Add driver for GeoDataFrame

.github/workflows/refresh-cache.yml

@@ -52,7 +52,7 @@ jobs:
           path: |
             /usr/share/miniconda3
             ~/pycache
-          key: test-py${{ matrix.python-version }}-${{ matrix.os }}-${{ hashFiles('environment.yml')  }}
+          key: test-${{ matrix.os }}-py${{ matrix.python-version }}-${{ hashFiles('environment.yml')  }}
         id: test-cache
 
   rebuild-docs-cache:

.github/workflows/tests.yml

@@ -2,6 +2,12 @@ name: Tests
 
 on:
   workflow_dispatch:
+    inputs:
+      debug_enabled:
+        type: boolean
+        description: 'Run the build with tmate debugging enabled (https://github.com/marketplace/actions/debugging-with-tmate)'
+        required: false
+        default: false
   push:
     branches: [main, v1]
     paths:
@@ -30,7 +36,7 @@ jobs:
       matrix:
         os: [ubuntu-latest, windows-latest]
         python-version: ['3.9','3.10','3.11']
-    name: py ${{ matrix.python-version }}
+    name: py ${{ matrix.python-version }} (${{ matrix.os}})
     runs-on: ${{ matrix.os }}
     timeout-minutes: 30
     concurrency:
@@ -53,9 +59,16 @@ jobs:
             ~/pycache
           # the below two settings mean we'll alway srestore the cache
           # but the cache hit output can tell us if we have to update afterwards
-          key: test-py${{ matrix.python-version }}-${{ hashFiles('environment.yml')  }}
+          key: test-${{ matrix.os }}-py${{ matrix.python-version }}-${{ hashFiles('environment.yml')  }}
           restore-keys: |
-            test-py${{ matrix.python-version }}
+            test-${{ matrix.os }}-py${{ matrix.python-version }}-${{ hashFiles('environment.yml')  }}
+            test-${{ matrix.os }}-py${{ matrix.python-version }}
+            test-${{ matrix.os }}
+
+      # Enable tmate debugging of manually-triggered workflows if the input option was provided
+      - name: Setup tmate session
+        uses: mxschmitt/action-tmate@v3
+        if: ${{ github.event_name == 'workflow_dispatch' && inputs.debug_enabled }}
 
       - name: Fail on no cache restore
         if: steps.cache.outputs.cache-matched-key == ''

docs/changelog.rst

@@ -6,6 +6,14 @@ All notable changes to this project will be documented in this page.
 The format is based on `Keep a Changelog`_, and this project adheres to
 `Semantic Versioning`_.
 
+V1
+==
+
+Added
+-----
+- Added Driver class for customizable io
+- Added MetaDataResolver class for customizable metadata discovery
+- Added DataSource class to represent and validate DataCatalog entries.
 
 
 Unreleased

hydromt/_typing/__init__.py

@@ -11,12 +11,14 @@
     Bbox,
     Crs,
     Data,
+    DataType,
     DeferedFileClose,
     ExportConfigDict,
     GeoDataframeSource,
     GeoDatasetSource,
     Geom,
     GeomBuffer,
+    GpdShapeGeom,
     ModeLike,
     Number,
     Predicate,
@@ -51,7 +53,9 @@
     "_exec_nodata_strat",
     "Variables",
     "Geom",
+    "GpdShapeGeom",
     "Data",
+    "DataType",
     "GeomBuffer",
     "Predicate",
 ]

hydromt/_typing/type_def.py

@@ -3,11 +3,54 @@
 from pathlib import Path
 from typing import Any, Dict, List, Literal, Tuple, TypedDict, Union
 
