add RTS to wuhan

packages/gnss-product-management/src/gnss_product_management/client/gnss_client.py

@@ -281,6 +281,7 @@ def resolve_dependencies(
         date: datetime.datetime,
         *,
         sink_id: str,
+        force_download: bool = False,
     ) -> tuple[DependencyResolution, AnyPath | None]:
         """Resolve all dependencies in a spec for the given date.
 
@@ -312,4 +313,9 @@ def resolve_dependencies(
             workspace=self._workspace,
             transport=self._transport,
         )
-        return pipeline.run(dep_spec, date, sink_id=sink_id)
+        return pipeline.run(
+            dep_spec,
+            date,
+            sink_id=sink_id,
+            force_download=force_download,
+        )

packages/gnss-product-management/src/gnss_product_management/client/product_query.py

@@ -24,6 +24,12 @@
 logger = logging.getLogger(__name__)
 
 
+def _remote_uri(protocol: str, hostname: str, directory: str, filename: str) -> str:
+    """Build a remote URI without duplicating schemes or path separators."""
+    base = hostname if "://" in hostname else f"{protocol}://{hostname}"
+    return f"{base.rstrip('/')}/{directory.strip('/')}/{filename}"
+
+
 class ProductQuery:
     """Fluent builder for constructing and executing a GNSS product search.
 
@@ -352,8 +358,11 @@ def search(self) -> list[FoundResource]:
                 )
             else:
                 proto = (rq.server.protocol or "ftp").lower()
-                uri = (
-                    f"{proto}://{hostname}/{rq.directory.value or rq.directory.pattern}/{filename}"  # type: ignore[union-attr]
+                uri = _remote_uri(
+                    proto,
+                    hostname,
+                    rq.directory.value or rq.directory.pattern,  # type: ignore[union-attr]
+                    filename,
                 )
             r = FoundResource(
                 product=rq.product.name,

packages/gnss-product-management/src/gnss_product_management/factories/connection_pool.py

@@ -2,17 +2,23 @@
 
 import logging
 import os
+import re
 import threading
 from contextlib import contextmanager
 from pathlib import Path
 from typing import Optional
-from urllib.parse import urlparse
+from urllib.parse import quote, urlparse
 
 import fsspec
 import fsspec.utils
 
 logger = logging.getLogger(__name__)
 
+_AIUB_DOWNLOAD_ROOT = "https://www.aiub.unibe.ch/download"
+_AIUB_LISTING_ENDPOINT = (
+    "https://code.aiub.unibe.ch/s3_script/aiub_s3_bucket_listing.php?path="
+)
+
 
 class ConnectionPool:
     """Thread-safe pool of fsspec filesystem instances for a single host.
@@ -249,6 +255,17 @@ def list_directory(self, hostname: str, directory: str) -> list[str]:
         full_path = pool.full_path(directory)
 
         def _ls(conn: "fsspec.AbstractFileSystem") -> list[str]:
+            if hostname.rstrip("/") == _AIUB_DOWNLOAD_ROOT:
+                listing_url = _AIUB_LISTING_ENDPOINT + quote(directory.strip("/"), safe="/")
+                html = conn.cat_file(listing_url)
+                if isinstance(html, bytes):
+                    html = html.decode("utf-8", errors="replace")
+                download_prefix = re.escape(
+                    f'{_AIUB_DOWNLOAD_ROOT}/{directory.strip("/")}/'
+                )
+                return sorted(
+                    set(re.findall(rf'href="{download_prefix}([^"/]+)"', html))
+                )
             raw = conn.ls(full_path, detail=False)
             return [Path(p).name for p in raw]
 

packages/gnss-product-management/src/gnss_product_management/factories/pipelines/download.py

@@ -61,6 +61,7 @@ def run(
         date: datetime.datetime,
         *,
         sink_id: str = "local_config",
+        force: bool = False,
     ) -> Path | None | list[Path | None]:
         """Download found resources to the workspace.
 
