Merge branch 'dev' into feature/python-typing/list.py

CHANGELOG.md

@@ -13,8 +13,12 @@
 
 ### Modules
 
+- Added stub test creation to `create_test_yml` ([#2476](https://github.com/nf-core/tools/pull/2476))
+
 ### Subworkflows
 
+- Added stub test creation to `create_test_yml` ([#2476](https://github.com/nf-core/tools/pull/2476))
+
 ### General
 
 # [v2.10 - Nickel Ostrich](https://github.com/nf-core/tools/releases/tag/2.10) + [2023-09-25]
@@ -45,6 +49,7 @@
 ### Linting
 
 - Add new command `nf-core subworkflows lint` ([#2379](https://github.com/nf-core/tools/pull/2379))
+- Check for existence of test profile ([#2478](https://github.com/nf-core/tools/pull/2478))
 
 ### Modules
 

nf_core/lint/nextflow_config.py

@@ -108,6 +108,11 @@ def nextflow_config(self):
 
                 lint:
                     nextflow_config: False
+
+    **The configuration should contain the following or the test will fail:**
+
+    * A ``test`` configuration profile should exist.
+
     """
     passed = []
     warned = []
@@ -312,4 +317,32 @@ def nextflow_config(self):
                 )
             )
 
+    # Check for the availability of the "test" configuration profile by parsing nextflow.config
+    with open(os.path.join(self.wf_path, "nextflow.config"), "r") as f:
+        content = f.read()
+
+        # Remove comments
+        cleaned_content = re.sub(r"//.*", "", content)
+        cleaned_content = re.sub(r"/\*.*?\*/", "", content, flags=re.DOTALL)
+
+        match = re.search(r"\bprofiles\s*{", cleaned_content)
+        if not match:
+            failed.append("nextflow.config does not contain `profiles` scope, but `test` profile is required")
+        else:
+            # Extract profiles scope content and check for test profile
+            start = match.end()
+            end = start
+            brace_count = 1
+            while brace_count > 0 and end < len(content):
+                if cleaned_content[end] == "{":
+                    brace_count += 1
+                elif cleaned_content[end] == "}":
+                    brace_count -= 1
+                end += 1
+            profiles_content = cleaned_content[start : end - 1].strip()
+            if re.search(r"\btest\s*{", profiles_content):
+                passed.append("nextflow.config contains configuration profile `test`")
+            else:
+                failed.append("nextflow.config does not contain configuration profile `test`")
+
     return {"passed": passed, "warned": warned, "failed": failed, "ignored": ignored}

nf_core/modules/test_yml_builder.py

@@ -142,12 +142,19 @@ def build_all_tests(self):
         """
         Go over each entry point and build structure
         """
+
+        # Build the other tests
         for entry_point in self.entry_points:
             ep_test = self.build_single_test(entry_point)
             if ep_test:
                 self.tests.append(ep_test)
 
-    def build_single_test(self, entry_point):
+        # Build the stub test
+        stub_test = self.build_single_test(self.entry_points[0], stub=True)
+        if stub_test:
+            self.tests.append(stub_test)
+
+    def build_single_test(self, entry_point, stub=False):
         """Given the supplied cli flags, prompt for any that are missing.
 
         Returns: Test command
@@ -158,6 +165,8 @@ def build_single_test(self, entry_point):
             "tags": [],
             "files": [],
         }
+        stub_option = " -stub" if stub else ""
+        stub_name = " stub" if stub else ""
 
         # Print nice divider line
         console = rich.console.Console()
@@ -166,7 +175,7 @@ def build_single_test(self, entry_point):
         log.info(f"Building test meta for entry point '{entry_point}'")
 
         while ep_test["name"] == "":
-            default_val = f"{self.module_name.replace('/', ' ')} {entry_point}"
+            default_val = f"{self.module_name.replace('/', ' ')} {entry_point}{stub_name}"
             if self.no_prompts:
                 ep_test["name"] = default_val
             else:
@@ -178,6 +187,7 @@ def build_single_test(self, entry_point):
             default_val = (
                 f"nextflow run ./tests/modules/{self.org}/{self.module_name} -entry {entry_point} "
                 f"-c ./tests/config/nextflow.config"
+                f"{stub_option}"
             )
             if self.no_prompts:
                 ep_test["command"] = default_val
@@ -200,7 +210,7 @@ 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(ep_test["command"])
+        ep_test["files"] = self.get_md5_sums(ep_test["command"], stub=stub)
 
         return ep_test
 
@@ -227,7 +237,7 @@ def _md5(self, fname):
         md5sum = hash_md5.hexdigest()
         return md5sum
 
-    def create_test_file_dict(self, results_dir, is_repeat=False):
+    def create_test_file_dict(self, results_dir, is_repeat=False, stub=False):
         """Walk through directory and collect md5 sums"""
         test_files = []
         for root, _, files in os.walk(results_dir, followlinks=True):
@@ -236,13 +246,13 @@ def create_test_file_dict(self, results_dir, is_repeat=False):
                 # add the key here so that it comes first in the dict
                 test_file = {"path": file_path}
                 # Check that this isn't an empty file
-                if self.check_if_empty_file(file_path):
+                if self.check_if_empty_file(file_path) and not stub:
                     if not is_repeat:
                         self.errors.append(f"Empty file found! '{os.path.basename(file_path)}'")
                 # Add the md5 anyway, linting should fail later and can be manually removed if needed.
                 #  Originally we skipped this if empty, but then it's too easy to miss the warning.
                 #  Equally, if a file is legitimately empty we don't want to prevent this from working.
-                if filename != "versions.yml":
+                if filename != "versions.yml" and not stub:
                     # Only add md5sum if the file is not versions.yml
                     file_md5 = self._md5(file_path)
                     test_file["md5sum"] = file_md5
@@ -254,7 +264,7 @@ def create_test_file_dict(self, results_dir, is_repeat=False):
 
         return test_files
 
-    def get_md5_sums(self, command, results_dir=None, results_dir_repeat=None):
+    def get_md5_sums(self, command, results_dir=None, results_dir_repeat=None, stub=False):
         """
         Recursively go through directories and subdirectories
         and generate tuples of (<file_path>, <md5sum>)
@@ -276,11 +286,11 @@ def get_md5_sums(self, command, results_dir=None, results_dir_repeat=None):
                     log.error(f"Directory '{results_dir}' does not exist")
                     results_dir = None
 
