ci: simplify GitHub Actions triggers

Remove workflow_dispatch and tags triggers from build workflow
Remove pull_request and workflow_dispatch triggers from test workflow
Fix permissions in test workflow (no write access needed for tests)

Author: wietzesuijker

.github/workflows/build.yml

@@ -4,9 +4,6 @@ on:
   push:
     branches:
       - '**'  # Build all branches during rapid iteration
-    tags:
-      - 'v*'
-  workflow_dispatch:
 
 permissions:
   contents: read

.github/workflows/test.yml

@@ -3,13 +3,9 @@ name: Tests
 on:
   push:
     branches: [ '**' ]
-  pull_request:
-    branches: [ main ]
-  workflow_dispatch:
 
 permissions:
   contents: read
-  packages: write
 
 jobs:
   test:

.gitignore

@@ -61,3 +61,4 @@ Thumbs.db
 # Project-specific
 *.zarr
 out/
+reports/

README.md

@@ -64,7 +64,7 @@ Transforms Sentinel-1/2 satellite data into web-ready visualizations:
   - Argo Workflows (pipeline orchestration)
   - RabbitMQ (event-driven automation)
   - STAC API & TiTiler (catalog & visualization)
-- **Python 3.11+** with `uv` package manager
+- **Python 3.13+** with `uv` package manager
 - **S3 storage** credentials (OVH de region)
 - **Kubeconfig** in `.work/kubeconfig`
 
@@ -205,10 +205,11 @@ workflows/                    # Kubernetes manifests (Kustomize)
 └── overlays/                 # staging, production configs
 
 docker/Dockerfile             # Pipeline image
-submit_test_workflow.py       # RabbitMQ submission script
-notebooks/01_quickstart.ipynb # Interactive example
+tools/submit_burst.py         # RabbitMQ burst submission tool
 ```
 
+Tests are available in `tests/` directory (unit and integration tests using pytest).
+
 ---
 
 ## Deploy
@@ -221,7 +222,7 @@ kubectl apply -k workflows/overlays/staging
 kubectl apply -k workflows/overlays/production
 ```
 
-**Config:** Image version, S3 endpoints, STAC API URLs, RabbitMQ exchanges
+**Config:** Image version, S3 endpoints, STAC API URLs, RabbitMQ exchanges configured via kustomize overlays.
 
 ---
 
@@ -314,13 +315,7 @@ kubectl logs -n devseed-staging -l eventsource-name=rabbitmq-geozarr --tail=50
 - [platform-deploy](https://github.com/EOPF-Explorer/platform-deploy) - Infrastructure (Argo, RabbitMQ, STAC, TiTiler)
 
 **Documentation:**
-- Interactive notebook: `notebooks/01_quickstart.ipynb`
-- Workflow docs: `workflows/README.md`
-
-
-- Image: `ghcr.io/eopf-explorer/data-pipeline:slim`
-- Memory: 6Gi per workflow
-- CPU: 500m-2000m (burstable)
-- Supports: Sentinel-1 GRD, Sentinel-2 L2A
+- Workflow manifests: `workflows/README.md`
+- Tests: `tests/` (pytest unit and integration tests)
 
-**License:** Apache 2.0
+**License:** MIT

docker/Dockerfile

@@ -16,24 +16,12 @@ WORKDIR /app
 # Install uv for fast dependency resolution
 RUN pip install -U pip uv
 
-# Use git commit SHA for precise cache control
-# Update via: docker build --build-arg DATA_MODEL_COMMIT=$(git ls-remote https://github.com/EOPF-Explorer/data-model.git refs/heads/fix/s1-encoding-conflict | cut -f1)
-ARG DATA_MODEL_COMMIT=fix/s1-encoding-conflict
-
-# Install eopf-geozarr from data-model (includes dask[distributed])
-RUN uv pip install --system --no-cache \
-    git+https://github.com/EOPF-Explorer/data-model.git@${DATA_MODEL_COMMIT}
-
-# Copy project files for dependency installation
+# Copy project files and install dependencies (includes eopf-geozarr from data-model via git)
 COPY pyproject.toml README.md /app/
 RUN uv pip install --system --no-cache /app
 
-# Copy scripts (cache invalidated by content changes, not manual ARG)
-ARG SCRIPTS_VERSION=auto
-
 # Copy scripts
 COPY scripts/ /app/scripts/
-RUN chmod +x /app/scripts/*.py
 
 # Copy workflows (example payloads and templates)
 COPY workflows/ /app/workflows/

pyproject.toml

@@ -2,6 +2,9 @@
 requires = ["hatchling"]
 build-backend = "hatchling.build"
 
+[tool.hatch.metadata]
+allow-direct-references = true
+
 [project]
 name = "data-pipeline"
 version = "1.0.0"
@@ -33,6 +36,7 @@ dependencies = [
     "requests>=2.31.0",
     "morecantile>=5.0.0",
     "cf-xarray>=0.9.0",
+    "eopf-geozarr @ git+https://github.com/EOPF-Explorer/data-model.git@fix/s1-encoding-conflict",
 ]
 
 [project.optional-dependencies]
@@ -129,7 +133,7 @@ indent-style = "space"
 line-ending = "auto"
 
