acryl-datahub 0.15.0.1rc17__py3-none-any.whl → 0.15.0.2__py3-none-any.whl

{acryl_datahub-0.15.0.1rc17.dist-info → acryl_datahub-0.15.0.2.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.7.0)
+Generator: setuptools (75.8.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

datahub/__init__.py

@@ -3,7 +3,7 @@ import warnings
 
 # Published at https://pypi.org/project/acryl-datahub/.
 __package_name__ = "acryl-datahub"
-__version__ = "0.15.0.1rc17"
+__version__ = "0.15.0.2"
 
 
 def is_dev_mode() -> bool:

datahub/api/entities/assertion/assertion_operator.py

@@ -20,15 +20,13 @@ class Operator(Protocol):
 
     operator: str
 
-    def id(self) -> str:
-        ...
+    def id(self) -> str: ...
 
-    def generate_parameters(self) -> AssertionStdParametersClass:
-        ...
+    def generate_parameters(self) -> AssertionStdParametersClass: ...
 
 
 def _generate_assertion_std_parameter(
-    value: Union[str, int, float, list]
+    value: Union[str, int, float, list],
 ) -> AssertionStdParameterClass:
     if isinstance(value, str):
         return AssertionStdParameterClass(

datahub/api/entities/corpgroup/corpgroup.py

@@ -114,7 +114,7 @@ class CorpGroup(BaseModel):
                 )
                 urns_created.add(m.urn)
             else:
-                logger.warn(
+                logger.warning(
                     f"Suppressing emission of member {m.urn} before we already emitted metadata for it"
                 )
 

datahub/api/entities/datacontract/assertion_operator.py

@@ -19,15 +19,13 @@ class Operator(Protocol):
 
     operator: str
 
-    def id(self) -> str:
-        ...
+    def id(self) -> str: ...
 
-    def generate_parameters(self) -> AssertionStdParametersClass:
-        ...
+    def generate_parameters(self) -> AssertionStdParametersClass: ...
 
 
 def _generate_assertion_std_parameter(
-    value: Union[str, int, float]
+    value: Union[str, int, float],
 ) -> AssertionStdParameterClass:
     if isinstance(value, str):
         return AssertionStdParameterClass(

datahub/api/entities/dataproduct/dataproduct.py

@@ -321,9 +321,9 @@ class DataProduct(ConfigModel):
 
     @classmethod
     def from_datahub(cls, graph: DataHubGraph, id: str) -> DataProduct:
-        data_product_properties: Optional[
-            DataProductPropertiesClass
-        ] = graph.get_aspect(id, DataProductPropertiesClass)
+        data_product_properties: Optional[DataProductPropertiesClass] = (
+            graph.get_aspect(id, DataProductPropertiesClass)
+        )
         domains: Optional[DomainsClass] = graph.get_aspect(id, DomainsClass)
         assert domains, "Data Product must have an associated domain. Found none."
         owners: Optional[OwnershipClass] = graph.get_aspect(id, OwnershipClass)
@@ -438,7 +438,7 @@ class DataProduct(ConfigModel):
             for replace_index, replace_value in patches_replace.items():
                 list_to_manipulate[replace_index] = replace_value
 
-            for drop_index, drop_value in patches_drop.items():
+            for drop_value in patches_drop.values():
                 list_to_manipulate.remove(drop_value)
 
             for add_value in patches_add:

datahub/api/entities/dataset/dataset.py

@@ -266,7 +266,8 @@ class Dataset(BaseModel):
             if self.schema_metadata.fields:
                 for field in self.schema_metadata.fields:
                     field_urn = field.urn or make_schema_field_urn(
-                        self.urn, field.id  # type: ignore[arg-type]
+                        self.urn,  # type: ignore[arg-type]
+                        field.id,  # type: ignore[arg-type]
                     )
                     assert field_urn.startswith("urn:li:schemaField:")
 

datahub/api/entities/structuredproperties/structuredproperties.py

@@ -1,7 +1,7 @@
 import logging
 from enum import Enum
 from pathlib import Path
-from typing import List, Optional
+from typing import Iterable, List, Optional
 
 import yaml
 from pydantic import validator
@@ -118,9 +118,9 @@ class StructuredProperties(ConfigModel):
         id = StructuredPropertyUrn.from_string(self.urn).id
         if self.qualified_name is not None:
             # ensure that qualified name and ID match
-            assert (
-                self.qualified_name == id
-            ), "ID in the urn and the qualified_name must match"
+            assert self.qualified_name == id, (
+                "ID in the urn and the qualified_name must match"
+            )
         return id
 