@@ -79,7 +80,7 @@ def run(
 
         paths: list[Path | None] = []
         for r in resources:
-            path = self._download_one(r, date, sink_id)
+            path = self._download_one(r, date, sink_id, force=force)
             paths.append(path)
 
         if single:
@@ -91,6 +92,8 @@ def _download_one(
         resource: FoundResource,
         date: datetime.datetime,
         sink_id: str,
+        *,
+        force: bool = False,
     ) -> Path | None:
         """Download a single resource and write its sidecar lockfile.
 
@@ -102,7 +105,7 @@ def _download_one(
         Returns:
             Path to the resolved file, or ``None`` on failure.
         """
-        if resource.is_local:
+        if resource.is_local and not force:
             local_path = resource.path
             if local_path and local_path.exists():
                 logger.debug("Already local: %s", local_path)
@@ -121,9 +124,10 @@ def _download_one(
             local_resource_id=sink_id,
             local_factory=self._planner._workspace,
             date=date,
+            force=force,
         )
         if path is not None:
-            if get_lock_product(path) is None:
+            if force or get_lock_product(path) is None:
                 lock = build_lock_product(sink=path, url=resource.uri, name=resource.product)
                 write_lock_product(lock)
             logger.info("Downloaded %s → %s", resource.product, path)

packages/gnss-product-management/src/gnss_product_management/factories/pipelines/resolve.py

@@ -94,6 +94,7 @@ def run(
         sink_id: str = "local_config",
         centers: list[str] | None = None,
         download: bool = True,
+        force_download: bool = False,
     ) -> tuple[DependencyResolution, AnyPath | None]:
         """Resolve all dependencies in *spec* for *date*.
 
@@ -126,7 +127,7 @@ def run(
             date=date,
             version=version,
         )
-        if existing is not None:
+        if existing is not None and not force_download:
             resolution = self._resolution_from_lockfile(existing, spec)
             if resolution.all_required_fulfilled:
                 logger.info(
@@ -151,6 +152,7 @@ def run(
             preferences=spec.preferences,
             centers=centers,
             download=download,
+            force_download=force_download,
         )
         with ThreadPoolExecutor(max_workers=15) as executor:
             resolved = list(executor.map(resolve_one, spec.dependencies))
@@ -176,6 +178,7 @@ def _resolve_one(
         preferences: list[SearchPreference],
         centers: list[str] | None,
         download: bool,
+        force_download: bool,
     ) -> ResolvedDependency:
         """Resolve a single dependency.
 
@@ -201,7 +204,10 @@ def _resolve_one(
             if centers:
                 q = q.sources(*centers)
             candidates = q.search()
-            found: FoundResource | None = candidates[0] if candidates else None
+            if force_download:
+                found = next((candidate for candidate in candidates if not candidate.is_local), None)
+            else:
+                found = candidates[0] if candidates else None
         except Exception as exc:
             logger.debug("No candidates for %s: %s", dep.spec, exc)
             return ResolvedDependency(spec=dep.spec, required=dep.required, status="missing")
@@ -227,7 +233,7 @@ def _resolve_one(
                 remote_url=found.uri,
             )
 
-        path = self._downloader.run(found, date, sink_id=sink_id)
+        path = self._downloader.run(found, date, sink_id=sink_id, force=force_download)
         if path is None:
             logger.warning("Download failed for dependency %s", dep.spec)
             return ResolvedDependency(spec=dep.spec, required=dep.required, status="missing")

packages/gnss-product-management/src/gnss_product_management/factories/remote_transport.py

@@ -299,6 +299,8 @@ def download_one(
         local_resource_id: str,
         local_factory: WorkSpace,
         date: datetime.datetime,
+        *,
+        force: bool = False,
     ) -> AnyPath | None:
         """Synchronously download matched files for one search target.
 
