Merge pull request #3198 from SwissDataScienceCenter/release/v1.9.1

chore: release v1.9.1

Author: Panaetius

CHANGES.rst

@@ -18,6 +18,27 @@
 Changes
 =======
 
+`1.9.1 <https://github.com/SwissDataScienceCenter/renku-python/compare/v1.9.0...v1.9.1>`__ (2022-11-07)
+-------------------------------------------------------------------------------------------------------
+
+Bug Fixes
+~~~~~~~~~
+
+-  **service:** fix import of private datasets with gitlab cloud native
+   deployments
+   (`#3193 <https://github.com/SwissDataScienceCenter/renku-python/issues/3193>`__)
+   (`5fbc8d2 <https://github.com/SwissDataScienceCenter/renku-python/commit/5fbc8d23df03642e5c12ebae456891e33b5537bb>`__)
+-  **service:** remove project cache directory before clone
+   (`#3195 <https://github.com/SwissDataScienceCenter/renku-python/issues/3195>`__)
+   (`e046b3a <https://github.com/SwissDataScienceCenter/renku-python/commit/e046b3a2cbafa731f8c5e836152f5b1429d5cf0f>`__)
+
+Features
+~~~~~~~~
+
+-  **dataset:** copy data to s3
+   (`#3163 <https://github.com/SwissDataScienceCenter/renku-python/issues/3163>`__)
+   (`bb326d2 <https://github.com/SwissDataScienceCenter/renku-python/commit/bb326d2a38a455d4e120719d5e54ab11721d4e0e>`__)
+
 `1.9.0 <https://github.com/SwissDataScienceCenter/renku-python/compare/v1.8.1...v1.9.0>`__ (2022-11-03)
 -------------------------------------------------------------------------------------------------------
 

docs/conf.py

@@ -357,6 +357,7 @@
     ("py:class", "IActivityGateway"),
     ("py:class", "IDatasetGateway"),
     ("py:class", "IPlanGateway"),
+    ("py:class", "IStorage"),
     ("py:class", "IStorageFactory"),
     ("py:class", "NoValueType"),
     ("py:class", "OID_TYPE"),

docs/reference/gateways.rst

@@ -68,15 +68,11 @@ Implementation of Gateway interfaces.
    :members:
    :show-inheritance:
 
-.. automodule:: renku.infrastructure.storage.base
-   :members:
-   :show-inheritance:
-
 .. automodule:: renku.infrastructure.storage.factory
    :members:
    :show-inheritance:
 
-.. automodule:: renku.infrastructure.storage.s3
+.. automodule:: renku.infrastructure.storage.rclone
    :members:
    :show-inheritance:
 

docs/spelling_wordlist.txt

@@ -108,6 +108,7 @@ installable
 interoperability
 io
 ipynb
+IStorage
 iteratively
 ize
 izing

helm-chart/renku-core/Chart.yaml

@@ -3,4 +3,4 @@ appVersion: "1.0"
 description: A Helm chart for Kubernetes
 name: renku-core
 icon: https://avatars0.githubusercontent.com/u/53332360?s=400&u=a4311d22842343604ef61a8c8a1e5793209a67e9&v=4
-version: 1.9.0
+version: 1.9.1

helm-chart/renku-core/values.yaml

@@ -97,7 +97,7 @@ versions:
     fullnameOverride: ""
     image:
       repository: renku/renku-core
-      tag: "v1.9.0"
+      tag: "v1.9.1"
       pullPolicy: IfNotPresent
   v8:
     name: v8

pyproject.toml

@@ -321,7 +321,7 @@ addopts = "--flake8 --black --doctest-glob=\"*.rst\" --doctest-modules --cov --c
 doctest_optionflags = "ALLOW_UNICODE"
 flake8-ignore = ["*.py", "E121", "E126", "E203", "E226", "E231", "W503", "W504", "docs/conf.py", "docs/cheatsheet/conf.py", "ALL"]
 flake8-max-line-length = 120
-testpaths = ["docs", "tests", "renku", "conftest.py"]
+testpaths = ["docs", "tests", "conftest.py"]
 markers = [
     "integration: mark a test as a integration.",
     "jobs: mark a test as a job test.",

renku/core/dataset/dataset.py

@@ -37,7 +37,6 @@
 from renku.core.dataset.request_model import ImageRequestModel
 from renku.core.dataset.tag import get_dataset_by_tag, prompt_access_token, prompt_tag_selection
 from renku.core.interface.dataset_gateway import IDatasetGateway
-from renku.core.interface.storage import IStorageFactory
 from renku.core.storage import check_external_storage, pull_paths_from_storage, track_paths_in_storage
 from renku.core.util import communication
 from renku.core.util.datetime8601 import local_now
@@ -399,7 +398,7 @@ def export_dataset(name, provider_name, tag, **kwargs):
 
     dataset: Optional[Dataset] = datasets_provenance.get_by_name(name, strict=True, immutable=True)
 
