Merge pull request #421 from OpenDataServices/419-wkt

create_template: Create a template suitable for WKT data

CHANGELOG.md

@@ -6,6 +6,12 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 
 ## [Unreleased]
 
+## [0.22.0] - 2023-06-27
+
+### Added
+
+- Generate templates that are correct for WKT <-> geojson conversion
+
 ## [0.21.0] - 2023-06-23
 
 ### Added

examples/help/create-template/expected.txt

@@ -5,6 +5,7 @@ usage: flatten-tool create-template [-h] -s SCHEMA [-f {csv,ods,xlsx,all}]
                                     [--no-deprecated-fields]
                                     [--truncation-length TRUNCATION_LENGTH]
                                     [--line-terminator LINE_TERMINATOR]
+                                    [--convert-wkt]
 
 optional arguments:
   -h, --help            show this help message and exit
@@ -36,3 +37,4 @@ optional arguments:
   --line-terminator LINE_TERMINATOR
                         The line terminator to use when writing CSV files:
                         CRLF or LF
+  --convert-wkt         Enable conversion of WKT to geojson

examples/wkt/create_template/cmd.txt

@@ -0,0 +1 @@
+$ flatten-tool create-template --convert-wkt --schema examples/wkt/create_template/schema.json -o examples/wkt/create_template/actual

examples/wkt/create_template/expected/main.csv

@@ -0,0 +1 @@
+geo

examples/wkt/create_template/schema.json

@@ -0,0 +1,11 @@
+{
+    "properties": {
+        "geo": {
+            "type": "object",
+            "properties": {
+                "type": {},
+                "coordinates": {}
+            }
+        }
+    }
+}

flattentool/__init__.py

@@ -25,6 +25,7 @@ def create_template(
     truncation_length=3,
     no_deprecated_fields=False,
     line_terminator="CRLF",
+    convert_wkt=False,
     **_,
 ):
     """
@@ -37,6 +38,8 @@ def create_template(
     if line_terminator not in LINE_TERMINATORS.keys():
         raise Exception(f"{line_terminator} is not a valid line terminator")
 
+    convert_flags = {"wkt": convert_wkt}
+
     parser = SchemaParser(
         schema_filename=schema,
         rollup=rollup,
@@ -45,6 +48,7 @@ def create_template(
         disable_local_refs=disable_local_refs,
         truncation_length=truncation_length,
         exclude_deprecated_fields=no_deprecated_fields,
+        convert_flags=convert_flags,
     )
     parser.parse()
 
@@ -121,6 +125,7 @@ def flatten(
             use_titles=use_titles,
             disable_local_refs=disable_local_refs,
             truncation_length=truncation_length,
+            convert_flags=convert_flags,
         )
         schema_parser.parse()
     else:
@@ -270,7 +275,9 @@ def unflatten(
         )
         if metatab_schema:
             parser = SchemaParser(
-                schema_filename=metatab_schema, disable_local_refs=disable_local_refs
+                schema_filename=metatab_schema,
+                disable_local_refs=disable_local_refs,
+                convert_flags=convert_flags,
             )
             parser.parse()
             spreadsheet_input.parser = parser

flattentool/cli.py

@@ -100,6 +100,11 @@ def create_parser():
         "--line-terminator",
         help="The line terminator to use when writing CSV files: CRLF or LF",
     )
+    parser_create_template.add_argument(
+        "--convert-wkt",
+        action="store_true",
+        help="Enable conversion of WKT to geojson",
+    )
 
     parser_flatten = subparsers.add_parser("flatten", help="Flatten a JSON file")
     parser_flatten.add_argument("input_name", help="Name of the input JSON file.")

flattentool/schema.py

@@ -225,12 +225,6 @@ def parse_schema_dict(
                         yield (field, child_title)
 
         elif "properties" in schema_dict:
-            if (
-                self.convert_flags.get("wkt")
-                and "type" in schema_dict["properties"]
-                and "coordinates" in schema_dict["properties"]
-            ):
-                self.flattened[parent_path.replace("/0/", "/").strip("/")] = "geojson"
             if "id" in schema_dict["properties"]:
                 if self.use_titles:
                     id_fields = parent_id_fields + [
@@ -271,6 +265,16 @@ def parse_schema_dict(
                     title_lookup[title].property_name = property_name
 
                 if "object" in property_type_set:
+                    if (
+                        self.convert_flags.get("wkt")
+                        and "type" in property_schema_dict.get("properties", {})
+                        and "coordinates" in property_schema_dict.get("properties", {})
+                    ):
+                        self.flattened[
+                            parent_path.replace("/0/", "/") + property_name
+                        ] = "geojson"
+                        yield (property_name, title)
+                        continue
                     self.flattened[parent_path + property_name] = "object"
                     for field, child_title in self.parse_schema_dict(
                         parent_path + property_name,

flattentool/tests/test_docs.py

@@ -136,7 +136,7 @@ def test_example_in_doc(root, filename):
 
 
 def test_expected_number_of_examples_in_docs_data():
-    expected = 66
+    expected = 67
     # See _get_examples_in_docs_data()
     if sys.version_info[:2] != (3, 8):
         expected -= 3

setup.py

@@ -45,7 +45,7 @@ def run(self):
 
 setup(
     name="flattentool",
-    version="0.21.0",
+    version="0.22.0",
     author="Open Data Services",
     author_email="[email protected]",
     packages=["flattentool"],