     @validator("urn", pre=True, always=True)
@@ -184,9 +184,9 @@ class StructuredProperties(ConfigModel):
 
     @classmethod
     def from_datahub(cls, graph: DataHubGraph, urn: str) -> "StructuredProperties":
-        structured_property: Optional[
-            StructuredPropertyDefinitionClass
-        ] = graph.get_aspect(urn, StructuredPropertyDefinitionClass)
+        structured_property: Optional[StructuredPropertyDefinitionClass] = (
+            graph.get_aspect(urn, StructuredPropertyDefinitionClass)
+        )
         if structured_property is None:
             raise Exception(
                 "StructuredPropertyDefinition aspect is None. Unable to create structured property."
@@ -226,3 +226,14 @@ class StructuredProperties(ConfigModel):
             yaml.indent(mapping=2, sequence=4, offset=2)
             yaml.default_flow_style = False
             yaml.dump(self.dict(), fp)
+
+    @staticmethod
+    def list_urns(graph: DataHubGraph) -> Iterable[str]:
+        return graph.get_urns_by_filter(
+            entity_types=["structuredProperty"],
+        )
+
+    @staticmethod
+    def list(graph: DataHubGraph) -> Iterable["StructuredProperties"]:
+        for urn in StructuredProperties.list_urns(graph):
+            yield StructuredProperties.from_datahub(graph, urn)

datahub/cli/cli_utils.py

@@ -3,7 +3,7 @@ import logging
 import time
 import typing
 from datetime import datetime
-from typing import Any, Dict, List, Optional, Tuple, Type, Union
+from typing import Any, Dict, List, Optional, Tuple, Type, TypeVar, Union
 
 import click
 import requests
@@ -33,6 +33,15 @@ def first_non_null(ls: List[Optional[str]]) -> Optional[str]:
     return next((el for el in ls if el is not None and el.strip() != ""), None)
 
 
+_T = TypeVar("_T")
+
+
+def get_or_else(value: Optional[_T], default: _T) -> _T:
+    # Normally we'd use `value or default`. However, that runs into issues
+    # when value is falsey but not None.
+    return value if value is not None else default
+
+
 def parse_run_restli_response(response: requests.Response) -> dict:
     response_json = response.json()
     if response.status_code != 200:
@@ -321,6 +330,8 @@ def get_frontend_session_login_as(
 def _ensure_valid_gms_url_acryl_cloud(url: str) -> str:
     if "acryl.io" not in url:
         return url
+    if url.endswith(":8080"):
+        url = url.replace(":8080", "")
     if url.startswith("http://"):
         url = url.replace("http://", "https://")
     if url.endswith("acryl.io"):
@@ -401,7 +412,7 @@ def generate_access_token(
 def ensure_has_system_metadata(
     event: Union[
         MetadataChangeProposal, MetadataChangeProposalWrapper, MetadataChangeEvent
-    ]
+    ],
 ) -> None:
     if event.systemMetadata is None:
         event.systemMetadata = SystemMetadataClass()

datahub/cli/delete_cli.py

@@ -1,8 +1,8 @@
 import logging
+import random
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from dataclasses import dataclass
 from datetime import datetime
-from random import choices
 from typing import Dict, List, Optional
 
 import click
@@ -457,11 +457,11 @@ def by_filter(
             click.echo("Found urns of multiple entity types")
             for entity_type, entity_urns in urns_by_type.items():
                 click.echo(
-                    f"- {len(entity_urns)} {entity_type} urn(s). Sample: {choices(entity_urns, k=min(5, len(entity_urns)))}"
+                    f"- {len(entity_urns)} {entity_type} urn(s). Sample: {random.sample(entity_urns, k=min(5, len(entity_urns)))}"
                 )
         else:
             click.echo(
-                f"Found {len(urns)} {entity_type} urn(s). Sample: {choices(urns, k=min(5, len(urns)))}"
+                f"Found {len(urns)} {entity_type} urn(s). Sample: {random.sample(urns, k=min(5, len(urns)))}"
             )
 
