the `view_media` property.
-    """
-    is_type_document = entity.type == ObjectTypes.DOCUMENT_DEFINITION
-    return is_type_document and is_dashboard(entity.view_media)
-
-
-def _is_report(entity: Entity) -> bool:
-    """
-    Returns True if the entity is a Report. Cubes share the same type as Reports,
-    so the subtype must be checked.
-    """
-    is_type_report = entity.type == ObjectTypes.REPORT_DEFINITION
-    is_subtype_cube = entity.subtype == ObjectSubTypes.OLAP_CUBE.value
-    return is_type_report and not is_subtype_cube
-
-
-def _safe_get_attribute(entity: Entity, attribute: str) -> Optional[str]:
-    """
-    Some properties may raise an error. Example: retrieving a Report's `sql` fails if the Report has not been published.
-    This safely returns the attribute value, or None if the retrieval fails.
-    """
-    try:
-        value = getattr(entity, attribute)
-    except IServerError as e:
-        logger.error(f"Could not get {attribute} for entity {entity.id}: {e}")
-        value = None
-    return value
-
-
-class Dependency(BaseModel):
-    id: str
-    name: str
-    subtype: int
-    type: int
-
-    model_config = ConfigDict(extra="ignore")
-
-
-def _list_dependencies(entity: Entity) -> list[dict]:
-    """Lists the entity's dependencies, keeping only relevant fields."""
-    dependencies: list[dict] = []
-
-    offset = 0
-    while True:
-        batch = entity.list_dependencies(offset=offset, limit=_BATCH_SIZE)
-        dependencies.extend(batch)
-        if len(batch) < _BATCH_SIZE:
-            break
-        offset += _BATCH_SIZE
-
-    return [
-        Dependency(**dependency).model_dump() for dependency in dependencies
-    ]
-
-
-def _level_1_folder_id(folders: list[dict]) -> str:
-    """Searches for the first enclosing folder and returns its ID."""
-    for folder in folders:
-        if folder["level"] == 1:
-            return folder["id"]
-
-    raise ValueError("No level 1 folder found")
-
 
 class StrategyClient:
     """Connect to Strategy through mstrio-py and fetch main assets."""
@@ -116,6 +48,7 @@ class StrategyClient:
             base_url=self.base_url,
             username=credentials.username,
             password=credentials.password,
+            verbose=False,
         )
 
         self.hostname = urlparse(self.base_url).hostname
@@ -129,36 +62,6 @@ class StrategyClient:
     def close(self):
         self.connection.close()
 
-    def _url(self, entity: Entity) -> str:
-        """
-        Formats the right URL.
-        * Dashboards : viewed in MicroStrategy
-        * Reports and Documents : viewed in MicroStrategy Web
-        * other (i.e. Cubes): the URL leads to the folder in MicroStrategy Web
-        """
-        if _is_dashboard(entity):
-            id_ = entity.id
-            template = URLTemplates.DASHBOARD
-
-        elif entity.type == ObjectTypes.DOCUMENT_DEFINITION:
-            id_ = entity.id
-            template = URLTemplates.DOCUMENT
-
-        elif _is_report(entity):
-            id_ = entity.id
-            template = URLTemplates.REPORT
-
-        else:
-            # default to folder URL
-            id_ = _level_1_folder_id(entity.ancestors)
-            template = URLTemplates.FOLDER
-
-        return template.value.format(
-            hostname=self.hostname,
-            id_=id_,
-            project_id=entity.project_id,
-        )
-
     def _common_entity_properties(
         self,
         entity: Entity,
@@ -169,7 +72,7 @@ class StrategyClient:
         Returns the entity's properties, including its dependencies
         and optional URL and/or description.
         """
-        dependencies = _list_dependencies(entity)
+        dependencies = list_dependencies(entity)
         owner_id = entity.owner.id if isinstance(entity.owner, User) else None
         properties = {
             "dependencies": dependencies,
@@ -182,23 +85,80 @@ class StrategyClient:
         }
 
