Add tests for nf-core subworkflows create-test-yml

CHANGELOG.md

@@ -29,6 +29,10 @@
 - Use `nfcore/gitpod:dev` container in the dev branch ([#2196](https://github.com/nf-core/tools/pull/2196))
 - Replace requests_mock with responses in test mocks ([#2165](https://github.com/nf-core/tools/pull/2165)).
 
+### Bug fixes, maintenance and tests
+
+- Add tests for `nf-core subworkflows create-test-yml` ([#2219](https://github.com/nf-core/tools/pull/2219))
+
 ## [v2.7.2 - Mercury Eagle Patch](https://github.com/nf-core/tools/releases/tag/2.7.2) - [2022-12-19]
 
 ### Template

README.md

@@ -612,7 +612,7 @@ The graphical interface is oganzised in groups and within the groups the single
 
 Now you can start to change the parameter itself. The `ID` of a new parameter should be defined in small letters without whitespaces. The description is a short free text explanation about the parameter, that appears if you run your pipeline with the `--help` flag. By clicking on the dictionary icon you can add a longer explanation for the parameter page of your pipeline. Usually, they contain a small paragraph about the parameter settings or a used datasource, like databases or references. If you want to specify some conditions for your parameter, like the file extension, you can use the nut icon to open the settings. This menu depends on the `type` you assigned to your parameter. For integers you can define a min and max value, and for strings the file extension can be specified.
 
-The `type` field is one of the most important points in your pipeline schema, since it defines the datatype of your input and how it will be interpreted. This allows extensive testing prior to starting the pipeline. 
+The `type` field is one of the most important points in your pipeline schema, since it defines the datatype of your input and how it will be interpreted. This allows extensive testing prior to starting the pipeline.
 
 The basic datatypes for a pipeline schema are:
 
@@ -621,7 +621,7 @@ The basic datatypes for a pipeline schema are:
 - `integer`
 - `boolean`
 
-For the `string` type you have three different options in the settings (nut icon): `enumerated values`, `pattern` and `format`. The first option, `enumerated values`, allows you to specify a list of specific input values. The list has to be separated with a pipe. The `pattern` and `format` settings can depend on each other. The `format` has to be either a directory or a file path. Depending on  the `format` setting selected, specifying the `pattern` setting can be the most efficient and time saving option, especially for `file paths`. The `number` and `integer` types share the same settings. Similarly to `string`, there is an `enumerated values` option with the possibility of specifying a `min` and `max` value. For the `boolean` there is no further settings and the default value is usually `false`. The `boolean` value can be switched to `true` by adding the flag to the command. This parameter type is often used to skip specific sections of a pipeline.
+For the `string` type you have three different options in the settings (nut icon): `enumerated values`, `pattern` and `format`. The first option, `enumerated values`, allows you to specify a list of specific input values. The list has to be separated with a pipe. The `pattern` and `format` settings can depend on each other. The `format` has to be either a directory or a file path. Depending on the `format` setting selected, specifying the `pattern` setting can be the most efficient and time saving option, especially for `file paths`. The `number` and `integer` types share the same settings. Similarly to `string`, there is an `enumerated values` option with the possibility of specifying a `min` and `max` value. For the `boolean` there is no further settings and the default value is usually `false`. The `boolean` value can be switched to `true` by adding the flag to the command. This parameter type is often used to skip specific sections of a pipeline.
 
 After filling the schema, click on the `Finished` button in the top right corner, this will automatically update your `nextflow_schema.json`. If this is not working, the schema can be copied from the graphical interface and pasted in your `nextflow_schema.json` file.
 

nf_core/subworkflow-template/tests/main.nf

@@ -4,7 +4,7 @@ nextflow.enable.dsl = 2
 
 include { {{ subworkflow_name|upper }} } from '../../../../subworkflows/{{ org }}/{{ subworkflow_dir }}/main.nf'
 
-workflow test_{{ subworkflow_name }} {
+workflow test_{{ component_name_underscore }} {
     {% if has_meta %}
     input = [
         [ id:'test' ], // meta map

nf_core/subworkflows/test_yml_builder.py

@@ -139,7 +139,7 @@ def scrape_workflow_entry_points(self):
                 if match:
                     self.entry_points.append(match.group(1))
         if len(self.entry_points) == 0:
-            raise UserWarning("No workflow entry points found in 'self.module_test_main'")
+            raise UserWarning(f"No workflow entry points found in '{self.subworkflow_test_main}'")
 
     def build_all_tests(self):
         """
@@ -195,7 +195,7 @@ def build_single_test(self, entry_point):
                     ).strip()
                     ep_test["tags"] = [t.strip() for t in prompt_tags.split(",")]
 