         if not force and not dry_run:

datahub/cli/docker_cli.py

@@ -296,9 +296,9 @@ def _restore(
     restore_indices: Optional[bool],
     primary_restore_file: Optional[str],
 ) -> int:
-    assert (
-        restore_primary or restore_indices
-    ), "Either restore_primary or restore_indices must be set"
+    assert restore_primary or restore_indices, (
+        "Either restore_primary or restore_indices must be set"
+    )
     msg = "datahub> "
     if restore_primary:
         msg += f"Will restore primary database from {primary_restore_file}. "
@@ -314,9 +314,9 @@ def _restore(
         assert primary_restore_file
         resolved_restore_file = os.path.expanduser(primary_restore_file)
         logger.info(f"Restoring primary db from backup at {resolved_restore_file}")
-        assert os.path.exists(
-            resolved_restore_file
-        ), f"File {resolved_restore_file} does not exist"
+        assert os.path.exists(resolved_restore_file), (
+            f"File {resolved_restore_file} does not exist"
+        )
         with open(resolved_restore_file) as fp:
             result = subprocess.run(
                 [

datahub/cli/ingest_cli.py

@@ -507,15 +507,11 @@ def list_source_runs(page_offset: int, page_size: int, urn: str, source: str) ->
         click.echo("No response received from the server.")
         return
 
-    # when urn or source filter does not match, exit gracefully
-    if (
-        not isinstance(data.get("data"), dict)
-        or "listIngestionSources" not in data["data"]
-    ):
-        click.echo("No matching ingestion sources found. Please check your filters.")
-        return
+    # a lot of responses can be null if there's errors in the run
+    ingestion_sources = (
+        data.get("data", {}).get("listIngestionSources", {}).get("ingestionSources", [])
+    )
 
-    ingestion_sources = data["data"]["listIngestionSources"]["ingestionSources"]
     if not ingestion_sources:
         click.echo("No ingestion sources or executions found.")
         return
@@ -526,18 +522,32 @@ def list_source_runs(page_offset: int, page_size: int, urn: str, source: str) ->
         name = ingestion_source.get("name", "N/A")
 
         executions = ingestion_source.get("executions", {}).get("executionRequests", [])
+
         for execution in executions:
+            if execution is None:
+                continue
+
             execution_id = execution.get("id", "N/A")
-            start_time = execution.get("result", {}).get("startTimeMs", "N/A")
-            start_time = (
-                datetime.fromtimestamp(start_time / 1000).strftime("%Y-%m-%d %H:%M:%S")
-                if start_time != "N/A"
-                else "N/A"
-            )
-            status = execution.get("result", {}).get("status", "N/A")
+            result = execution.get("result") or {}
+            status = result.get("status", "N/A")
+
+            try:
+                start_time = (
+                    datetime.fromtimestamp(
+                        result.get("startTimeMs", 0) / 1000
+                    ).strftime("%Y-%m-%d %H:%M:%S")
+                    if status != "DUPLICATE" and result.get("startTimeMs") is not None
+                    else "N/A"
+                )
+            except (TypeError, ValueError):
+                start_time = "N/A"
 
             rows.append([execution_id, name, start_time, status, urn])
 
+    if not rows:
+        click.echo("No execution data found.")
+        return
+
     click.echo(
         tabulate(
             rows,

datahub/cli/lite_cli.py

@@ -176,7 +176,7 @@ def get(
             )
         )
     end_time = time.time()
-    logger.debug(f"Time taken: {int((end_time - start_time)*1000.0)} millis")
+    logger.debug(f"Time taken: {int((end_time - start_time) * 1000.0)} millis")
 