         if with_url:
-            properties["url"] = self._url(entity)
+            assert self.hostname
+            properties["url"] = format_url(
+                entity=entity, hostname=self.hostname
+            )
 
         if with_description:
-            properties["description"] = _safe_get_attribute(
-                entity, "description"
+            properties["description"] = safe_get_property(entity, "description")
+
+        return properties
+
+    def _attributes_properties(self, attribute: Attribute) -> dict[str, Any]:
+        """
+        Attributes have a lookup table, which we need to compute the table lineage.
+        """
+        return {
+            **self._common_entity_properties(attribute, with_url=False),
+            "lookup_table_id": lookup_table_id(attribute),
+        }
+
+    def _physical_table_properties(
+        self, table: Optional[PhysicalTable]
+    ) -> Optional[dict[str, Any]]:
+        """
+        Returns the properties of the physical table, including its columns.
+        A physical table can have 1 of these types:
+          * "normal": meaning it matches 1 warehouse table
+          * "sql": it is based on an SQL statement
+        Other types are not supported (and they technically shouldn't be possible.)
+        """
+        if not table:
+            return None
+
+        properties = {
+            "id": table.id,
+            "name": table.name,
+            "type": table.table_type.value,
+            "columns": column_properties(table.columns),
+        }
+
+        if table.table_type == PhysicalTableType.SQL:
+            physical_table = PhysicalTable(
+                connection=self.connection,
+                id=table.id,
             )
+            properties["sql_statement"] = physical_table.sql_statement
+
+        elif table.table_type == PhysicalTableType.NORMAL:
+            properties["table_prefix"] = table.table_prefix
+            properties["table_name"] = table.table_name
 
         return properties
 
+    def _logical_table_properties(self, table: LogicalTable) -> dict[str, Any]:
+        """
+        Returns properties for:
+            * the logical table itself
+            * its physical table (though it may not be accessible)
+            * the columns of the physical table
+        """
+        physical_table = safe_get_property(table, "physical_table")
+        return {
+            "id": table.id,
+            "name": table.name,
+            "physical_table": self._physical_table_properties(physical_table),
+        }
+
     def _report_properties(self, report: Report) -> dict[str, Any]:
         """
         Report properties contain an optional SQL source query. Due to a typing
         bug in the mstrio package, the typing must be ignored.
         """
         properties = self._common_entity_properties(report)  # type: ignore
-        properties["url"] = self._url(report)  # type: ignore
-        properties["sql"] = _safe_get_attribute(report, "sql")  # type: ignore
+        properties["url"] = format_url(entity=report, hostname=self.hostname)  # type: ignore
+        properties["sql"] = safe_get_property(report, "sql")  # type: ignore
         return properties
 
     @staticmethod
@@ -243,7 +203,7 @@ class StrategyClient:
     def _fetch_attributes(self) -> Iterator[dict[str, Any]]:
         return self._fetch_entities(
             list_attributes,
-            with_url=False,
+            custom_property_extractor=self._attributes_properties,
         )
 
     def _fetch_cubes(self) -> Iterator[dict[str, Any]]:
@@ -263,6 +223,15 @@ class StrategyClient:
             with_description=False,
         )
 
+    def _fetch_logical_tables(self) -> Iterator[dict[str, Any]]:
+        """
+        Yields all logical tables, including their physical tables and their columns.
+        """
+        return self._fetch_entities(
+            list_logical_tables,
+            custom_property_extractor=self._logical_table_properties,
+        )
+
     def _fetch_metrics(self) -> Iterator[dict[str, Any]]:
         return self._fetch_entities(
             list_metrics,
@@ -298,6 +267,9 @@ class StrategyClient:
         elif asset == StrategyAsset.FACT:
             yield from self._fetch_facts()
 
