HDXDSYS-2660 Add facility to allow writing resources with headers and no data (#105)

* Add no_empty parameter to generate_resource

Author: mcarans

pyproject.toml

@@ -38,7 +38,7 @@ dependencies = [
     "defopt>=7.0.0",
     "email_validator",
     "hdx-python-country>=4.0.1",
-    "hdx-python-utilities>=4.0.2",
+    "hdx-python-utilities>=4.0.3",
     "libhxl>=5.2.2",
     "makefun",
     "quantulum3",

requirements.txt

@@ -36,7 +36,7 @@ frictionless==5.18.1
     # via hdx-python-utilities
 hdx-python-country==4.0.1
     # via hdx-python-api (pyproject.toml)
-hdx-python-utilities==4.0.2
+hdx-python-utilities==4.0.3
     # via
     #   hdx-python-api (pyproject.toml)
     #   hdx-python-country
@@ -86,6 +86,8 @@ num2words==0.5.14
     # via quantulum3
 openpyxl==3.1.5
     # via hdx-python-utilities
+packaging==26.0
+    # via wheel
 petl==1.7.17
     # via frictionless
 ply==3.11
@@ -144,7 +146,7 @@ rpds-py==0.30.0
     #   referencing
 ruamel-yaml==0.19.1
     # via hdx-python-utilities
-setuptools==80.9.0
+setuptools==80.10.1
     # via ckanapi
 shellingham==1.5.4
     # via typer
@@ -194,7 +196,7 @@ urllib3==2.6.3
     #   requests
 validators==0.35.0
     # via frictionless
-wheel==0.45.1
+wheel==0.46.3
     # via libhxl
 xlrd==2.0.2
     # via hdx-python-utilities

src/hdx/data/dataset.py

@@ -2410,6 +2410,7 @@ def generate_resource(
         datecol: int | str | None = None,
         yearcol: int | str | None = None,
         date_function: Callable[[dict], dict | None] | None = None,
+        no_empty: bool = True,
     ) -> tuple[bool, dict]:
         """Write rows to file and create resource, adding it to the dataset. The headers
         argument is either a row number (rows start counting at 1), or the actual
@@ -2440,9 +2441,10 @@ def generate_resource(
             columns: Columns to write. Defaults to all.
             format: Format to write. Defaults to csv.
             encoding: Encoding to use. Defaults to None (infer encoding).
-            datecol: Optional[Union[int, str]] = None,
-            yearcol: Optional[Union[int, str]] = None,
-            date_function: Optional[Callable[[Dict], Optional[Dict]]] = None,
+            datecol: Date column for setting time period. Defaults to None (don't set).
+            yearcol: Year column for setting dataset year range. Defaults to None (don't set).
+            date_function: Date function to call for each row. Defaults to None.
+            no_empty: Don't generate resource if there are no data rows. Defaults to True.
 
         Returns:
             (True if resource added, dictionary of results)
@@ -2504,6 +2506,7 @@ def process_row(row: Sequence | Mapping) -> Sequence | Mapping | None:
             format=format,
             encoding=encoding,
             row_function=process_row,
+            no_empty=no_empty,
         )
         if not rows:
             logger.error(f"No data rows in {filename}!")
@@ -2731,6 +2734,7 @@ def download_generate_resource(
         datecol: int | str | None = None,
         yearcol: int | str | None = None,
         date_function: Callable[[dict], dict | None] | None = None,
+        no_empty: bool = True,
         **kwargs: Any,
     ) -> tuple[bool, dict]:
         """Download url, write rows to csv and create resource, adding to it
@@ -2769,9 +2773,10 @@ def download_generate_resource(
             columns: Columns to write. Defaults to all.
             format: Format to write. Defaults to csv.
             encoding: Encoding to use. Defaults to None (infer encoding).
-            datecol: Optional[Union[int, str]] = None,
-            yearcol: Optional[Union[int, str]] = None,
-            date_function: Optional[Callable[[Dict], Optional[Dict]]] = None,
+            datecol: Date column for setting time period. Defaults to None (don't set).
+            yearcol: Year column for setting dataset year range. Defaults to None (don't set).
+            date_function: Date function to call for each row. Defaults to None.
+            no_empty: Don't generate resource if there are no data rows. Defaults to True.
             **kwargs: Any additional args to pass to downloader.get_tabular_rows
 
         Returns:
@@ -2797,6 +2802,7 @@ def download_generate_resource(
             datecol=datecol,
             yearcol=yearcol,
             date_function=date_function,
+            no_empty=no_empty,
         )
 
     def download_and_generate_resource(

tests/hdx/data/test_dataset_resource_generation.py

@@ -500,6 +500,23 @@ def process_year(row):
                 )
                 assert success is False
                 url = "https://raw.githubusercontent.com/OCHA-DAP/hdx-python-api/main/tests/fixtures/gen_resource/test_data_no_data.csv"
+                success, results = dataset.download_generate_resource(
+                    downloader,
+                    url,
+                    folder,
+                    filename,
+                    resourcedata,
+                )
+                assert success is False
+                success, results = dataset.download_generate_resource(
+                    downloader,
+                    url,
+                    folder,
+                    filename,
+                    resourcedata,
+                    no_empty=False,
+                )
+                assert success is True
                 success, results = dataset.download_generate_resource(
                     downloader,
                     url,