 
 @lite.command()
@@ -228,7 +228,7 @@ def ls(path: Optional[str]) -> None:
     try:
         browseables = lite.ls(path)
         end_time = time.time()
-        logger.debug(f"Time taken: {int((end_time - start_time)*1000.0)} millis")
+        logger.debug(f"Time taken: {int((end_time - start_time) * 1000.0)} millis")
         auto_complete: List[AutoComplete] = [
             b.auto_complete for b in browseables if b.auto_complete is not None
         ]

datahub/cli/migrate.py

@@ -179,7 +179,7 @@ def dataplatform2instance_func(
 
     if not force and not dry_run:
         # get a confirmation from the operator before proceeding if this is not a dry run
-        sampled_urns_to_migrate = random.choices(
+        sampled_urns_to_migrate = random.sample(
             urns_to_migrate, k=min(10, len(urns_to_migrate))
         )
         sampled_new_urns: List[str] = [
@@ -193,7 +193,7 @@ def dataplatform2instance_func(
             if key
         ]
         click.echo(
-            f"Will migrate {len(urns_to_migrate)} urns such as {random.choices(urns_to_migrate, k=min(10, len(urns_to_migrate)))}"
+            f"Will migrate {len(urns_to_migrate)} urns such as {random.sample(urns_to_migrate, k=min(10, len(urns_to_migrate)))}"
         )
         click.echo(f"New urns will look like {sampled_new_urns}")
         click.confirm("Ok to proceed?", abort=True)
@@ -426,9 +426,9 @@ def batch_get_ids(
             entities_yielded += 1
             log.debug(f"yielding {x}")
             yield x
-        assert (
-            entities_yielded == num_entities
-        ), "Did not delete all entities, try running this command again!"
+        assert entities_yielded == num_entities, (
+            "Did not delete all entities, try running this command again!"
+        )
     else:
         log.error(f"Failed to execute batch get with {str(response.content)}")
         response.raise_for_status()

datahub/cli/specific/assertions_cli.py

@@ -136,9 +136,9 @@ def extras_list_to_dict(extras: List[str]) -> Dict[str, str]:
     extra_properties: Dict[str, str] = dict()
     for x in extras:
         parts = x.split("=")
-        assert (
-            len(parts) == 2
-        ), f"Invalid value for extras {x}, should be in format key=value"
+        assert len(parts) == 2, (
+            f"Invalid value for extras {x}, should be in format key=value"
+        )
         extra_properties[parts[0]] = parts[1]
     return extra_properties
 

datahub/cli/specific/structuredproperties_cli.py

@@ -1,9 +1,11 @@
 import json
 import logging
 from pathlib import Path
+from typing import Iterable
 
 import click
 from click_default_group import DefaultGroup
+from ruamel.yaml import YAML
 
 from datahub.api.entities.structuredproperties.structuredproperties import (
     StructuredProperties,
@@ -61,3 +63,85 @@ def get(urn: str, to_file: str) -> None:
                 )
         else:
             click.secho(f"Structured property {urn} does not exist")
+
+
+@properties.command(
+    name="list",
+)
+@click.option("--details/--no-details", is_flag=True, default=True)
+@click.option("--to-file", required=False, type=str)
+@telemetry.with_telemetry()
+def list(details: bool, to_file: str) -> None:
+    """List structured properties in DataHub"""
+
+    def to_yaml_list(
+        objects: Iterable[StructuredProperties],  # iterable of objects to dump
+        file: Path,
+    ) -> None:
+        # if file exists, first we read it
+        yaml = YAML(typ="rt")  # default, if not specfied, is 'rt' (round-trip)
+        yaml.indent(mapping=2, sequence=4, offset=2)
+        yaml.default_flow_style = False
+        serialized_objects = []
+        if file.exists():
+            with open(file, "r") as fp:
+                existing_objects = yaml.load(fp)  # this is a list of dicts
+                existing_objects = [
+                    StructuredProperties.parse_obj(obj) for obj in existing_objects
+                ]
+                objects = [obj for obj in objects]
+                # do a positional update of the existing objects
+                existing_urns = {obj.urn for obj in existing_objects}
+                # existing_urns = {obj["urn"] if "urn" in obj else f"urn:li:structuredProperty:{obj['id']}" for obj in existing_objects}
+                for i, obj in enumerate(existing_objects):
+                    # existing_urn = obj["urn"] if "urn" in obj else f"urn:li:structuredProperty:{obj['id']}"
+                    existing_urn = obj.urn
+                    # breakpoint()
+                    if existing_urn in {obj.urn for obj in objects}:
+                        existing_objects[i] = next(
+                            obj.dict(exclude_unset=True, exclude_none=True)
+                            for obj in objects
+                            if obj.urn == existing_urn
+                        )
+                new_objects = [
+                    obj.dict(exclude_unset=True, exclude_none=True)
+                    for obj in objects
+                    if obj.urn not in existing_urns
+                ]
+                serialized_objects = existing_objects + new_objects
+        else:
+            serialized_objects = [
+                obj.dict(exclude_unset=True, exclude_none=True) for obj in objects
+            ]
+
+        with open(file, "w") as fp:
+            yaml.dump(serialized_objects, fp)
+
+    with get_default_graph() as graph:
+        if details:
+            logger.info(
+                "Listing structured properties with details. Use --no-details for urns only"
+            )
+            structuredproperties = StructuredProperties.list(graph)
+            if to_file:
+                to_yaml_list(structuredproperties, Path(to_file))
+            else:
+                for structuredproperty in structuredproperties:
+                    click.secho(
+                        f"{json.dumps(structuredproperty.dict(exclude_unset=True, exclude_none=True), indent=2)}"
+                    )
+        else:
+            logger.info(
+                "Listing structured property urns only, use --details for more information"
+            )
+            structured_property_urns = StructuredProperties.list_urns(graph)
+            if to_file:
+                with open(to_file, "w") as f:
+                    for urn in structured_property_urns:
+                        f.write(f"{urn}\n")
+                click.secho(
+                    f"Structured property urns written to {to_file}", fg="green"
+                )
+            else:
+                for urn in structured_property_urns:
+                    click.secho(f"{urn}")