+        elif asset == StrategyAsset.LOGICAL_TABLE:
+            yield from self._fetch_logical_tables()
+
         elif asset == StrategyAsset.METRIC:
             yield from self._fetch_metrics()
 

castor_extractor/visualization/strategy/client/properties.py

@@ -0,0 +1,145 @@
+import logging
+from enum import Enum
+from typing import Any, Optional
+
+from mstrio.helpers import IServerError  # type: ignore
+from mstrio.modeling import (  # type: ignore
+    Attribute,
+    TableColumn,
+)
+from mstrio.types import ObjectSubTypes, ObjectTypes  # type: ignore
+from mstrio.utils.entity import Entity  # type: ignore
+from mstrio.utils.helper import is_dashboard  # type: ignore
+from pydantic import BaseModel, ConfigDict
+
+logger = logging.getLogger(__name__)
+
+_BATCH_SIZE: int = 100
+
+
+class URLTemplates(Enum):
+    DASHBOARD = "https://{hostname}/MicroStrategyLibrary/app/{project_id}/{id_}"
+    DOCUMENT = "https://{hostname}/MicroStrategy/servlet/mstrWeb?documentID={id_}&projectID={project_id}"
+    REPORT = "https://{hostname}/MicroStrategy/servlet/mstrWeb?reportID={id_}&projectID={project_id}"
+    FOLDER = "https://{hostname}/MicroStrategy/servlet/mstrWeb?folderID={id_}&projectID={project_id}"
+
+
+class Dependency(BaseModel):
+    id: str
+    name: str
+    subtype: int
+    type: int
+
+    model_config = ConfigDict(extra="ignore")
+
+
+def list_dependencies(entity: Entity) -> list[dict]:
+    """Lists the entity's dependencies, keeping only relevant fields."""
+    dependencies: list[dict] = []
+
+    offset = 0
+    while True:
+        batch = entity.list_dependencies(offset=offset, limit=_BATCH_SIZE)
+        dependencies.extend(batch)
+        if len(batch) < _BATCH_SIZE:
+            break
+        offset += _BATCH_SIZE
+
+    return [
+        Dependency(**dependency).model_dump() for dependency in dependencies
+    ]
+
+
+def _is_dashboard(entity: Entity) -> bool:
+    """
+    Returns True if the entity is a Dashboard. They can only be distinguished
+    from Documents by checking the `view_media` property.
+    """
+    is_type_document = entity.type == ObjectTypes.DOCUMENT_DEFINITION
+    return is_type_document and is_dashboard(entity.view_media)
+
+
+def _is_report(entity: Entity) -> bool:
+    """
+    Returns True if the entity is a Report. Cubes share the same type as Reports,
+    so the subtype must be checked.
+    """
+    is_type_report = entity.type == ObjectTypes.REPORT_DEFINITION
+    is_subtype_cube = entity.subtype == ObjectSubTypes.OLAP_CUBE.value
+    return is_type_report and not is_subtype_cube
+
+
+def format_url(entity: Entity, hostname: str) -> str:
+    """
+    Formats the right URL.
+    * Dashboards : viewed in MicroStrategy
+    * Reports and Documents : viewed in MicroStrategy Web
+    * other (i.e. Cubes): the URL leads to the folder in MicroStrategy Web
+    """
+    if _is_dashboard(entity):
+        id_ = entity.id
+        template = URLTemplates.DASHBOARD
+
+    elif entity.type == ObjectTypes.DOCUMENT_DEFINITION:
+        id_ = entity.id
+        template = URLTemplates.DOCUMENT
+
+    elif _is_report(entity):
+        id_ = entity.id
+        template = URLTemplates.REPORT
+
+    else:
+        # default to folder URL
+        id_ = level_1_folder_id(entity.ancestors)
+        template = URLTemplates.FOLDER
+
+    return template.value.format(
+        hostname=hostname,
+        id_=id_,
+        project_id=entity.project_id,
+    )
+
+
+def safe_get_property(entity: Entity, attribute: str) -> Optional[str]:
+    """
+    Some properties may raise an error. Example: retrieving a Report's `sql` fails if the Report has not been published.
+    This safely returns the attribute value, or None if the retrieval fails.
+    """
+    try:
+        value = getattr(entity, attribute)
+    except IServerError as e:
+        logger.error(f"Could not get {attribute} for entity {entity.id}: {e}")
+        value = None
+    return value
+
+
+def column_properties(columns: list[TableColumn]) -> list[dict[str, Any]]:
+    """Returns the properties of a physical table's columns."""
+    properties: list[dict[str, Any]] = []
+
+    for column in columns:
+        column_properties = {
+            "id": column.id,
+            "name": column.name,
+            "column_name": column.column_name,
+        }
+        properties.append(column_properties)
+
+    return properties
+
+
+def level_1_folder_id(folders: list[dict]) -> str:
+    """Searches for the first enclosing folder and returns its ID."""
+    for folder in folders:
+        if folder["level"] == 1:
+            return folder["id"]
+
+    raise ValueError("No level 1 folder found")
+
+
+def lookup_table_id(attribute: Attribute):
+    """Returns the lookup table's ID, if there is one."""
+    lookup_table = attribute.attribute_lookup_table
+    if not lookup_table:
+        return None
+    return lookup_table.object_id

