Issue #108 - Added some behave tests for the CLI.

csvqb/csvqb/tests/behaviour/cli.feature

@@ -0,0 +1,41 @@
+Feature: Test the csvqb Command Line Interface.
+
+  Scenario: The csvqb build command should take an info.json and CSV and output CSV-Ws into the default './out' directory.
+    Given the existing test-case file "configloaders/data.csv"
+    And the existing test-case file "configloaders/info.json"
+    When the csvqb CLI is run with "build --config configloaders/info.json configloaders/data.csv"
+    Then the csvqb CLI should succeed
+    And the csvqb CLI should print "Creating output directory"
+    And the file at "out/ons-international-trade-in-services-by-subnational-areas-of-the-uk.csv" should exist
+    And the file at "out/ons-international-trade-in-services-by-subnational-areas-of-the-uk.csv-metadata.json" should exist
+    And the file at "out/validation-errors.json" should not exist
+
+  Scenario: The csvqb build command should not output CSV-Ws when validation errors occur
+    Given the existing test-case file "configloaders/validation-error/data.csv"
+    And the existing test-case file "configloaders/validation-error/info.json"
+    When the csvqb CLI is run with "build --config configloaders/validation-error/info.json configloaders/validation-error/data.csv"
+    Then the csvqb CLI should fail with status code 1
+    And the csvqb CLI should print "Validation Error"
+    And the csvqb CLI should print "Column 'Measure Type' not found in data provided"
+    And the file at "out/validation-error-output.csv" should not exist
+    And the file at "out/validation-error-output.csv-metadata.json" should not exist
+    And the file at "out/validation-errors.json" should not exist
+
+  Scenario: The csvqb build command should output validation errors to file
+    Given the existing test-case file "configloaders/validation-error/data.csv"
+    And the existing test-case file "configloaders/validation-error/info.json"
+    When the csvqb CLI is run with "build --validation-errors-to-file --config configloaders/validation-error/info.json configloaders/validation-error/data.csv"
+    Then the csvqb CLI should fail with status code 1
+    And the file at "out/validation-errors.json" should exist
+    And the validation-errors.json file in the "out" directory should contain
+    """
+      Column 'Measure Type' not found in data provided
+    """
+
+  Scenario: The csvqb build command should still output CSV-Ws if the user overrides validation errors
+    Given the existing test-case file "configloaders/validation-error/data.csv"
+    And the existing test-case file "configloaders/validation-error/info.json"
+    When the csvqb CLI is run with "build --ignore-validation-errors --config configloaders/validation-error/info.json  configloaders/validation-error/data.csv"
+    Then the csvqb CLI should succeed
+    And the file at "out/validation-error-output.csv" should exist
+    And the file at "out/validation-error-output.csv-metadata.json" should exist

csvqb/csvqb/tests/behaviour/steps/cli.py

@@ -0,0 +1,59 @@
+from behave import when, then
+import subprocess
+from typing import Tuple
+from devtools.behave.temporarydirectory import get_context_temp_dir_path
+
+
+@when('the csvqb CLI is run with "{arguments}"')
+def step_impl(context, arguments: str):
+    command: str = f"csvqb {arguments.strip()}"
+    (status_code, response) = run_command_in_temp_dir(context, command)
+    context.csvqb_cli_result = (status_code, response)
+
+
+@then("the csvqb CLI should succeed")
+def step_impl(context):
+    (status_code, response) = context.csvqb_cli_result
+    assert status_code == 0, status_code
+    assert "Build Complete" in response, response
+
+
+@then("the csvqb CLI should fail with status code {status_code}")
+def step_impl(context, status_code: str):
+    (status_code, response) = context.csvqb_cli_result
+    assert status_code == int(status_code), status_code
+
+
+@then('the csvqb CLI should print "{printed_text}"')
+def step_impl(context, printed_text: str):
+    (status_code, response) = context.csvqb_cli_result
+    assert printed_text in response, response
+
+
+@then('the validation-errors.json file in the "{out_dir}" directory should contain')
+def step_impl(context, out_dir: str):
+    tmp_dir_path = get_context_temp_dir_path(context)
+    expected_text_contents: str = context.text.strip()
+    validation_errors_file = tmp_dir_path / out_dir / "validation-errors.json"
+    assert validation_errors_file.exists()
+
+    with open(validation_errors_file, "r") as f:
+        file_contents = f.read()
+
+    assert expected_text_contents in file_contents, file_contents
+
+
+def run_command_in_temp_dir(context, command: str) -> Tuple[int, str]:
+    tmp_dir_path = get_context_temp_dir_path(context)
+    process = subprocess.Popen(
+        command,
+        shell=True,
+        cwd=tmp_dir_path,
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE,
+    )
+    status_code = process.wait()
+    response = process.stdout.read().decode("utf-8") + process.stderr.read().decode(
+        "utf-8"
+    )
+    return status_code, response

