DM-52947: Add a ResourcePath.get_info() method

.github/workflows/build.yaml

@@ -68,9 +68,26 @@ jobs:
         run: |
           uv pip install --system --no-deps -v -e .
 
+      - name: Start fake-gcs-server
+        run: |
+          docker run -d --name fake-gcs-server -p 4443:4443 \
+            fsouza/fake-gcs-server -scheme http -filesystem-root /tmp/fake-gcs-server
+
+          for i in $(seq 1 30); do
+            if curl --silent --fail http://127.0.0.1:4443/storage/v1/b >/dev/null; then
+              exit 0
+            fi
+            sleep 1
+          done
+
+          docker logs fake-gcs-server
+          exit 1
+
       - name: Run tests
         env:
           S3_ENDPOINT_URL: "https://google.com"
+          STORAGE_EMULATOR_HOST: "http://127.0.0.1:4443"
+          GOOGLE_CLOUD_PROJECT: "test-project"
         run: |
           pytest -r a -v -n 3 --cov=lsst.resources\
                  --cov=tests --cov-report=xml --cov-report=term --cov-branch \

.gitignore

@@ -3,6 +3,7 @@
 *.bck
 *.pyc
 .eggs
+.env
 *.egg-info
 version.py
 _build.*

doc/changes/DM-52947.feature.rst

@@ -0,0 +1 @@
+Added a ``ResourcePath.get_info()`` method to provide a general interface for obtaining information about a resource including the size, modification date, and any checksums available.

pyproject.toml

@@ -174,6 +174,7 @@ select = [
 ]
 extend-select = [
     "RUF100", # Warn about unused noqa
+    "D212",  # Docstring starts without newline after quotes.
 ]
 
 [tool.ruff.lint.isort]

python/lsst/resources/__init__.py

@@ -14,6 +14,7 @@
 
 __all__ = (
     "ResourceHandleProtocol",
+    "ResourceInfo",
     "ResourcePath",
     "ResourcePathExpression",
 )
@@ -22,5 +23,5 @@
 from ._resourceHandles import ResourceHandleProtocol
 
 # Should only expose ResourcePath and its input type alias
-from ._resourcePath import ResourcePath, ResourcePathExpression
+from ._resourcePath import ResourceInfo, ResourcePath, ResourcePathExpression
 from .version import *

python/lsst/resources/_resourceHandles/_davResourceHandle.py

@@ -197,8 +197,8 @@ class DavReadAheadCache:
     Parameters
     ----------
     client : `lsst.resources.davutils.DavClient`
-        webDAV client to interact with the server to download data.
-    backend_url : `str`
+        The webDAV client to interact with the server to download data.
+    url : `str`
         URL of the resource to download data from.
     filesize : `int`
         Size in bytes of the remote file.

python/lsst/resources/_resourcePath.py

@@ -11,11 +11,13 @@
 
 from __future__ import annotations
 
-__all__ = ("ResourcePath", "ResourcePathExpression")
+__all__ = ("ResourceInfo", "ResourcePath", "ResourcePathExpression")
 
 import concurrent.futures
 import contextlib
 import copy
+import dataclasses
+import datetime
 import io
 import locale
 import logging
@@ -131,6 +133,27 @@ def _patch_environ(new_values: dict[str, str]) -> Iterator[None]:
                 os.environ[k] = old_values[k]
 
 
+@dataclasses.dataclass(frozen=True)
+class ResourceInfo:
+    """Information about this resource."""
+
+    uri: str
+    """URI in string form of the resource from which this information is
+    derived.
+    """
+    is_file: bool
+    """Indicate whether the resource is a file or a directory."""
+    size: int
+    """Size of the file in bytes. A directory or a URI that has no concept
+    of size returns 0."""
+    last_modified: datetime.datetime | None
+    """Modification date of the resource, if known."""
+    checksums: dict[str, Any]
+    """Checksums for this file. Supported checksum implementations are
+    backend dependent.
+    """
+
+
 class ResourcePath:  # numpydoc ignore=PR02
     """Convenience wrapper around URI parsers.
 
@@ -1931,6 +1954,17 @@ def _copy_extra_attributes(self, original_uri: ResourcePath) -> None:
         # ResourcePath constructor by passing in a ResourcePath object.
         pass
 
+    def get_info(self) -> ResourceInfo:
+        """Return lightweight metadata about this resource.
+
+        Returns
+        -------
+        info : `ResourceInfo`
+            The information about this resource that can be obtained from
+            the backend. Will not read the file contents.
+        """
+        raise NotImplementedError("")
+
 
 ResourcePathExpression = str | urllib.parse.ParseResult | ResourcePath | Path
 """Type-annotation alias for objects that can be coerced to ResourcePath.

python/lsst/resources/dav.py

@@ -39,7 +39,7 @@
 
 from ._resourceHandles import ResourceHandleProtocol
 from ._resourceHandles._davResourceHandle import DavReadResourceHandle
-from ._resourcePath import ResourcePath, ResourcePathExpression
+from ._resourcePath import ResourceInfo, ResourcePath, ResourcePathExpression
 from .davutils import (
     DavClient,
     DavClientPool,
@@ -139,8 +139,7 @@ def __init__(self) -> None:
         self._reset()
 
     def _reset(self) -> None:
-        """
-        Initialize all the globals.
+        """Initialize all the globals.
 
         This method is a helper for reinitializing globals in tests.
         """
@@ -289,11 +288,22 @@ def size(self) -> int:
 
         return 0 if self.isdir() else self._client.size(self._internal_url)
 
-    def info(self) -> dict[str, Any]:
-        """Return metadata details about this resource."""
-        log.debug("info %s [%#x]", self, id(self))
+    @override
+    def get_info(self) -> ResourceInfo:
+        """Return lightweight metadata details about this resource."""
+        log.debug("get_info %s [%#x]", self, id(self))
+
+        info = self._client.info(self._internal_url)
+        if info["type"] is None:
+            raise FileNotFoundError(f"Resource {self} does not exist")
 
-        return self._client.info(self._internal_url, name=str(self))
+        return ResourceInfo(
+            uri=str(self),
+            is_file=info["type"] == "file",
+            size=info["size"],
+            last_modified=info["last_modified"],
+            checksums=info["checksums"],
+        )
 
     @override
     def read(self, size: int = -1) -> bytes: