Issue/cleaner docker (#22)

* update lock file

* make docker more simple

* remove cpu only

* tidy

* remove venv

* update entrypoint

* fix docker (runs locally)

* update generation data for pv

* update work flow

* fix

* update github workflow

* add logging

* tidy up

* remove "as base" from docker file

* no-cache

* ContainerFile

* include ContainerFile

* remove main ref

* use v1 docker buildx

* remove docker file

.dockerignore

@@ -4,3 +4,4 @@
 !poetry.lock
 !pyproject.toml
 !README.md
+!ContainerFile

.github/workflows/docker-build.yml

@@ -3,6 +3,15 @@ name: Docker Build
 on: push
 
 jobs:
-
   DockerBuild:
-    uses: openclimatefix/.github/.github/workflows/[email protected]
+    name: Build docker
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v1
+
+      - name: Build
+        run: docker build . -f ContainerFile --no-cache

ContainerFile

@@ -0,0 +1,31 @@
+FROM python:3.11-slim
+
+RUN apt-get update
+RUN apt-get install -y git
+
+ENV PYTHONFAULTHANDLER=1 \
+	PYTHONHASHSEED=random \
+	PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+RUN apt-get update
+RUN apt-get install -y gdal-bin libgdal-dev g++
+
+ENV PIP_DEFAULT_TIMEOUT=100 \
+	PIP_DISABLE_PIP_VERSION_CHECK=1 \
+	PIP_NO_CACHE_DIR=1 \
+	POETRY_VERSION=1.7.1
+
+RUN pip install "poetry==$POETRY_VERSION"
+
+COPY pyproject.toml poetry.lock README.md .
+RUN poetry install --no-dev --no-root
+
+COPY india_forecast_app ./india_forecast_app
+RUN poetry build
+RUN poetry install
+
+COPY nwp.zarr ./nwp.zarr
+
+ENTRYPOINT ["poetry", "run" , "python3", "india_forecast_app/app.py", "--write-to-db"]

india_forecast_app/app.py

@@ -15,7 +15,7 @@
 from pvsite_datamodel.write import insert_forecast_values
 from sqlalchemy.orm import Session
 
-from .models import DummyModel, PVNetModel
+from india_forecast_app.models import DummyModel, PVNetModel
 
 log = logging.getLogger(__name__)
 
@@ -60,6 +60,7 @@ def get_generation_data(
     # pad by 1 second to ensure get_pv_generation_by_sites returns correct data
     end = timestamp + dt.timedelta(seconds=1)
 
+    log.info(f'Getting generation data for sites: {site_uuids}, from {start=} to {end=}')
     generation_data = get_pv_generation_by_sites(
         session=db_session, site_uuids=site_uuids, start_utc=start, end_utc=end
     )
@@ -243,7 +244,10 @@ def app(timestamp: dt.datetime | None, write_to_db: bool, log_level: str):
             asset_sites = pv_sites if asset_type == "pv" else wind_sites
             if len(asset_sites) > 0:
                 log.info(f"Reading latest historic {asset_type} generation data...")
-                generation_data = get_generation_data(session, asset_sites, timestamp)
+                if asset_type == "wind":
+                    generation_data = get_generation_data(session, asset_sites, timestamp)
+                else:
+                    generation_data = {"data": pd.DataFrame(), "metadata": pd.DataFrame()}
                 log.info(f"Loading {asset_type} model...")
                 models[asset_type] = get_model(asset_type, timestamp, generation_data)
                 log.info(f"{asset_type} model loaded")