-        ep_test["files"] = self.get_md5_sums(entry_point, ep_test["command"])
+        ep_test["files"] = self.get_md5_sums(ep_test["command"])
 
         return ep_test
 
@@ -272,7 +272,7 @@ def create_test_file_dict(self, results_dir, is_repeat=False):
 
         return test_files
 
-    def get_md5_sums(self, entry_point, command, results_dir=None, results_dir_repeat=None):
+    def get_md5_sums(self, command, results_dir=None, results_dir_repeat=None):
         """
         Recursively go through directories and subdirectories
         and generate tuples of (<file_path>, <md5sum>)

tests/subworkflows/create_test_yml.py

@@ -0,0 +1,95 @@
+import os
+from unittest import mock
+
+import pytest
+
+import nf_core.subworkflows
+
+from ..utils import with_temporary_folder
+
+
+@with_temporary_folder
+def test_subworkflows_custom_yml_dumper(self, out_dir):
+    """Try to create a yml file with the custom yml dumper"""
+    yml_output_path = os.path.join(out_dir, "test.yml")
+    meta_builder = nf_core.subworkflows.SubworkflowTestYmlBuilder(
+        subworkflow="test/tool",
+        directory=self.pipeline_dir,
+        test_yml_output_path=yml_output_path,
+        no_prompts=True,
+    )
+    meta_builder.test_yml_output_path = yml_output_path
+    meta_builder.tests = [{"testname": "myname"}]
+    meta_builder.print_test_yml()
+    assert os.path.isfile(yml_output_path)
+
+
+@with_temporary_folder
+def test_subworkflows_test_file_dict(self, test_file_dir):
+    """Create dict of test files and create md5 sums"""
+    meta_builder = nf_core.subworkflows.SubworkflowTestYmlBuilder(
+        subworkflow="test/tool",
+        directory=self.pipeline_dir,
+        test_yml_output_path="./",
+        no_prompts=True,
+    )
+    with open(os.path.join(test_file_dir, "test_file.txt"), "w") as fh:
+        fh.write("this line is just for testing")
+    test_files = meta_builder.create_test_file_dict(test_file_dir)
+    assert len(test_files) == 1
+    assert test_files[0]["md5sum"] == "2191e06b28b5ba82378bcc0672d01786"
+
+
+@with_temporary_folder
+def test_subworkflows_create_test_yml_get_md5(self, test_file_dir):
+    """Get md5 sums from a dummy output"""
+    meta_builder = nf_core.subworkflows.SubworkflowTestYmlBuilder(
+        subworkflow="test/tool",
+        directory=self.pipeline_dir,
+        test_yml_output_path="./",
+        no_prompts=True,
+    )
+    with open(os.path.join(test_file_dir, "test_file.txt"), "w") as fh:
+        fh.write("this line is just for testing")
+    test_files = meta_builder.get_md5_sums(
+        command="dummy",
+        results_dir=test_file_dir,
+        results_dir_repeat=test_file_dir,
+    )
+    assert test_files[0]["md5sum"] == "2191e06b28b5ba82378bcc0672d01786"
+
+
+def test_subworkflows_create_test_yml_entry_points(self):
+    """Test extracting test entry points from a main.nf file"""
+    subworkflow = "test_subworkflow"
+    meta_builder = nf_core.subworkflows.SubworkflowTestYmlBuilder(
+        subworkflow=f"{subworkflow}/test",
+        directory=self.pipeline_dir,
+        test_yml_output_path="./",
+        no_prompts=True,
+    )
+    meta_builder.subworkflow_test_main = os.path.join(
+        self.nfcore_modules, "tests", "subworkflows", "nf-core", subworkflow, "main.nf"
+    )
+    meta_builder.scrape_workflow_entry_points()
+    assert meta_builder.entry_points[0] == f"test_{subworkflow}"
+
+
+def test_subworkflows_create_test_yml_check_inputs(self):
+    """Test the check_inputs() function - raise UserWarning because test.yml exists"""
+    cwd = os.getcwd()
+    os.chdir(self.nfcore_modules)
+    subworkflow = "test_subworkflow"
+    meta_builder = nf_core.subworkflows.SubworkflowTestYmlBuilder(
+        subworkflow=f"{subworkflow}",
+        directory=self.pipeline_dir,
+        test_yml_output_path="./",
+        no_prompts=True,
+    )
+    meta_builder.subworkflow_test_main = os.path.join(
+        self.nfcore_modules, "tests", "subworkflows", "nf-core", subworkflow, "main.nf"
+    )
+    with pytest.raises(UserWarning) as excinfo:
+        meta_builder.check_inputs()
+    os.chdir(cwd)
+    assert "Test YAML file already exists!" in str(excinfo.value)

tests/test_subworkflows.py

@@ -93,6 +93,13 @@ def tearDown(self):
         test_subworkflows_create_nfcore_modules,
         test_subworkflows_create_succeed,
     )
+    from .subworkflows.create_test_yml import (
+        test_subworkflows_create_test_yml_check_inputs,
+        test_subworkflows_create_test_yml_entry_points,
+        test_subworkflows_create_test_yml_get_md5,
+        test_subworkflows_custom_yml_dumper,
+        test_subworkflows_test_file_dict,
+    )
     from .subworkflows.info import (
         test_subworkflows_info_in_modules_repo,
         test_subworkflows_info_local,