csvqb/csvqb/tests/test-cases/configloaders/validation-error/data.csv

@@ -0,0 +1,2 @@
+Period,Location,Flow,Industry Grouping,Marker,Value,Undefined Column
+2020-01,West Midlands,Imports,Barbed Wire,,20,Undefined Column Value

csvqb/csvqb/tests/test-cases/configloaders/validation-error/info.json

@@ -0,0 +1,42 @@
+{
+    "id": "validation-error-output",
+    "title": "Validation Error Output",
+    "publisher": "HM Revenue & Customs",
+    "description": "All bulletins provide details on percentage of one litre or less & more than one litre bottles. This information is provided on a yearly basis.",
+    "landingPage": "https://www.gov.uk/government/statistics/bottles-bulletin",
+    "datasetNotes": [
+        "\"UK bottles-bulletin Tables\" Excel file, latest version on page"
+    ],
+    "published": "2019-02-28",
+    "families": [
+        "Trade"
+    ],
+    "extract": {
+        "source": "XLS",
+        "stage": "Done"
+    },
+    "transform": {
+        "airtable": "recys4OhEtE0gE14P",
+        "columns": {
+            "Period": {
+                "parent": "http://purl.org/linked-data/sdmx/2009/dimension#refPeriod",
+                "value": "http://reference.data.gov.uk/id/{+period}"
+            },
+            "Measure Type": {
+                "dimension": "http://purl.org/linked-data/cube#measureType",
+                "value": "http://gss-data.org.uk/def/x/{+measure_type}",
+                "types": ["one-litre-and-less", "more-than-one-litre", "number-of-bottles"]
+            },
+            "Unit": {
+                "attribute": "http://purl.org/linked-data/sdmx/2009/attribute#unitMeasure",
+                "value": "http://gss-data.org.uk/def/concept/measurement-units/{+unit}"
+            },
+            "Value": {
+                "datatype": "integer"
+            }
+        },
+        "main_issue": 67
+    },
+    "sizingNotes": "",
+    "notes": ""
+}

devtools/devtools/behave/file.py

@@ -10,9 +10,31 @@
 from .temporarydirectory import get_context_temp_dir_path
 
 
-@Given('the existing test-case files "{file}"')
+@Given('the existing test-case file "{file}"')
 def step_impl(context, file):
     test_cases_dir = _get_test_cases_dir()
+    test_case_file = test_cases_dir / file
+
+    if not test_case_file.exists():
+        raise Exception(f"Could not find test-case file {file}")
+
+    temp_dir = get_context_temp_dir_path(context)
+    output_file_path = temp_dir.joinpath(test_case_file.relative_to(test_cases_dir))
+
+    _ensure_directory_hierarchy_exists(output_file_path.parent)
+
+    shutil.copy(test_case_file, output_file_path)
+
+
+def _ensure_directory_hierarchy_exists(directory: Path):
+    if not directory.exists():
+        _ensure_directory_hierarchy_exists(directory.parent)
+        directory.mkdir()
+
+
+@Given('the existing test-case files "{files_glob}"')
+def step_impl(context, files_glob: str):
+    test_cases_dir = _get_test_cases_dir()
 
     matching_files = list(test_cases_dir.rglob(f"**/{file}"))
     if len(matching_files) == 0:
@@ -32,7 +54,7 @@ def step_impl(context, file):
 @then('the file at "{file}" should exist')
 def step_impl(context, file):
     temp_dir = get_context_temp_dir_path(context)
-    assert (temp_dir / file).exists()
+    assert (temp_dir / file).exists(), file
 
 
 def _get_test_cases_dir(start_dir: Path = Path(".")):