push code for 0.24 (#1168)

* Allow injection of sessions (#1111)
* show server info (#1118)
* Fix bug in exposing ExcelRequestOptions and test (#1123)
* Fix a few pylint errors (#1124)
* fix behavior when url has no protocol (#1125)
* Add permission control for Data Roles and Metrics (Issue #1063) (#1120)
* add option to pass specific datasources (#1150)
* allow user agent to be set by caller (#1166)
* Fix issues with connections publishing workbooks (#1171)
* Allow download to file-like objects (#1172)
* Add updated_at to JobItem class (#1182)
* fix revision references where xml returned does not match docs (#1176)
* Do not create empty connections list (#1178)
---------

Co-authored-by: Marwan Baghdad <[email protected]>
Co-authored-by: jorwoods <[email protected]>
Co-authored-by: Brian Cantoni <[email protected]>
Co-authored-by: TrimPeachu <[email protected]>
Co-authored-by: Stu Tomlinson <[email protected]>
Co-authored-by: Jeremy Harris <[email protected]>

Author: 7 people

tableauserverclient/models/datasource_item.py

@@ -34,7 +34,7 @@ class AskDataEnablement:
         Disabled = "Disabled"
         SiteDefault = "SiteDefault"
 
-    def __init__(self, project_id: str, name: str = None) -> None:
+    def __init__(self, project_id: str, name: Optional[str] = None) -> None:
         self._ask_data_enablement = None
         self._certified = None
         self._certification_note = None

tableauserverclient/models/job_item.py

@@ -34,6 +34,7 @@ def __init__(
         workbook_id: Optional[str] = None,
         datasource_id: Optional[str] = None,
         flow_run: Optional[FlowRunItem] = None,
+        updated_at: Optional["datetime.datetime"] = None,
     ):
         self._id = id_
         self._type = job_type
@@ -47,6 +48,7 @@ def __init__(
         self._workbook_id = workbook_id
         self._datasource_id = datasource_id
         self._flow_run = flow_run
+        self._updated_at = updated_at
 
     @property
     def id(self) -> str:
@@ -113,9 +115,13 @@ def flow_run(self):
     def flow_run(self, value):
         self._flow_run = value
 
+    @property
+    def updated_at(self) -> Optional["datetime.datetime"]:
+        return self._updated_at
+
     def __repr__(self):
         return (
-            "<Job#{_id} {_type} created_at({_created_at}) started_at({_started_at}) completed_at({_completed_at})"
+            "<Job#{_id} {_type} created_at({_created_at}) started_at({_started_at}) updated_at({_updated_at}) completed_at({_completed_at})"
             " progress ({_progress}) finish_code({_finish_code})>".format(**self.__dict__)
         )
 
@@ -144,6 +150,7 @@ def _parse_element(cls, element, ns):
         datasource = element.find(".//t:datasource[@id]", namespaces=ns)
         datasource_id = datasource.get("id") if datasource is not None else None
         flow_run = None
+        updated_at = parse_datetime(element.get("updatedAt", None))
         for flow_job in element.findall(".//t:runFlowJobType", namespaces=ns):
             flow_run = FlowRunItem()
             flow_run._id = flow_job.get("flowRunId", None)
@@ -163,6 +170,7 @@ def _parse_element(cls, element, ns):
             workbook_id,
             datasource_id,
             flow_run,
+            updated_at,
         )
 
 

tableauserverclient/models/revision_item.py

@@ -67,10 +67,10 @@ def from_response(cls, resp: bytes, ns, resource_item) -> List["RevisionItem"]:
             revision_item._resource_id = resource_item.id
             revision_item._resource_name = resource_item.name
             revision_item._revision_number = revision_xml.get("revisionNumber", None)
-            revision_item._current = string_to_bool(revision_xml.get("isCurrent", ""))
-            revision_item._deleted = string_to_bool(revision_xml.get("isDeleted", ""))
-            revision_item._created_at = parse_datetime(revision_xml.get("createdAt", None))
-            for user in revision_xml.findall(".//t:user", namespaces=ns):
+            revision_item._current = string_to_bool(revision_xml.get("current", ""))
+            revision_item._deleted = string_to_bool(revision_xml.get("deleted", ""))
+            revision_item._created_at = parse_datetime(revision_xml.get("publishedAt", None))
+            for user in revision_xml.findall(".//t:publisher", namespaces=ns):
                 revision_item._user_id = user.get("id", None)
                 revision_item._user_name = user.get("name", None)
 

tableauserverclient/models/site_item.py

@@ -50,9 +50,9 @@ def __init__(
         self,
         name: str,
         content_url: str,
-        admin_mode: str = None,
-        user_quota: int = None,
-        storage_quota: int = None,
+        admin_mode: Optional[str] = None,
+        user_quota: Optional[int] = None,
+        storage_quota: Optional[int] = None,
         disable_subscriptions: bool = False,
         subscribe_others_enabled: bool = True,
         revision_history_enabled: bool = False,

tableauserverclient/models/workbook_item.py

@@ -33,7 +33,7 @@
 
 
 class WorkbookItem(object):
-    def __init__(self, project_id: str, name: str = None, show_tabs: bool = False) -> None:
+    def __init__(self, project_id: str, name: Optional[str] = None, show_tabs: bool = False) -> None:
         self._connections = None
         self._content_url = None
         self._webpage_url = None

tableauserverclient/server/endpoint/datasources_endpoint.py

@@ -31,22 +31,9 @@
 )
 from ...models import ConnectionCredentials, RevisionItem
 from ...models.job_item import JobItem
-from ...models import ConnectionCredentials
 
-io_types = (io.BytesIO, io.BufferedReader)
-
-from pathlib import Path
-from typing import (
-    List,
-    Mapping,
-    Optional,
-    Sequence,
-    Tuple,
-    TYPE_CHECKING,
-    Union,
-)
-
-io_types = (io.BytesIO, io.BufferedReader)
+io_types_r = (io.BytesIO, io.BufferedReader)
+io_types_w = (io.BytesIO, io.BufferedWriter)
 
 # The maximum size of a file that can be published in a single request is 64MB
 FILESIZE_LIMIT = 1024 * 1024 * 64  # 64MB
@@ -61,8 +48,10 @@
     from .schedules_endpoint import AddResponse
 
 FilePath = Union[str, os.PathLike]
-FileObject = Union[io.BufferedReader, io.BytesIO]
-PathOrFile = Union[FilePath, FileObject]
+FileObjectR = Union[io.BufferedReader, io.BytesIO]
+FileObjectW = Union[io.BufferedWriter, io.BytesIO]
+PathOrFileR = Union[FilePath, FileObjectR]
+PathOrFileW = Union[FilePath, FileObjectW]
 
 
 class Datasources(QuerysetEndpoint):
@@ -80,7 +69,7 @@ def baseurl(self) -> str:
 
     # Get all datasources
     @api(version="2.0")
-    def get(self, req_options: RequestOptions = None) -> Tuple[List[DatasourceItem], PaginationItem]:
+    def get(self, req_options: Optional[RequestOptions] = None) -> Tuple[List[DatasourceItem], PaginationItem]:
         logger.info("Querying all datasources on site")
         url = self.baseurl
         server_response = self.get_request(url, req_options)
@@ -135,39 +124,11 @@ def delete(self, datasource_id: str) -> None:
     def download(
         self,
         datasource_id: str,
-        filepath: FilePath = None,
+        filepath: Optional[PathOrFileW] = None,
         include_extract: bool = True,
         no_extract: Optional[bool] = None,
     ) -> str:
-        if not datasource_id:
-            error = "Datasource ID undefined."
-            raise ValueError(error)
-        url = "{0}/{1}/content".format(self.baseurl, datasource_id)
-
-        if no_extract is False or no_extract is True:
-            import warnings
-
-            warnings.warn(
-                "no_extract is deprecated, use include_extract instead.",
-                DeprecationWarning,
-            )
-            include_extract = not no_extract
-
-        if not include_extract:
-            url += "?includeExtract=False"
-
-        with closing(self.get_request(url, parameters={"stream": True})) as server_response:
-            _, params = cgi.parse_header(server_response.headers["Content-Disposition"])
-            filename = to_filename(os.path.basename(params["filename"]))
-
-            download_path = make_download_path(filepath, filename)
-
-            with open(download_path, "wb") as f:
-                for chunk in server_response.iter_content(1024):  # 1KB
-                    f.write(chunk)
-
-        logger.info("Downloaded datasource to {0} (ID: {1})".format(download_path, datasource_id))
-        return os.path.abspath(download_path)
+        return self.download_revision(datasource_id, None, filepath, include_extract, no_extract)
 
     # Update datasource
     @api(version="2.0")
@@ -232,10 +193,10 @@ def delete_extract(self, datasource_item: DatasourceItem) -> None:
     def publish(
         self,
         datasource_item: DatasourceItem,
-        file: PathOrFile,
+        file: PathOrFileR,
         mode: str,
-        connection_credentials: ConnectionCredentials = None,
-        connections: Sequence[ConnectionItem] = None,
+        connection_credentials: Optional[ConnectionCredentials] = None,
+        connections: Optional[Sequence[ConnectionItem]] = None,
         as_job: bool = False,
     ) -> Union[DatasourceItem, JobItem]:
 
@@ -255,8 +216,7 @@ def publish(
                 error = "Only {} files can be published as datasources.".format(", ".join(ALLOWED_FILE_EXTENSIONS))
                 raise ValueError(error)
 
-        elif isinstance(file, io_types):
-
+        elif isinstance(file, io_types_r):
             if not datasource_item.name:
                 error = "Datasource item must have a name when passing a file object"
                 raise ValueError(error)
@@ -302,7 +262,7 @@ def publish(
             if isinstance(file, (Path, str)):
                 with open(file, "rb") as f:
                     file_contents = f.read()
-            elif isinstance(file, io_types):
+            elif isinstance(file, io_types_r):
                 file_contents = file.read()
             else:
                 raise TypeError("file should be a filepath or file object.")
@@ -433,14 +393,17 @@ def download_revision(
         self,
         datasource_id: str,
         revision_number: str,
-        filepath: Optional[PathOrFile] = None,
+        filepath: Optional[PathOrFileW] = None,
         include_extract: bool = True,
         no_extract: Optional[bool] = None,
-    ) -> str:
+    ) -> PathOrFileW:
         if not datasource_id:
             error = "Datasource ID undefined."
             raise ValueError(error)
-        url = "{0}/{1}/revisions/{2}/content".format(self.baseurl, datasource_id, revision_number)
+        if revision_number is None:
+            url = "{0}/{1}/content".format(self.baseurl, datasource_id)
+        else:
+            url = "{0}/{1}/revisions/{2}/content".format(self.baseurl, datasource_id, revision_number)
         if no_extract is False or no_extract is True:
             import warnings
 
@@ -455,18 +418,22 @@ def download_revision(
 
         with closing(self.get_request(url, parameters={"stream": True})) as server_response:
             _, params = cgi.parse_header(server_response.headers["Content-Disposition"])
-            filename = to_filename(os.path.basename(params["filename"]))
-
-            download_path = make_download_path(filepath, filename)
-
-            with open(download_path, "wb") as f:
+            if isinstance(filepath, io_types_w):
                 for chunk in server_response.iter_content(1024):  # 1KB
-                    f.write(chunk)
+                    filepath.write(chunk)
+                return_path = filepath
+            else:
+                filename = to_filename(os.path.basename(params["filename"]))
+                download_path = make_download_path(filepath, filename)
+                with open(download_path, "wb") as f:
+                    for chunk in server_response.iter_content(1024):  # 1KB
+                        f.write(chunk)
+                return_path = os.path.abspath(download_path)
 
         logger.info(
-            "Downloaded datasource revision {0} to {1} (ID: {2})".format(revision_number, download_path, datasource_id)
+            "Downloaded datasource revision {0} to {1} (ID: {2})".format(revision_number, return_path, datasource_id)
         )
-        return os.path.abspath(download_path)
+        return return_path
 
