feat: run notebooks in data service (#375)

Co-authored-by: Samuel Gaist <[email protected]>

squashme: resolve package version conflicts

.devcontainer/devcontainer.json

@@ -20,7 +20,10 @@
 		"ghcr.io/devcontainers/features/kubectl-helm-minikube:1": {
 			"minikube": "none"
 		},
-		"ghcr.io/eitsupi/devcontainer-features/jq-likes:2": {},
+		"ghcr.io/eitsupi/devcontainer-features/jq-likes:2": {
+			"jqVersion": "latest",
+			"yqVersion": "latest"
+		},
 		"ghcr.io/dhoeric/features/k9s:1": {},
 		"ghcr.io/EliiseS/devcontainer-features/bash-profile:1": {
 			"command": "alias k=kubectl"

.devcontainer/docker-compose.yml

@@ -23,6 +23,8 @@ services:
       ZED_TOKEN: renku
       ZED_INSECURE: "true"
       POETRY_CACHE_DIR: "/poetry_cache"
+      NB_SERVER_OPTIONS__DEFAULTS_PATH: /workspace/server_defaults.json
+      NB_SERVER_OPTIONS__UI_CHOICES_PATH: /workspace/server_options.json
     network_mode: service:db
     depends_on:
       - db
@@ -43,6 +45,7 @@ services:
       - "8080:8080"
       - "5678:5678"
       - "50051:50051"
+      - "8888:80"
 
   swagger:
     image: swaggerapi/swagger-ui

.github/workflows/acceptance-tests.yml

@@ -25,6 +25,8 @@ jobs:
       renku-graph: ${{ steps.deploy-comment.outputs.renku-graph}}
       renku-notebooks: ${{ steps.deploy-comment.outputs.renku-notebooks}}
       renku-ui: ${{ steps.deploy-comment.outputs.renku-ui}}
+      amalthea-sessions: ${{ steps.deploy-comment.outputs.amalthea-sessions}}
+      amalthea: ${{ steps.deploy-comment.outputs.amalthea}}
       test-enabled: ${{ steps.deploy-comment.outputs.test-enabled}}
       test-cypress-enabled: ${{ steps.deploy-comment.outputs.test-cypress-enabled}}
       persist: ${{ steps.deploy-comment.outputs.persist}}
@@ -84,6 +86,8 @@ jobs:
           renku_graph: "${{ needs.check-deploy.outputs.renku-graph }}"
           renku_notebooks: "${{ needs.check-deploy.outputs.renku-notebooks }}"
           renku_data_services: "@${{ github.head_ref }}"
+          amalthea: "${{ needs.check-deploy.outputs.amalthea }}"
+          amalthea_sessions: "${{ needs.check-deploy.outputs.amalthea-sessions }}"
           extra_values: "${{ needs.check-deploy.outputs.extra-values }}"
 
   selenium-acceptance-tests:

.github/workflows/save_cache.yml

@@ -0,0 +1,41 @@
+name: Create cache from commits on main
+
+on:
+  push:
+    branches:
+      - main
+      - chore-add-kind
+  workflow_dispatch:
+
+
+jobs:
+  save-poetry-cache:
+    runs-on: ubuntu-latest
+    env:
+      CACHE_KEY: main-branch-poetry-cache-ubuntu
+      CACHE_PATH: .devcontainer/.poetry_cache
+      DEVCONTAINER_IMAGE_CACHE: ghcr.io/swissdatasciencecenter/renku-data-services/devcontainer
+
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+      - name: Login to Docker Hub
+        uses: docker/login-action@v2
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+      - name: Install python deps
+        uses: devcontainers/[email protected]
+        with:
+          runCmd: poetry install --with dev
+          push: always
+          skipContainerUserIdUpdate: false
+          imageName: ${{ env.DEVCONTAINER_IMAGE_CACHE }}
+          cacheFrom: ${{ env.DEVCONTAINER_IMAGE_CACHE }}
+      - uses: actions/cache/save@v3
+        name: Create cache
+        with:
+          path: ${{ env.CACHE_PATH }}
+          key: ${{ env.CACHE_KEY }}