@@ -329,10 +331,13 @@ def download_one(
         destination_dir.mkdir(parents=True, exist_ok=True)
         destination_path = destination_dir / query.product.filename.value  # type: ignore[union-attr]
 
-        # Prefer an already-decompressed version on disk
+        # Prefer an already-decompressed version on disk unless the caller
+        # explicitly requested a fresh copy from the remote source.
         if destination_path.suffix == ".gz":
             decompressed_path = destination_path.with_suffix("")
             if (
+                not force
+                and
                 decompressed_path.exists()
                 and decompressed_path.stat().st_size > 0
                 and self._validate_cached_or_evict(decompressed_path)
@@ -348,6 +353,8 @@ def download_one(
         # describes the file as served, so it only applies to the
         # non-decompressed destination path.
         if (
+            not force
+            and
             destination_path.exists()
             and destination_path.stat().st_size > 0
             and self._validate_cached_or_evict(destination_path, query.checksum)

packages/gnss-product-management/test/test_code_https.py

@@ -0,0 +1,37 @@
+"""Regression tests for CODE's S3-backed HTTPS product archive."""
+
+from unittest.mock import MagicMock
+
+from gnss_product_management.client.product_query import _remote_uri
+from gnss_product_management.factories.connection_pool import ConnectionPoolFactory
+
+
+def test_code_https_listing_extracts_download_filenames(monkeypatch) -> None:
+    html = b"""
+    <a href="https://www.aiub.unibe.ch/download/CODE/COD0OPSRAP_20262390000_01D_05M_ORB.SP3">
+      COD0OPSRAP_20262390000_01D_05M_ORB.SP3
+    </a>
+    """
+    factory = ConnectionPoolFactory(max_connections=1)
+    hostname = "https://www.aiub.unibe.ch/download"
+    factory.add_connection(hostname)
+    pool = factory._pools[hostname]
+    connection = MagicMock()
+    connection.cat_file.return_value = html
+    monkeypatch.setattr(pool, "_connect", lambda: connection)
+
+    assert factory.list_directory(hostname, "CODE/") == [
+        "COD0OPSRAP_20262390000_01D_05M_ORB.SP3"
+    ]
+
+
+def test_remote_uri_keeps_existing_https_scheme() -> None:
+    assert _remote_uri(
+        "https",
+        "https://www.aiub.unibe.ch/download",
+        "CODE/",
+        "COD0OPSRAP_20262390000_01D_05M_ORB.SP3",
+    ) == (
+        "https://www.aiub.unibe.ch/download/CODE/"
+        "COD0OPSRAP_20262390000_01D_05M_ORB.SP3"
+    )

packages/gnss-product-management/test/test_download_integrity.py

@@ -96,13 +96,20 @@ def env(tmp_path: Path):
     }
 
 
-def _download(env, filename: str = "TEST.SP3", checksum: str | None = None) -> Path | None:
+def _download(
+    env,
+    filename: str = "TEST.SP3",
+    checksum: str | None = None,
+    *,
+    force: bool = False,
+) -> Path | None:
     query = _make_query(env["remote_root"], filename, checksum)
     return env["wormhole"].download_one(
         query=query,
         local_resource_id="local_config",
         local_factory=env["workspace"],
         date=TEST_DATE,
+        force=force,
     )
 
 
@@ -191,6 +198,16 @@ def test_cache_without_sidecar_is_trusted(self, env, monkeypatch) -> None:
         assert _download(env) == cached
         assert calls == []
 
+    def test_force_redownload_replaces_valid_cache(self, env) -> None:
+        cached = env["sink_dir"] / "TEST.SP3"
+        cached.parent.mkdir(parents=True)
+        cached.write_bytes(b"valid but stale product")
+
+        result = _download(env, force=True)
+
+        assert result == cached
+        assert result.read_bytes() == REMOTE_CONTENT
+
     def test_corrupt_cache_is_evicted_and_redownloaded(self, env) -> None:
         """A cached file whose hash no longer matches its sidecar must be
         evicted (with its stale sidecar) and fetched fresh."""