{castor_extractor-0.24.22.dist-info → castor_extractor-0.24.27.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: castor-extractor
-Version: 0.24.22
+Version: 0.24.27
 Summary: Extract your metadata assets.
 Home-page: https://www.castordoc.com/
 License: EULA
@@ -215,6 +215,26 @@ For any questions or bug report, contact us at [support@coalesce.io](mailto:supp
 
 # Changelog
 
+## 0.24.27 - 2025-06-20
+
+* Strategy: extract logical tables
+
+## 0.24.26 - 2025-06-16
+
+* Coalesce: increase _MAX_ERRORS client parameter
+
+## 0.24.25 - 2025-06-12
+
+* DBT: Fix API base url
+
+## 0.24.24 - 2025-06-06
+
+* Power BI: handle rate limit issues when extracting pages
+
+## 0.24.23 - 2025-06-05
+
+* Salesforce: print response's error message when authentication fails
+
 ## 0.24.22 - 2025-05-27
 
 * Add retry for `Request.Timeout` on **ApiClient**

{castor_extractor-0.24.22.dist-info → castor_extractor-0.24.27.dist-info}/RECORD

@@ -1,4 +1,4 @@
-CHANGELOG.md,sha256=8EyATqcpMtcJ3LhQgQadRhcDiEULwOXYw40137Y5jgs,17936
+CHANGELOG.md,sha256=ThBExaezvHQ577dZ2fnUkr6cggKn1PutV0WZYR3sMBc,18305
 Dockerfile,sha256=xQ05-CFfGShT3oUqaiumaldwA288dj9Yb_pxofQpufg,301
 DockerfileUsage.md,sha256=2hkJQF-5JuuzfPZ7IOxgM6QgIQW7l-9oRMFVwyXC4gE,998
 LICENCE,sha256=sL-IGa4hweyya1HgzMskrRdybbIa2cktzxb5qmUgDg8,8254
@@ -76,7 +76,7 @@ castor_extractor/transformation/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm
 castor_extractor/transformation/coalesce/__init__.py,sha256=CW_qdtEfwgJRsCyBlk5hNlxwEO-VV6mBXZvkRbND_J8,112
 castor_extractor/transformation/coalesce/assets.py,sha256=pzccYPP66c9PAnVroemx7-6MeRHw7Ft1OlTC6jIamAA,363
 castor_extractor/transformation/coalesce/client/__init__.py,sha256=VRmVpH29rOghtDQnCN7dAdA0dI0Lxseu4BC8rnwM9dU,80
-castor_extractor/transformation/coalesce/client/client.py,sha256=o-BlVqdzcQ2RhrmShsNFXVhsmJF4i0pP6DmiWGAE2dY,6172
+castor_extractor/transformation/coalesce/client/client.py,sha256=7EVJDDxnIm5_uMHLFZ2PD6JzfebVglKST9IiURwn4vs,6524
 castor_extractor/transformation/coalesce/client/credentials.py,sha256=jbJxjbdPspf-dzYKfeb7oqL_8TXd1nvkJrjAcdAnLPc,548
 castor_extractor/transformation/coalesce/client/endpoint.py,sha256=0uLh7dpA1vsR9qr_50SEYV_-heQE4BwED9oNMgYsL-w,1272
