Merge pull request #24 from davemfish/exp/data-package

Major refactor to use dataclasses that loosely follow DataPackage - Resource models

README.md

@@ -1,4 +1,10 @@
-A Python library for creating [Metadata Control Files](https://geopython.github.io/pygeometa/reference/mcf/)
+A Python library for creating human and machine-readable metadata for geospatial data.
+
+Supported datatypes include:
+* everything supported by GDAL
+* tabular formats supported by `frictionless`
+* compressed formats supported by `frictionless`
+
 
 See `requirements.txt` for dependencies
 
@@ -7,48 +13,50 @@ See `requirements.txt` for dependencies
 #### Creating & adding metadata to file:
 
 ```python
-from geometamaker import MetadataControl
+import geometamaker
 
 data_path = 'data/watershed_gura.shp'
-mc = MetadataControl(data_path)
+resource = geometamaker.describe(data_path)
 
-mc.set_title('My Dataset')
-mc.set_abstract('all about my dataset')
-mc.set_keywords(['hydrology', 'watersheds'])
+resource.set_title('My Dataset')
+resource.set_description('all about my dataset')
+resource.set_keywords(['hydrology', 'watersheds'])
 
 # For a vector:
-mc.set_field_description(
+resource.set_field_description(
     'field_name',  # the name of an actual field in the vector's table
-    abstract='something about the field',
+    description='something about the field',
     units='mm')
 
 # or for a raster:
-mc.set_band_description(
+data_path = 'data/dem.tif'
+resource = geometamaker.describe(data_path)
+resource.set_band_description(
     1,  # a raster band index, starting at 1
     name='band name',
-    abstract='something about the band',
+    description='something about the band',
     units='mm')
 
 
-mc.validate()
-mc.write()
+resource.write()
 ```
 
 #### Creating metadata for a batch of files:
 ```python
 import os
 
-from geometamaker import MetadataControl
+import geometamaker
 
 data_dir = 'C:/Users/dmf/projects/invest/data/invest-sample-data'
 for path, dirs, files in os.walk(data_dir):
     for file in files:
-        if file.endswith(('.shp', '.gpkg', '.tif')):
-            filepath = os.path.join(path, file)
-            print(filepath)
-            mc = MetadataControl(filepath)
-            mc.validate()
-            mc.write()
+        filepath = os.path.join(path, file)
+        print(filepath)
+        try:
+            resource = geometamaker.describe(filepath)
+        except ValueError as err:
+            print(err)
+        resource.write()
 ```
 
 #### For a complete list of methods:

docs/environment-rtd.yml

@@ -8,13 +8,11 @@ name: env-readthedocs
 channels:
     - conda-forge
 dependencies:
-    - python=3.8
+    - python=3.10
+    - frictionless
     - fsspec
     - gdal>=3
-    - jsonschema
     - numpy
-    - pygeometa
     - pygeoprocessing>=2.4.2
-    - shapely
     - pyyaml
     - sphinx_rtd_theme

docs/source/conf.py

@@ -3,13 +3,12 @@
 # For the full list of built-in configuration values, see the documentation:
 # https://www.sphinx-doc.org/en/master/usage/configuration.html
 
-import datetime
 import os
 import sys
 import sphinx.ext.apidoc
 from pkg_resources import get_distribution
 
-sys.path.insert(0, os.path.abspath('../../src/geometamaker'))
+sys.path.insert(0, os.path.abspath('../../src'))
 
 # -- Project information -----------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#project-information

requirements.txt

@@ -5,10 +5,8 @@
 aiohttp
 fsspec
 GDAL
-jsonschema
+frictionless
 numpy
-pygeometa
 pygeoprocessing>=2.4.3
 pyyaml
 requests
-shapely

src/geometamaker/__init__.py

@@ -1 +1,6 @@
-from .geometamaker import MetadataControl
+import importlib.metadata
+
+from .geometamaker import describe
+
+
+__version__ = importlib.metadata.version('geometamaker')