feat: added renku service with cache and datasets

Author: jsam

Dockerfile renamed to Dockerfile.cli

@@ -1,4 +1,4 @@
-FROM python:3.6-alpine as base
+FROM python:3.7-alpine as base
 
 RUN apk add --no-cache git && \
     pip install --no-cache --upgrade pip

Dockerfile.svc

@@ -0,0 +1,18 @@
+FROM python:3.7-alpine
+
+RUN apk add --update --no-cache alpine-sdk g++ gcc linux-headers libxslt-dev python3-dev build-base openssl-dev libffi-dev git && \
+    pip install --no-cache --upgrade pip setuptools pipenv requirements-builder
+
+RUN apk add --no-cache --allow-untrusted \
+    --repository http://dl-cdn.alpinelinux.org/alpine/latest-stable/community \
+    --repository http://dl-cdn.alpinelinux.org/alpine/latest-stable/main \
+    --repository http://nl.alpinelinux.org/alpine/edge/community \
+    git-lfs && \
+    git lfs install
+
+COPY . /code/renku
+WORKDIR /code/renku
+RUN requirements-builder -e all --level=pypi setup.py > requirements.txt && pip install -r requirements.txt && pip install -e . && pip install gunicorn
+
+
+ENTRYPOINT ["gunicorn", "renku.service.entrypoint:app", "-b", "0.0.0.0:8080"]

MANIFEST.in

@@ -39,6 +39,7 @@ include babel.ini
 include brew.py
 include pytest.ini
 include snap/snapcraft.yaml
+recursive-include renku *.json
 recursive-include .github CODEOWNERS
 recursive-include .travis *.sh
 recursive-include docs *.bat

Pipfile.lock

@@ -30,6 +30,8 @@
 import responses
 from click.testing import CliRunner
 
+from renku.service.entrypoint import create_app
+
 
 @pytest.fixture(scope='module')
 def renku_path(tmpdir_factory):
@@ -474,3 +476,40 @@ def sleep_after():
     import time
     yield
     time.sleep(0.5)
+
+
+@pytest.fixture(scope='module')
+def svc_client():
+    """Renku service client."""
+    flask_app = create_app()
+
+    testing_client = flask_app.test_client()
+    testing_client.testing = True
+
+    ctx = flask_app.app_context()
+    ctx.push()
+
+    yield testing_client
+
+    ctx.pop()
+
+
+@pytest.fixture(scope='function')
+def svc_client_with_repo(svc_client):
+    """Renku service remote repository."""
+    access_token = 'contact:EcfPJvEqjJepyu6XyqKZ'
+    remote_url = 'https://{0}@renkulab.io/gitlab/contact/integration-tests.git'
+    headers = {'Authorization': 'Bearer b4b4de0eda0f471ab82702bd5c367fa7'}
+
+    params = {'git_url': remote_url.format(access_token), 'force': 1}
+
+    response = svc_client.get(
+        '/cache/project-clone', query_string=params, headers=headers
+    )
+
+    assert response
+    assert 'result' in response.json
+    assert 'error' not in response.json
+    assert 'integration-tests' == response.json['result']['project_id']
+
+    yield svc_client, headers

conftest.py

@@ -25,6 +25,8 @@
 import yaml
 
 from renku.core.management import LocalClient
+from renku.core.management.config import RENKU_HOME
+from renku.core.management.repository import default_path
 
 from .git import get_git_isolation
 
@@ -63,8 +65,17 @@ def pass_local_client(
         )
 
     def new_func(*args, **kwargs):
-        ctx = click.get_current_context()
-        client = ctx.ensure_object(LocalClient)
+        ctx = click.get_current_context(silent=True)
+        if not ctx:
+            client = LocalClient(
+                path=default_path(),
+                renku_home=RENKU_HOME,
+                use_external_storage=True,
+            )
+            ctx = click.Context(click.Command(method))
+        else:
+            client = ctx.ensure_object(LocalClient)
+
         stack = contextlib.ExitStack()
 
         # Handle --isolation option:
@@ -85,8 +96,11 @@ def new_func(*args, **kwargs):
         if lock or (lock is None and commit):
             stack.enter_context(client.lock)
 
-        with stack:
-            result = ctx.invoke(method, client, *args, **kwargs)
+        result = None
+        if ctx:
+            with stack:
+                result = ctx.invoke(method, client, *args, **kwargs)
+
         return result
 
     return functools.update_wrapper(new_func, method)

renku/core/commands/client.py

@@ -106,7 +106,7 @@ def create_dataset(client, name, handle_duplicate_fn=None):
     :raises: ``renku.core.errors.ParameterError``
     """
     existing = client.load_dataset(name=name)
-    if (not existing or handle_duplicate_fn and handle_duplicate_fn(existing)):
+    if not existing or handle_duplicate_fn and handle_duplicate_fn(existing):
         with client.with_dataset(name=name) as dataset:
             creator = Creator.from_git(client.repo)
             if creator not in dataset.creator:
@@ -143,9 +143,12 @@ def add_file(
     sources=(),
     destination='',
     with_metadata=None,
-    urlscontext=contextlib.nullcontext
+    urlscontext=contextlib.nullcontext,
+    use_external_storage=True
 ):
     """Add data file to a dataset."""
+    client.use_external_storage = use_external_storage
+
     add_to_dataset(
         client, urls, name, link, force, sources, destination, with_metadata,
         urlscontext

renku/core/commands/dataset.py

@@ -296,7 +296,8 @@ def _add_from_url(self, dataset, dataset_path, url, link, destination):
         mode = dst.stat().st_mode & 0o777
         dst.chmod(mode & ~(stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH))
 
-        self.track_paths_in_storage(str(dst.relative_to(self.path)))
+        if self.has_external_storage:
+            self.track_paths_in_storage(str(dst.relative_to(self.path)))
 
         return [{
             'path': dst.relative_to(self.path),

renku/core/management/datasets.py

@@ -47,13 +47,18 @@ def default_path():
         return '.'
 
 
+def path_converter(path):
+    """Converter for path in PathMixin."""
+    return Path(path).resolve()
+
+
 @attr.s
 class PathMixin:
     """Define a default path attribute."""
 
     path = attr.ib(
         default=default_path,
-        converter=lambda arg: Path(arg).resolve().absolute(),
+        converter=path_converter,
     )
 
     @path.validator

renku/core/management/repository.py

@@ -0,0 +1,6 @@
+FROM python:3.6
+ADD . /app
+WORKDIR /app
+RUN pip install -e .
+EXPOSE 8000
+CMD ["gunicorn", "-b", "0.0.0.0:8000", "app"]