 castor_extractor/transformation/coalesce/client/type.py,sha256=oiiVP9NL0ijTXyQmaB8aJVYckc7m-m8ZgMyNIAduUKE,43
@@ -84,8 +84,8 @@ castor_extractor/transformation/coalesce/client/utils.py,sha256=jbxh3OCbYm3fKZD1
 castor_extractor/transformation/coalesce/client/utils_test.py,sha256=Q00Y1n0Q_sZ0LFnYn98yDGFumBsifzVJSc7_3PSBMfI,1543
 castor_extractor/transformation/dbt/__init__.py,sha256=LHQROlMqYWCc7tcmhdjXtROFpJqUvCg9jPC8avHgD4I,107
 castor_extractor/transformation/dbt/assets.py,sha256=JY1nKEGySZ84wNoe7dnizwAYw2q0t8NVaIfqhB2rSw0,148
-castor_extractor/transformation/dbt/client.py,sha256=KSlMHaesDOTYjTKs_ZFSYG1Udxjj5jn0m5zaCf5AZZk,5526
-castor_extractor/transformation/dbt/client_test.py,sha256=YK86romOhTRFqzA2Gs262mDUfmvz0cmhiN3jC9nB8P8,4527
+castor_extractor/transformation/dbt/client.py,sha256=BIue1DNAn2b7kHeiXBkGNosq8jZA2DrgjP7Gi5epAPE,5684
+castor_extractor/transformation/dbt/client_test.py,sha256=RLL7y_pLDv2QBM03qBht8yYEooeT_woRADHcb8vgBQ4,4535
 castor_extractor/transformation/dbt/credentials.py,sha256=pGq7GqFQTw9TwN1DXSHC-0yJ2H6B_wMAbHyQTLqJVh0,543
 castor_extractor/types.py,sha256=nHel2hv6NoHmdpOX_heEfO2-DnZPoYA2x0eJdbFvT0s,1276
 castor_extractor/uploader/__init__.py,sha256=A4bq_SrEtKAsl0r_D_duSTvL5WIQjVfsMy7tDx9IKg0,87
@@ -96,7 +96,7 @@ castor_extractor/uploader/settings.py,sha256=3MvOX-UFRqrLZoiT7wYn9jUGro7NX4RCafY
 castor_extractor/uploader/upload.py,sha256=PSQfkO_7LSE0WBo9Tm_hlS2ONepKeB0cBFdJXySnues,4310
 castor_extractor/uploader/upload_test.py,sha256=7fwstdQe7FjuwGilsCdFpEQr1qLoR2WTRUzyy93fISw,402
 castor_extractor/uploader/utils.py,sha256=otAaySj5aeem6f0CTd0Te6ioJ6uP2J1p348j-SdIwDI,802
-castor_extractor/utils/__init__.py,sha256=ybzci46Myi9LABZGBK2qXOiGd00llcpmBpvAp5LEVHc,1597
+castor_extractor/utils/__init__.py,sha256=_hC54hBfPH41TTuWMsqQcyYVF7SojrOevW3OAv8M05E,1652
 castor_extractor/utils/argument_parser.py,sha256=S4EcIh3wNDjs3fOrQnttCcPsAmG8m_Txl7xvEh0Q37s,283
 castor_extractor/utils/argument_parser_test.py,sha256=wnyLFJ74iEiPxxLSbwFtckR7FIHxsFOVU38ljs9gqRA,633
 castor_extractor/utils/batch.py,sha256=SFlLmJgVjV2nVhIrjVIEp8wJ9du4dKKHq8YVYubnwQQ,448
@@ -106,7 +106,7 @@ castor_extractor/utils/client/abstract.py,sha256=CWF7_afNpEZ3jor-22wXbKIvM20ukHk