.github/workflows/test_publish.yml

@@ -70,6 +70,11 @@ jobs:
       - uses: actions/checkout@v4
         with:
           fetch-depth: 0
+      - uses: actions/cache/restore@v3
+        name: Restore cache
+        with:
+          path: ${{ env.CACHE_PATH }}
+          key: ${{ env.CACHE_KEY }}
       - name: Set Git config
         shell: bash
         run: |
@@ -111,6 +116,11 @@ jobs:
       - uses: actions/checkout@v4
         with:
           fetch-depth: 0
+      - uses: actions/cache/restore@v3
+        name: Restore cache
+        with:
+          path: ${{ env.CACHE_PATH }}
+          key: ${{ env.CACHE_KEY }}
       - name: Set Git config
         shell: bash
         run: |
@@ -155,6 +165,11 @@ jobs:
       - uses: actions/checkout@v4
         with:
           fetch-depth: 0
+      - uses: actions/cache/restore@v3
+        name: Restore cache
+        with:
+          path: ${{ env.CACHE_PATH }}
+          key: ${{ env.CACHE_KEY }}
       - name: Set Git config
         shell: bash
         run: |

Makefile

@@ -1,8 +1,8 @@
 .PHONY: schemas tests test_setup main_tests schemathesis_tests collect_coverage style_checks pre_commit_checks run download_avro check_avro avro_models update_avro kind_cluster install_amaltheas all
 
-AMALTHEA_JS_VERSION ?= 0.11.0
-AMALTHEA_SESSIONS_VERSION ?= 0.0.1-new-operator-chart
-codegen_params = --input-file-type openapi --output-model-type pydantic_v2.BaseModel --use-double-quotes --target-python-version 3.12 --collapse-root-models --field-constraints --strict-nullable --openapi-scopes schemas paths parameters --set-default-enum-member --use-one-literal-as-default --use-default
+AMALTHEA_JS_VERSION ?= 0.12.2
+AMALTHEA_SESSIONS_VERSION ?= 0.0.9-new-operator-chart
+codegen_params = --input-file-type openapi --output-model-type pydantic_v2.BaseModel --use-double-quotes --target-python-version 3.12 --collapse-root-models --field-constraints --strict-nullable --set-default-enum-member --openapi-scopes schemas paths parameters --set-default-enum-member --use-one-literal-as-default --use-default
 
 define test_apispec_up_to_date
 	$(eval $@_NAME=$(1))
@@ -161,7 +161,13 @@ kind_cluster:  ## Creates a kind cluster for testing
 	sleep 15
 	kubectl wait --namespace ingress-nginx --for=condition=ready pod --selector=app.kubernetes.io/component=controller --timeout=90s
 
-install_amaltheas:  ## Installs both version of amalthea in the currently active k8s context.
+install_amaltheas:  ## Installs both version of amalthea in the. NOTE: It uses the currently active k8s context.
 	helm repo add renku https://swissdatasciencecenter.github.io/helm-charts
-	helm install amalthea-js renku/amalthea --version $(AMALTHEA_JS_VERSION)
-	helm install amalthea-sessions renku/amalthea-sessions --version $(AMALTHEA_SESSIONS_VERSION)
+	helm repo update
+	helm upgrade --install amalthea-js renku/amalthea --version $(AMALTHEA_JS_VERSION)
+	helm upgrade --install amalthea-sessions amalthea-sessions-0.0.9-new-operator-chart.tgz --version $(AMALTHEA_SESSIONS_VERSION)
+
+# TODO: Add the version variables from the top of the file here when the charts are fully published
+amalthea_schema:  ## Updates generates pydantic classes from CRDs
+	curl https://raw.githubusercontent.com/SwissDataScienceCenter/amalthea/feat-add-cloud-storage/config/crd/bases/amalthea.dev_amaltheasessions.yaml | yq '.spec.versions[0].schema.openAPIV3Schema' | poetry run datamodel-codegen --input-file-type jsonschema --output-model-type pydantic_v2.BaseModel --output components/renku_data_services/notebooks/cr_amalthea_session.py --use-double-quotes --target-python-version 3.12 --collapse-root-models --field-constraints --strict-nullable --base-class renku_data_services.notebooks.cr_base.BaseCRD --allow-extra-fields --use-default-kwarg
+	curl https://raw.githubusercontent.com/SwissDataScienceCenter/amalthea/main/controller/crds/jupyter_server.yaml | yq '.spec.versions[0].schema.openAPIV3Schema' | poetry run datamodel-codegen --input-file-type jsonschema --output-model-type pydantic_v2.BaseModel --output components/renku_data_services/notebooks/cr_jupyter_server.py --use-double-quotes --target-python-version 3.12 --collapse-root-models --field-constraints --strict-nullable --base-class renku_data_services.notebooks.cr_base.BaseCRD --allow-extra-fields --use-default-kwarg