datahub/cli/timeline_cli.py

@@ -50,7 +50,7 @@ def pretty_id(id: Optional[str]) -> str:
     if id.startswith("urn:li:dataset"):
         dataset_key = dataset_urn_to_key(id)
         if dataset_key:
-            return f"{click.style('dataset', fg='cyan')}:{click.style(dataset_key.platform[len('urn:li:dataPlatform:'):], fg='white')}:{click.style(dataset_key.name, fg='white')}"
+            return f"{click.style('dataset', fg='cyan')}:{click.style(dataset_key.platform[len('urn:li:dataPlatform:') :], fg='white')}:{click.style(dataset_key.name, fg='white')}"
     # failed to prettify, return original
     return id
 

datahub/configuration/common.py

@@ -200,8 +200,7 @@ class IgnorableError(MetaError):
 
 @runtime_checkable
 class ExceptionWithProps(Protocol):
-    def get_telemetry_props(self) -> Dict[str, Any]:
-        ...
+    def get_telemetry_props(self) -> Dict[str, Any]: ...
 
 
 def should_show_stack_trace(exc: Exception) -> bool:

datahub/configuration/config_loader.py

@@ -19,64 +19,87 @@ from datahub.configuration.yaml import YamlConfigurationMechanism
 Environ = Mapping[str, str]
 
 
-def _resolve_element(element: str, environ: Environ) -> str:
-    if re.search(r"(\$\{).+(\})", element):
-        return expand(element, nounset=True, environ=environ)
-    elif element.startswith("$"):
-        try:
-            return expand(element, nounset=True, environ=environ)
-        except UnboundVariable:
-            return element
-    else:
-        return element
-
-
-def _resolve_list(ele_list: list, environ: Environ) -> list:
-    new_v: list = []
-    for ele in ele_list:
-        if isinstance(ele, str):
-            new_v.append(_resolve_element(ele, environ=environ))
-        elif isinstance(ele, list):
-            new_v.append(_resolve_list(ele, environ=environ))
-        elif isinstance(ele, dict):
-            new_v.append(resolve_env_variables(ele, environ=environ))
-        else:
-            new_v.append(ele)
-    return new_v
-
-
 def resolve_env_variables(config: dict, environ: Environ) -> dict:
-    new_dict: Dict[Any, Any] = {}
-    for k, v in config.items():
-        if isinstance(v, dict):
-            new_dict[k] = resolve_env_variables(v, environ=environ)
-        elif isinstance(v, list):
-            new_dict[k] = _resolve_list(v, environ=environ)
-        elif isinstance(v, str):
-            new_dict[k] = _resolve_element(v, environ=environ)
-        else:
-            new_dict[k] = v
-    return new_dict
+    # TODO: This is kept around for backwards compatibility.
+    return EnvResolver(environ).resolve(config)
 
 
 def list_referenced_env_variables(config: dict) -> Set[str]:
-    # This is a bit of a hack, but expandvars does a bunch of escaping
-    # and other logic that we don't want to duplicate here.
+    # TODO: This is kept around for backwards compatibility.
+    return EnvResolver(environ=os.environ).list_referenced_variables(config)
+
+
+class EnvResolver:
+    def __init__(self, environ: Environ, strict_env_syntax: bool = False):
+        self.environ = environ
+        self.strict_env_syntax = strict_env_syntax
 