 castor_extractor/utils/client/api/__init__.py,sha256=vlG7WXznYgLTn3XyMGsyUkgRkup8FbKM14EXJ8mv-b0,264
 castor_extractor/utils/client/api/auth.py,sha256=lq0K3UEl1vwIIa_vKTdlpIQPdE5K1-5DXmCwO4dKzng,1890
 castor_extractor/utils/client/api/auth_test.py,sha256=LlyXytnatg6ZzR4Zkvzk0BH99FYhHX7qn_nyr2MSnDI,1305
-castor_extractor/utils/client/api/client.py,sha256=En1V9eglhB71q5jZwi5XhNZanLb7IUKfd4AT-sSexJk,4866
+castor_extractor/utils/client/api/client.py,sha256=qmj7KoNqt6F-cmpdaMiz_aVxzwMCgbDNcgzXSbCdu1Y,5183
 castor_extractor/utils/client/api/client_test.py,sha256=FM3ZxsLLfMOBn44cXX6FIgnA31-5TTNIyp9D4LBwtXE,1222
 castor_extractor/utils/client/api/pagination.py,sha256=ph5TYqPiyFGgygsIhCATAHPIQ9UJNZyiTcqlyRdGEno,2460
 castor_extractor/utils/client/api/pagination_test.py,sha256=jCOgXFXrH-jrCxe2dfk80ZksJF-EtmpJPU11BGabsqk,1385
@@ -143,7 +143,7 @@ castor_extractor/utils/retry_test.py,sha256=j_6IJStBomEhxmGpIY9IIlESgMxhcDpmIKj2
 castor_extractor/utils/safe.py,sha256=gvIMRIoggdVeYMl222IYqXnHVDninDklFMlAHt-WldA,1948
 castor_extractor/utils/safe_test.py,sha256=IHN1Z761tYMFslYC-2HAfkXmFPh4LYSqNLs4QZwykjk,2160
 castor_extractor/utils/salesforce/__init__.py,sha256=fZ2U6t6AFFAIC-DLXvFHBgpBDjTvX0tFgZ8zJoehPAc,88
-castor_extractor/utils/salesforce/client.py,sha256=Cd_6gCJRafl2kUa5qm_VuAYyscpdosJBZsg5ka0AqWE,2377
+castor_extractor/utils/salesforce/client.py,sha256=wcbJScclvSHjMf6wYNVnHjmpoC22dSshmZW9rDxXKF0,3211
 castor_extractor/utils/salesforce/client_test.py,sha256=T3gUnZ0cRvnL_4dVc4lInRSO9Ti2WeLkLWV1scps4IY,668
 castor_extractor/utils/salesforce/constants.py,sha256=7yPmUeyn4IHQiHLDutXE0L_OBd41E5080vFxqA_s4Dc,58
 castor_extractor/utils/salesforce/credentials.py,sha256=m_11LIaBrYVgH2bLo-QnxaIY5KhEdtfVXz9r2lb_fd0,1123
@@ -155,6 +155,8 @@ castor_extractor/utils/string_test.py,sha256=u3P2tAPhyfCLvD19rH_JcpHhPuWTHUdg0z_
 castor_extractor/utils/time.py,sha256=jmP1QWg4lv21Jp_Oy71lfJ47hjNOSgHiBOFf964RMPU,1732
 castor_extractor/utils/time_test.py,sha256=pH8DSosNlwDYZXZNNjYDcL0WbmZc_c212LEEn88Oqew,647
 castor_extractor/utils/type.py,sha256=Sd8JlEgbGkBUZnRqCUDtREeBkOMTXtlNMyCph90_J0Q,328
+castor_extractor/utils/url.py,sha256=0YaKAz3EC5PgTb5A2TNOlxf1DANK40yw6hs7ArEtJaU,1238
+castor_extractor/utils/url_test.py,sha256=LWzNdOZqjrDeLmvhPBYmP35mzhm7jGAXi021thiro1Y,1425