bases/renku_data_services/data_api/app.py

@@ -18,6 +18,7 @@
 from renku_data_services.data_connectors.blueprints import DataConnectorsBP
 from renku_data_services.message_queue.blueprints import MessageQueueBP
 from renku_data_services.namespace.blueprints import GroupsBP
+from renku_data_services.notebooks.blueprints import NotebooksBP, NotebooksNewBP
 from renku_data_services.platform.blueprints import PlatformConfigBP
 from renku_data_services.project.blueprints import ProjectsBP
 from renku_data_services.repositories.blueprints import RepositoriesBP
@@ -130,6 +131,24 @@ def register_all_handlers(app: Sanic, config: Config) -> Sanic:
         authenticator=config.authenticator,
         internal_gitlab_authenticator=config.gitlab_authenticator,
     )
+    notebooks = NotebooksBP(
+        name="notebooks_old",
+        url_prefix=url_prefix,
+        authenticator=config.authenticator,
+        nb_config=config.nb_config,
+        internal_gitlab_authenticator=config.gitlab_authenticator,
+        git_repo=config.git_repositories_repo,
+    )
+    notebooks_new = NotebooksNewBP(
+        name="notebooks",
+        url_prefix=url_prefix,
+        authenticator=config.authenticator,
+        nb_config=config.nb_config,
+        project_repo=config.project_repo,
+        session_repo=config.session_repo,
+        rp_repo=config.rp_repo,
+        internal_gitlab_authenticator=config.gitlab_authenticator,
+    )
     platform_config = PlatformConfigBP(
         name="platform_config",
         url_prefix=url_prefix,
@@ -175,6 +194,8 @@ def register_all_handlers(app: Sanic, config: Config) -> Sanic:
             oauth2_clients.blueprint(),
             oauth2_connections.blueprint(),
             repositories.blueprint(),
+            notebooks.blueprint(),
+            notebooks_new.blueprint(),
             platform_config.blueprint(),
             message_queue.blueprint(),
             data_connectors.blueprint(),

components/renku_data_services/app_config/config.py

@@ -59,6 +59,7 @@
 from renku_data_services.message_queue.interface import IMessageQueue
 from renku_data_services.message_queue.redis_queue import RedisQueue
 from renku_data_services.namespace.db import GroupRepository
+from renku_data_services.notebooks.config import _NotebooksConfig
 from renku_data_services.platform.db import PlatformRepository
 from renku_data_services.project.db import ProjectMemberRepository, ProjectRepository
 from renku_data_services.repositories.db import GitRepositoriesRepository
@@ -151,6 +152,7 @@ class Config:
     kc_api: IKeycloakAPI
     message_queue: IMessageQueue
     gitlab_url: str | None
+    nb_config: _NotebooksConfig
 
     secrets_service_public_key: rsa.RSAPublicKey
     """The public key of the secrets service, used to encrypt user secrets that only it can decrypt."""
@@ -228,6 +230,10 @@ def load_apispec() -> dict[str, Any]:
         with open(spec_file) as f:
             repositories = safe_load(f)
 
+        spec_file = Path(renku_data_services.notebooks.__file__).resolve().parent / "api.spec.yaml"
+        with open(spec_file) as f:
+            repositories = safe_load(f)
+
         spec_file = Path(renku_data_services.platform.__file__).resolve().parent / "api.spec.yaml"
         with open(spec_file) as f:
             platform = safe_load(f)
@@ -466,8 +472,8 @@ def from_env(cls, prefix: str = "") -> "Config":
         gitlab_client: base_models.GitlabAPIProtocol
         user_preferences_config: UserPreferencesConfig
         version = os.environ.get(f"{prefix}VERSION", "0.0.1")
-        server_options_file = os.environ.get("SERVER_OPTIONS")
-        server_defaults_file = os.environ.get("SERVER_DEFAULTS")
+        server_options_file = os.environ.get("NB_SERVER_OPTIONS__UI_CHOICES_PATH")
+        server_defaults_file = os.environ.get("NB_SERVER_OPTIONS__DEFAULTS_PATH")
         k8s_namespace = os.environ.get("K8S_NAMESPACE", "default")
         max_pinned_projects = int(os.environ.get(f"{prefix}MAX_PINNED_PROJECTS", "10"))
         user_preferences_config = UserPreferencesConfig(max_pinned_projects=max_pinned_projects)
@@ -549,6 +555,7 @@ def from_env(cls, prefix: str = "") -> "Config":
         sentry = SentryConfig.from_env(prefix)
         trusted_proxies = TrustedProxiesConfig.from_env(prefix)
         message_queue = RedisQueue(redis)
+        nb_config = _NotebooksConfig.from_env(db)
 