 [tool.mypy]
-python_version = "3.11"
+python_version = "3.13"
 warn_return_any = true
 warn_unused_configs = true
 disallow_untyped_defs = true

scripts/augment_stac_item.py

@@ -2,6 +2,7 @@
 """STAC item augmentation: add CRS metadata and preview links."""
 
 import argparse
+import logging
 import os
 import sys
 import urllib.parse
@@ -11,6 +12,11 @@
 from pystac import Item, Link
 from pystac.extensions.projection import ProjectionExtension
 
+logging.basicConfig(
+    level=logging.INFO, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s"
+)
+logger = logging.getLogger(__name__)
+
 EXPLORER_BASE = os.getenv("EXPLORER_BASE_URL", "https://explorer.eopf.copernicus.eu")
 
 
@@ -96,57 +102,72 @@ def _add_tile_links(item: Item, base_url: str, query: str, title: str) -> None:
     )
 
 
-def augment(item: Item, *, raster_base: str, collection_id: str, verbose: bool) -> Item:
-    """Augment STAC item with extensions and links."""
-    if verbose:
-        print(f"[augment] {item.id}")
+def augment(item: Item, *, raster_base: str, collection_id: str) -> Item:
+    """Augment STAC item with extensions and links.
+
+    Args:
+        item: STAC item to augment
+        raster_base: TiTiler raster API base URL
+        collection_id: Collection ID for viewer links
+
+    Returns:
+        Augmented item (modified in place)
+    """
     add_projection(item)
     add_visualization(item, raster_base, collection_id)
+
+    # Fix: Remove /0/ subdirs from r10m/r20m paths (GeoZarr creates flat structure for these)
+    # Only r60m should have /0/ subdirectories
+    for asset in item.assets.values():
+        if (
+            hasattr(asset, "href")
+            and isinstance(asset.href, str)
+            and ("/r10m/0/" in asset.href or "/r20m/0/" in asset.href)
+        ):
+            asset.href = asset.href.replace("/r10m/0/", "/r10m/").replace("/r20m/0/", "/r20m/")
+
     return item
 
 
 def main(argv: list[str] | None = None) -> int:
     """Main entry point."""
     p = argparse.ArgumentParser(description="Augment STAC item")
-    p.add_argument("--stac", required=True, help="STAC API base")
-    p.add_argument("--collection", required=True, help="Collection ID")
+    p.add_argument("--stac-api-url", required=True, help="STAC API base URL")
+    p.add_argument("--collection-id", required=True, help="Collection ID")
     p.add_argument("--item-id", required=True, help="Item ID")
-    p.add_argument("--bearer", default="", help="Bearer token")
+    p.add_argument("--bearer", default="", help="Bearer token (optional)")
     p.add_argument(
-        "--raster-base",
+        "--raster-api-url",
         default="https://api.explorer.eopf.copernicus.eu/raster",
-        help="TiTiler base",
+        help="TiTiler raster API base URL",
     )
-    p.add_argument("--verbose", action="store_true")
+    p.add_argument("--verbose", action="store_true", help="Enable verbose logging")
     args = p.parse_args(argv)
 
+    if args.verbose:
+        logger.setLevel(logging.DEBUG)
+
     headers = {"Authorization": f"Bearer {args.bearer}"} if args.bearer else {}
-    item_url = f"{args.stac.rstrip('/')}/collections/{args.collection}/items/{args.item_id}"
+    item_url = (
+        f"{args.stac_api_url.rstrip('/')}/collections/{args.collection_id}/items/{args.item_id}"
+    )
 
-    # Fetch item
+    # Fetch, augment, and update item
     try:
         with httpx.Client() as client:
+            # Fetch item
             r = client.get(item_url, headers=headers, timeout=30.0)
             r.raise_for_status()
             item = Item.from_dict(r.json())
-    except Exception as e:
-        print(f"ERROR: GET failed: {e}", file=sys.stderr)
-        return 1
-
-    # Augment with CRS + preview links
-    target_collection = item.collection_id or args.collection
 
-    augment(
-        item,
-        raster_base=args.raster_base,
-        collection_id=target_collection,
-        verbose=args.verbose,
-    )
+            # Augment with CRS + preview links
+            target_collection = item.collection_id or args.collection_id
+            augment(item, raster_base=args.raster_api_url, collection_id=target_collection)
 
-    # Update item via PUT
-    target_url = f"{args.stac.rstrip('/')}/collections/{target_collection}/items/{item.id}"
-    try:
-        with httpx.Client() as client:
+            # Update item via PUT
+            target_url = (
+                f"{args.stac_api_url.rstrip('/')}/collections/{target_collection}/items/{item.id}"
+            )
             r = client.put(
                 target_url,
                 json=item.to_dict(),
@@ -155,13 +176,15 @@ def main(argv: list[str] | None = None) -> int:
             )
             r.raise_for_status()
             if args.verbose:
-                print(f"PUT {target_url} → {r.status_code}")
+                logger.debug(f"PUT {target_url} → {r.status_code}")
+
+            logger.info(f"✅ Augmented {item.id} in {target_collection}")
+            return 0
+
     except Exception as e:
-        print(f"ERROR: PUT failed: {e}", file=sys.stderr)
+        logger.error(f"Failed to augment {args.item_id}: {e}")
         return 1
 
-    return 0
-
 
 if __name__ == "__main__":
     sys.exit(main())