-from geopandas import GeoDataFrame, GeoSeries
+import geopandas as gpd
+from dateutil.parser import parse
+from pydantic.functional_validators import AfterValidator, BeforeValidator
+from shapely.geometry.base import BaseGeometry
+from typing_extensions import Annotated
 from xarray import DataArray, Dataset
 
 from hydromt._typing.model_mode import ModelMode
 
+
+def _time_tuple_from_str(
+    t: Tuple[Union[str, datetime], Union[str, datetime]],
+) -> "TimeRange":
+    if isinstance(t[0], str):
+        t0 = parse(t[0])
+    else:
+        t0 = t[0]
+    if isinstance(t[1], str):
+        t1 = parse(t[1])
+    else:
+        t1 = t[1]
+    return (t0, t1)
+
+
+def _timerange_validate(tr: tuple[datetime, datetime]) -> tuple[datetime, datetime]:
+    assert tr[0] >= tr[1], f"timerange t0: '{tr[0]}' should be less than t1: '{tr[1]}'"
+    return tr
+
+
+def _validate_bbox(
+    bbox: tuple[float, float, float, float],
+) -> tuple[float, float, float, float]:
+    assert (
+        bbox[0] < bbox[2]
+    ), f"bbox minx: '{bbox[0]}' should be less than maxx: '{bbox[2]}'."
+    assert (
+        bbox[1] < bbox[3]
+    ), f"bbox miny: '{bbox[1]}' should be less than maxy: '{bbox[3]}'."
+
+
+DataType = Literal[
+    "DataFrame", "DataSet", "GeoDataFrame", "GeoDataSet", "RasterDataSet"
+]
+GeoDataframeSource = Union[str, Path]
+GeoDatasetSource = Union[str, Path]
+RasterDatasetSource = Union[str, Path]
+Bbox = Annotated[Tuple[float, float, float, float], _validate_bbox]
+
 StrPath = Union[str, Path]
 GeoDataframeSource = StrPath
 GeoDatasetSource = StrPath
@@ -17,7 +60,11 @@
 Bbox = Tuple[float, float, float, float]
 Crs = int
 TotalBounds = Tuple[Bbox, Crs]
-TimeRange = Tuple[datetime, datetime]
+TimeRange = Annotated[
+    Tuple[datetime, datetime],
+    BeforeValidator(_time_tuple_from_str),
+    AfterValidator(_timerange_validate),
+]
 Number = Union[int, float]
 SourceSpecDict = TypedDict(
     "SourceSpecDict", {"source": str, "provider": str, "version": Union[str, int]}
@@ -38,13 +85,13 @@
     "intersects", "within", "contains", "overlaps", "crosses", "touches"
 ]
 
-Geom = Union[GeoDataFrame, GeoSeries]
+Geom = Union[gpd.GeoDataFrame, gpd.GeoSeries]
+GpdShapeGeom = Union[Geom, BaseGeometry]
 
 Data = Union[Dataset, DataArray]
 
 Variables = Union[str, List[str]]
 
 GeomBuffer = int
 
-
 ModeLike = Union[ModelMode, str]

hydromt/data_adapter/__init__.py

@@ -1,6 +1,6 @@
-# -*- coding: utf-8 -*-
-"""HydroMT data adapter."""
+"""Data Adapters are generic for its HydroMT type and perform transformations."""
 
+# TODO: correct imports after deprecation of old adapters
 from .caching import cache_vrt_tiles
 from .data_adapter import PREPROCESSORS, DataAdapter
 from .dataframe import DataFrameAdapter

hydromt/data_adapter/caching.py

@@ -20,6 +20,7 @@
 HYDROMT_DATADIR = join(Path.home(), ".hydromt_data")
 
 
+# TODO: move this to an appropriate module (has nothing to do with caching)
 def _uri_validator(uri: Union[str, Path]) -> bool:
     """Check if uri is valid."""
     try:

hydromt/data_adapter/data_adapter_base.py

@@ -0,0 +1,10 @@
+"""Base class for data adapters."""
+from pydantic import BaseModel
+
+from hydromt.data_sources.data_source import DataSource
+
+
+class DataAdapterBase(BaseModel):
+    """Base class for data adapters."""
+
+    source: DataSource