-    vars = set()
+    def resolve(self, config: dict) -> dict:
+        return self._resolve_dict(config)
 
-    def mock_get_env(key: str, default: Optional[str] = None) -> str:
-        vars.add(key)
-        if default is not None:
-            return default
-        return "mocked_value"
+    @classmethod
+    def list_referenced_variables(
+        cls,
+        config: dict,
+        strict_env_syntax: bool = False,
+    ) -> Set[str]:
+        # This is a bit of a hack, but expandvars does a bunch of escaping
+        # and other logic that we don't want to duplicate here.
 
-    mock = unittest.mock.MagicMock()
-    mock.get.side_effect = mock_get_env
+        vars = set()
 
-    resolve_env_variables(config, environ=mock)
+        def mock_get_env(key: str, default: Optional[str] = None) -> str:
+            vars.add(key)
+            if default is not None:
+                return default
+            return "mocked_value"
+
+        mock = unittest.mock.MagicMock()
+        mock.get.side_effect = mock_get_env
+
+        resolver = EnvResolver(environ=mock, strict_env_syntax=strict_env_syntax)
+        resolver._resolve_dict(config)
+
+        return vars
+
+    def _resolve_element(self, element: str) -> str:
+        if re.search(r"(\$\{).+(\})", element):
+            return expand(element, nounset=True, environ=self.environ)
+        elif not self.strict_env_syntax and element.startswith("$"):
+            try:
+                return expand(element, nounset=True, environ=self.environ)
+            except UnboundVariable:
+                # TODO: This fallback is kept around for backwards compatibility, but
+                # doesn't make a ton of sense from first principles.
+                return element
+        else:
+            return element
 
-    return vars
+    def _resolve_list(self, ele_list: list) -> list:
+        new_v: list = []
+        for ele in ele_list:
+            if isinstance(ele, str):
+                new_v.append(self._resolve_element(ele))
+            elif isinstance(ele, list):
+                new_v.append(self._resolve_list(ele))
+            elif isinstance(ele, dict):
+                new_v.append(self._resolve_dict(ele))
+            else:
+                new_v.append(ele)
+        return new_v
+
+    def _resolve_dict(self, config: dict) -> dict:
+        new_dict: Dict[Any, Any] = {}
+        for k, v in config.items():
+            if isinstance(v, dict):
+                new_dict[k] = self._resolve_dict(v)
+            elif isinstance(v, list):
+                new_dict[k] = self._resolve_list(v)
+            elif isinstance(v, str):
+                new_dict[k] = self._resolve_element(v)
+            else:
+                new_dict[k] = v
+        return new_dict
 
 
 WRITE_TO_FILE_DIRECTIVE_PREFIX = "__DATAHUB_TO_FILE_"
@@ -159,7 +182,7 @@ def load_config_file(
 
     config = raw_config.copy()
     if resolve_env_vars:
-        config = resolve_env_variables(config, environ=os.environ)
+        config = EnvResolver(environ=os.environ).resolve(config)
     if process_directives:
         config = _process_directives(config)
 

datahub/configuration/git.py

@@ -121,9 +121,9 @@ class GitInfo(GitReference):
 
         repo: str = values["repo"]
         if repo.startswith(_GITHUB_PREFIX):
-            return f"git@github.com:{repo[len(_GITHUB_PREFIX):]}.git"
+            return f"git@github.com:{repo[len(_GITHUB_PREFIX) :]}.git"
         elif repo.startswith(_GITLAB_PREFIX):
-            return f"git@gitlab.com:{repo[len(_GITLAB_PREFIX):]}.git"
+            return f"git@gitlab.com:{repo[len(_GITLAB_PREFIX) :]}.git"
         else:
             raise ValueError(
                 "Unable to infer repo_ssh_locator from repo. Please set repo_ssh_locator manually."