     @api(version="2.3")
     def delete_revision(self, datasource_id: str, revision_number: str) -> None:

tableauserverclient/server/endpoint/endpoint.py

@@ -3,7 +3,7 @@
 from packaging.version import Version
 from functools import wraps
 from xml.etree.ElementTree import ParseError
-from typing import Any, Callable, Dict, Optional, TYPE_CHECKING
+from typing import Any, Callable, Dict, Optional, TYPE_CHECKING, Mapping
 
 from .exceptions import (
     ServerResponseError,
@@ -35,15 +35,35 @@ def __init__(self, parent_srv: "Server"):
         self.parent_srv = parent_srv
 
     @staticmethod
-    def _make_common_headers(auth_token, content_type):
-        _client_version: Optional[str] = get_versions()["version"]
-        headers = {}
+    def set_parameters(http_options, auth_token, content, content_type, parameters) -> Dict[str, Any]:
+        parameters = parameters or {}
+        parameters.update(http_options)
+        if "headers" not in parameters:
+            parameters["headers"] = {}
+
         if auth_token is not None:
-            headers[TABLEAU_AUTH_HEADER] = auth_token
+            parameters["headers"][TABLEAU_AUTH_HEADER] = auth_token
         if content_type is not None:
-            headers[CONTENT_TYPE_HEADER] = content_type
-        headers[USER_AGENT_HEADER] = "Tableau Server Client/{}".format(_client_version)
-        return headers
+            parameters["headers"][CONTENT_TYPE_HEADER] = content_type
+
+        Endpoint.set_user_agent(parameters)
+        if content is not None:
+            parameters["data"] = content
+        return parameters or {}
+
+    @staticmethod
+    def set_user_agent(parameters):
+        if USER_AGENT_HEADER not in parameters["headers"]:
+            if USER_AGENT_HEADER in parameters:
+                parameters["headers"][USER_AGENT_HEADER] = parameters[USER_AGENT_HEADER]
+            else:
+                # only set the TSC user agent if not already populated
+                _client_version: Optional[str] = get_versions()["version"]
+                parameters["headers"][USER_AGENT_HEADER] = "Tableau Server Client/{}".format(_client_version)
+
+        # result: parameters["headers"]["User-Agent"] is set
+        # return explicitly for testing only
+        return parameters
 
     def _make_request(
         self,
@@ -54,18 +74,14 @@ def _make_request(
         content_type: Optional[str] = None,
         parameters: Optional[Dict[str, Any]] = None,
     ) -> "Response":
-        parameters = parameters or {}
-        if "headers" not in parameters:
-            parameters["headers"] = {}
-        parameters.update(self.parent_srv.http_options)
-        parameters["headers"].update(Endpoint._make_common_headers(auth_token, content_type))
-
-        if content is not None:
-            parameters["data"] = content
+        parameters = Endpoint.set_parameters(
+            self.parent_srv.http_options, auth_token, content, content_type, parameters
+        )
 
-        logger.debug("request {}, url: {}".format(method.__name__, url))
+        logger.debug("request {}, url: {}".format(method, url))
         if content:
-            logger.debug("request content: {}".format(helpers.strings.redact_xml(content[:1000])))
+            redacted = helpers.strings.redact_xml(content[:1000])
+            logger.debug("request content: {}".format(redacted))
 
         server_response = method(url, **parameters)
         self._check_status(server_response, url)
@@ -78,7 +94,7 @@ def _make_request(
 
         return server_response
 
-    def _check_status(self, server_response, url: str = None):
+    def _check_status(self, server_response, url: Optional[str] = None):
         if server_response.status_code >= 500:
             raise InternalServerError(server_response, url)
         elif server_response.status_code not in Success_codes:

tableauserverclient/server/endpoint/exceptions.py

@@ -1,4 +1,5 @@
 from defusedxml.ElementTree import fromstring
+from typing import Optional
 
 
 class TableauError(Exception):
@@ -33,7 +34,7 @@ def from_response(cls, resp, ns, url=None):
 
 
 class InternalServerError(TableauError):
-    def __init__(self, server_response, request_url: str = None):
+    def __init__(self, server_response, request_url: Optional[str] = None):
         self.code = server_response.status_code
         self.content = server_response.content
         self.url = request_url or "server"