 castor_extractor/utils/validation.py,sha256=dRvC9SoFVecVZuLQNN3URq37yX2sBSW3-NxIxkcol5o,1894
 castor_extractor/utils/validation_test.py,sha256=A7P6VmI0kYX2aGIeEN12y7LsY7Kpm8pE4bdVFhbBAMw,1184
 castor_extractor/utils/write.py,sha256=Z_RYm47XeHiUPPUMYMuAjQrVZ18CAkL3daQHQG1XPlM,2148
@@ -236,8 +238,8 @@ castor_extractor/visualization/powerbi/__init__.py,sha256=hoZ73ngLhMc9edqxO9PUIE
 castor_extractor/visualization/powerbi/assets.py,sha256=IB_XKwgdN1pZYGZ4RfeHrLjflianTzWf_6tg-4CIwu0,742
 castor_extractor/visualization/powerbi/client/__init__.py,sha256=UPIhMaCCdNxhiLdkItC0IPFE_AMi-SgqI_ahwjB9utI,151
 castor_extractor/visualization/powerbi/client/authentication.py,sha256=cTohunKr1nUDfvxB0sejJSyfE2BdCtwT1WMPecWlbyU,1045
-castor_extractor/visualization/powerbi/client/client.py,sha256=MbqqUF4yadjbAQ_I0iwOmdqR0qC1L3yfbDGgIZqZ0hQ,7348
-castor_extractor/visualization/powerbi/client/client_test.py,sha256=MhQfg6kj__zpARXfv9-VrJXt1fXj6Eri91y8KA9Sn9E,5694
+castor_extractor/visualization/powerbi/client/client.py,sha256=CWCYmj2spYin74qq9T8v2ZJ5TcxBuEy5EjArhCVZjLM,8141
+castor_extractor/visualization/powerbi/client/client_test.py,sha256=Ox_bHpCSckEpT6IiR7drx2c9fmaVl1btUZxnwEmamGQ,5718
 castor_extractor/visualization/powerbi/client/constants.py,sha256=88R_aGachNNUZh6OSH2fkDwZtY4KTStzKm_g7HNCqqo,387
 castor_extractor/visualization/powerbi/client/credentials.py,sha256=OVWdhZSNODzTdLysY-sbpBZ3uUkLokeayQZnbJAqt2I,1386
 castor_extractor/visualization/powerbi/client/credentials_test.py,sha256=TzFqxsWVQ3sXR_n0bJsexK9Uz7ceXCEPVqDGWTJzW60,993
@@ -276,10 +278,11 @@ castor_extractor/visualization/sigma/client/endpoints.py,sha256=DBFphbgoH78_MZUG
 castor_extractor/visualization/sigma/client/pagination.py,sha256=kNEhNq08tTGbypyMjxs0w4uvDtQc_iaWpOZweaa_FsU,690
 castor_extractor/visualization/sigma/extract.py,sha256=XIT1qsj6g6dgBWP8HPfj_medZexu48EaY9tUwi14gzM,2298
 castor_extractor/visualization/strategy/__init__.py,sha256=HOMv4JxqF5ZmViWi-pDE-PSXJRLTdXal_jtpHG_rlR8,123
-castor_extractor/visualization/strategy/assets.py,sha256=tqB3GOtp-r7IOnYO8UxZgrldoSMImJnv5KeIwDFxg68,302
+castor_extractor/visualization/strategy/assets.py,sha256=yFXF_dX01patC0HQ1eU7Jo_4DZ4m6IJEg0uCB71tMoI,480