         return cls(
             version=version,
@@ -569,4 +576,5 @@ def from_env(cls, prefix: str = "") -> "Config":
             encryption_key=encryption_key,
             secrets_service_public_key=secrets_service_public_key,
             gitlab_url=gitlab_url,
+            nb_config=nb_config,
         )

components/renku_data_services/authn/dummy.py

@@ -7,6 +7,7 @@
 from typing import Optional
 
 from sanic import Request
+from ulid import ULID
 
 import renku_data_services.base_models as base_models
 
@@ -39,10 +40,22 @@ class DummyAuthenticator:
     """
 
     token_field = "Authorization"  # nosec: B105
+    anon_id_header_key: str = "Renku-Auth-Anon-Id"
+    anon_id_cookie_name: str = "Renku-Auth-Anon-Id"
 
-    @staticmethod
-    async def authenticate(access_token: str, request: Request) -> base_models.APIUser:
+    async def authenticate(self, access_token: str, request: Request) -> base_models.APIUser:
         """Indicates whether the user has successfully logged in."""
+        access_token = request.headers.get(self.token_field) or ""
+        if not access_token or len(access_token) == 0:
+            # Try to get an anonymous user ID if the validation of keycloak credentials failed
+            anon_id = request.headers.get(self.anon_id_header_key)
+            if anon_id is None:
+                anon_id = request.cookies.get(self.anon_id_cookie_name)
+            if anon_id is None:
+                anon_id = f"anon-{str(ULID())}"
+            return base_models.AnonymousAPIUser(id=str(anon_id))
+
+        access_token = access_token.removeprefix("Bearer ").removeprefix("bearer ")
         user_props = {}
         with contextlib.suppress(Exception):
             user_props = json.loads(access_token)
@@ -64,4 +77,5 @@ async def authenticate(access_token: str, request: Request) -> base_models.APIUs
             last_name=user_props.get("last_name", "Doe") if is_set else None,
             email=user_props.get("email", "[email protected]") if is_set else None,
             full_name=user_props.get("full_name", "John Doe") if is_set else None,
+            refresh_token=request.headers.get("Renku-Auth-Refresh-Token"),
         )

components/renku_data_services/authn/gitlab.py

@@ -2,10 +2,13 @@
 
 import contextlib
 import urllib.parse as parse
+from contextlib import suppress
 from dataclasses import dataclass
+from datetime import datetime
 
 import gitlab
 from sanic import Request
+from sanic.compat import Header
 
 import renku_data_services.base_models as base_models
 from renku_data_services import errors
@@ -23,6 +26,7 @@ class GitlabAuthenticator:
     gitlab_url: str
 
     token_field: str = "Gitlab-Access-Token"
+    expires_at_field: str = "Gitlab-Access-Token-Expires-At"
 
     def __post_init__(self) -> None:
         """Properly set gitlab url."""
@@ -36,10 +40,10 @@ async def authenticate(self, access_token: str, request: Request) -> base_models
         if self.token_field != "Authorization":  # nosec: B105
             access_token = str(request.headers.get(self.token_field))
 
-        result = await self._get_gitlab_api_user(access_token)
+        result = await self._get_gitlab_api_user(access_token, request.headers)
         return result
 
-    async def _get_gitlab_api_user(self, access_token: str) -> base_models.APIUser:
+    async def _get_gitlab_api_user(self, access_token: str, headers: Header) -> base_models.APIUser:
         """Get and validate a Gitlab API User."""
         client = gitlab.Gitlab(self.gitlab_url, oauth_token=access_token)
         try:
@@ -69,12 +73,18 @@ async def _get_gitlab_api_user(self, access_token: str) -> base_models.APIUser:
             if len(name_parts) >= 1:
                 last_name = " ".join(name_parts)
 
+        expires_at: datetime | None = None
+        expires_at_raw: str | None = headers.get(self.expires_at_field)
+        if expires_at_raw is not None and len(expires_at_raw) > 0:
+            with suppress(ValueError):
+                expires_at = datetime.fromtimestamp(float(expires_at_raw))
+
         return base_models.APIUser(
-            is_admin=False,
             id=str(user_id),
             access_token=access_token,
             first_name=first_name,
             last_name=last_name,
             email=email,
             full_name=full_name,
+            access_token_expires_at=expires_at,
         )