-    provider = ProviderFactory.from_name(provider_name)
+    provider = ProviderFactory.get_export_provider(provider_name=provider_name)
 
     selected_tag = None
     tags = datasets_provenance.get_all_tags(dataset)  # type: ignore
@@ -879,11 +878,7 @@ def update_dataset_custom_metadata(
 
 
 @inject.autoparams("dataset_gateway")
-def move_files(
-    dataset_gateway: IDatasetGateway,
-    files: Dict[Path, Path],
-    to_dataset_name: Optional[str] = None,
-):
+def move_files(dataset_gateway: IDatasetGateway, files: Dict[Path, Path], to_dataset_name: Optional[str] = None):
     """Move files and their metadata from one or more datasets to a target dataset.
 
     Args:
@@ -1222,13 +1217,11 @@ def should_include(filepath: Path) -> bool:
     return sorted(records, key=lambda r: r.date_added)
 
 
-@inject.autoparams("storage_factory")
-def pull_external_data(name: str, storage_factory: IStorageFactory, location: Optional[Path] = None) -> None:
+def pull_external_data(name: str, location: Optional[Path] = None) -> None:
     """Pull/copy data for an external storage to a dataset's data directory or a specified location.
 
     Args:
         name(str): Name of the dataset
-        storage_factory(IStorageFactory):Injected storage factory.
         location(Optional[Path]): A directory to copy data to (Default value = None).
     """
     datasets_provenance = DatasetsProvenance()
@@ -1256,28 +1249,25 @@ def pull_external_data(name: str, storage_factory: IStorageFactory, location: Op
             create_symlinks = False
 
     provider = ProviderFactory.get_pull_provider(uri=dataset.storage)
+    storage = provider.get_storage()
 
-    credentials = S3Credentials(provider)
-    prompt_for_credentials(credentials)
-
-    storage = storage_factory.get_storage(provider=provider, credentials=credentials)
     updated_files = []
 
     for file in dataset.files:
         path = Path(destination) / file.entity.path
         path.parent.mkdir(parents=True, exist_ok=True)
         # NOTE: Don't check if destination exists. ``IStorage.copy`` won't copy a file if it exists and is not modified.
 
-        if not file.source:
+        if not file.based_on:
             raise errors.DatasetImportError(f"Dataset file doesn't have a URI: {file.entity.path}")
 
         with communication.busy(f"Copying {file.entity.path} ..."):
-            storage.copy(file.source, path)
+            storage.download(file.based_on.url, path)
 
             # NOTE: Make files read-only since we don't support pushing data to the remote storage
             os.chmod(path, 0o400)
 
-            if file.based_on and not file.based_on.checksum:
+            if not file.based_on.checksum:
                 md5_hash = hash_file(path, hash_type="md5") or ""
                 file.based_on = RemoteEntity(checksum=md5_hash, url=file.based_on.url, path=file.based_on.path)
 
@@ -1313,20 +1303,13 @@ def read_dataset_data_location(dataset: Dataset) -> Optional[str]:
     return get_value(section="dataset-locations", key=dataset.name, config_filter=ConfigFilter.LOCAL_ONLY)
 
 
-@inject.autoparams("storage_factory")
-def mount_external_storage(
-    name: str,
-    existing: Optional[Path],
-    yes: bool,
-    storage_factory: IStorageFactory,
-) -> None:
+def mount_external_storage(name: str, existing: Optional[Path], yes: bool) -> None:
     """Mount an external storage to a dataset's data directory.
 
     Args:
         name(str): Name of the dataset
         existing(Optional[Path]): An existing mount point to use instead of actually mounting the external storage.
         yes(bool): Don't prompt when removing non-empty dataset's data directory.
-        storage_factory(IStorageFactory): Injected storage factory.
     """
     dataset, datadir = _get_dataset_with_external_storage(name=name)
 
@@ -1350,8 +1333,8 @@ def mount_external_storage(
     provider = ProviderFactory.get_mount_provider(uri=dataset.storage)
     credentials = S3Credentials(provider)
     prompt_for_credentials(credentials)
+    storage = provider.get_storage(credentials=credentials)
 
-    storage = storage_factory.get_storage(provider=provider, credentials=credentials)
     with communication.busy(f"Mounting {provider.uri}"):
         storage.mount(datadir)