 castor_extractor/visualization/strategy/client/__init__.py,sha256=XWP0yF5j6JefDJkDfX-RSJn3HF2ceQ0Yx1PLCfB3BBo,80
-castor_extractor/visualization/strategy/client/client.py,sha256=_K7JkatG0DYtbQOJULTNYKHWuBZ11KMR_rQjx8LiR5c,10242
+castor_extractor/visualization/strategy/client/client.py,sha256=6DJO0Fh67FXxmwY5h_X9cu5sEq3GhM19b9hwn_fvhSE,9460
 castor_extractor/visualization/strategy/client/credentials.py,sha256=urFfNxWX1JG6wwFMYImufQzHa5g-sgjdlVGzi63owwg,1113
+castor_extractor/visualization/strategy/client/properties.py,sha256=PaCFnrc8aDtIELjZbPoPHuEyN6n4A9Kuw6Rmqjhm6qo,4486
 castor_extractor/visualization/strategy/extract.py,sha256=2fBuvS2xiOGXRpxXnZsE_C3en6t1-BlM5TbusjHyEkg,1166
 castor_extractor/visualization/tableau/__init__.py,sha256=eFI_1hjdkxyUiAYiy3szwyuwn3yJ5C_KbpBU0ySJDcQ,138
 castor_extractor/visualization/tableau/assets.py,sha256=HbCRd8VCj1WBEeqg9jwnygnT7xOFJ6PQD7Lq7sV-XR0,635
@@ -425,8 +428,8 @@ castor_extractor/warehouse/sqlserver/queries/table.sql,sha256=kbBQP-TdG5px1IVgyx
 castor_extractor/warehouse/sqlserver/queries/user.sql,sha256=gOrZsMVypusR2dc4vwVs4E1a-CliRsr_UjnD2EbXs-A,94
 castor_extractor/warehouse/sqlserver/query.py,sha256=g0hPT-RmeGi2DyenAi3o72cTlQsLToXIFYojqc8E5fQ,533
 castor_extractor/warehouse/synapse/queries/column.sql,sha256=lNcFoIW3Y0PFOqoOzJEXmPvZvfAsY0AP63Mu2LuPzPo,1351
-castor_extractor-0.24.22.dist-info/LICENCE,sha256=sL-IGa4hweyya1HgzMskrRdybbIa2cktzxb5qmUgDg8,8254
-castor_extractor-0.24.22.dist-info/METADATA,sha256=VTqwYHqBEng4wSZ5N88tT1z6Y2OdS9uETv4qIE3CZVw,25389
-castor_extractor-0.24.22.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
-castor_extractor-0.24.22.dist-info/entry_points.txt,sha256=_F-qeZCybjoMkNb9ErEhnyqXuG6afHIFQhakdBHZsr4,1803
-castor_extractor-0.24.22.dist-info/RECORD,,
+castor_extractor-0.24.27.dist-info/LICENCE,sha256=sL-IGa4hweyya1HgzMskrRdybbIa2cktzxb5qmUgDg8,8254
+castor_extractor-0.24.27.dist-info/METADATA,sha256=0-IaR_-6k2rDoWNgoST4QUHVZWsxKGCD8qQytZ6J-Vo,25758
+castor_extractor-0.24.27.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+castor_extractor-0.24.27.dist-info/entry_points.txt,sha256=_F-qeZCybjoMkNb9ErEhnyqXuG6afHIFQhakdBHZsr4,1803
+castor_extractor-0.24.27.dist-info/RECORD,,