-        test_files = self.create_test_file_dict(results_dir=results_dir)
+        test_files = self.create_test_file_dict(results_dir=results_dir, stub=stub)
 
         # If test was repeated, compare the md5 sums
         if results_dir_repeat:
-            test_files_repeat = self.create_test_file_dict(results_dir=results_dir_repeat, is_repeat=True)
+            test_files_repeat = self.create_test_file_dict(results_dir=results_dir_repeat, is_repeat=True, stub=stub)
 
             # Compare both test.yml files
             for i in range(len(test_files)):

nf_core/subworkflows/test_yml_builder.py

@@ -145,12 +145,19 @@ def build_all_tests(self):
         """
         Go over each entry point and build structure
         """
+
+        # Build the other tests
         for entry_point in self.entry_points:
             ep_test = self.build_single_test(entry_point)
             if ep_test:
                 self.tests.append(ep_test)
 
-    def build_single_test(self, entry_point):
+        # Build the stub test
+        stub_test = self.build_single_test(self.entry_points[0], stub=True)
+        if stub_test:
+            self.tests.append(stub_test)
+
+    def build_single_test(self, entry_point, stub=False):
         """Given the supplied cli flags, prompt for any that are missing.
 
         Returns: Test command
@@ -161,6 +168,8 @@ def build_single_test(self, entry_point):
             "tags": [],
             "files": [],
         }
+        stub_option = " -stub" if stub else ""
+        stub_name = " stub" if stub else ""
 
         # Print nice divider line
         console = rich.console.Console()
@@ -169,14 +178,14 @@ def build_single_test(self, entry_point):
         log.info(f"Building test meta for entry point '{entry_point}'")
 
         while ep_test["name"] == "":
-            default_val = f"{self.subworkflow} {entry_point}"
+            default_val = f"{self.subworkflow} {entry_point}{stub_name}"
             if self.no_prompts:
                 ep_test["name"] = default_val
             else:
                 ep_test["name"] = rich.prompt.Prompt.ask("[violet]Test name", default=default_val).strip()
 
         while ep_test["command"] == "":
-            default_val = f"nextflow run ./tests/subworkflows/{self.modules_repo.repo_path}/{self.subworkflow} -entry {entry_point} -c ./tests/config/nextflow.config"
+            default_val = f"nextflow run ./tests/subworkflows/{self.modules_repo.repo_path}/{self.subworkflow} -entry {entry_point} -c ./tests/config/nextflow.config{stub_option}"
             if self.no_prompts:
                 ep_test["command"] = default_val
             else:
@@ -195,7 +204,7 @@ 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(ep_test["command"])
+        ep_test["files"] = self.get_md5_sums(ep_test["command"], stub=stub)
 
         return ep_test
 
@@ -244,7 +253,7 @@ def _md5(self, fname):
         md5sum = hash_md5.hexdigest()
         return md5sum
 
-    def create_test_file_dict(self, results_dir, is_repeat=False):
+    def create_test_file_dict(self, results_dir, is_repeat=False, stub=False):
         """Walk through directory and collect md5 sums"""
         test_files = []
         for root, _, files in os.walk(results_dir, followlinks=True):
@@ -256,14 +265,15 @@ def create_test_file_dict(self, results_dir, is_repeat=False):
                 # add the key here so that it comes first in the dict
                 test_file = {"path": file_path}
                 # Check that this isn't an empty file
-                if self.check_if_empty_file(file_path):
+                if self.check_if_empty_file(file_path) and not stub:
                     if not is_repeat:
                         self.errors.append(f"Empty file found! '{os.path.basename(file_path)}'")
                 # Add the md5 anyway, linting should fail later and can be manually removed if needed.
                 #  Originally we skipped this if empty, but then it's too easy to miss the warning.
                 #  Equally, if a file is legitimately empty we don't want to prevent this from working.
-                file_md5 = self._md5(file_path)
-                test_file["md5sum"] = file_md5
+                if not stub:
+                    file_md5 = self._md5(file_path)
+                    test_file["md5sum"] = file_md5
                 # Switch out the results directory path with the expected 'output' directory
                 test_file["path"] = file_path.replace(results_dir, "output")
                 test_files.append(test_file)
@@ -272,7 +282,7 @@ def create_test_file_dict(self, results_dir, is_repeat=False):
 
         return test_files
 
-    def get_md5_sums(self, command, results_dir=None, results_dir_repeat=None):
+    def get_md5_sums(self, command, results_dir=None, results_dir_repeat=None, stub=False):
         """
         Recursively go through directories and subdirectories
         and generate tuples of (<file_path>, <md5sum>)
@@ -294,11 +304,11 @@ def get_md5_sums(self, command, results_dir=None, results_dir_repeat=None):
                     log.error(f"Directory '{results_dir}' does not exist")
                     results_dir = None
 
-        test_files = self.create_test_file_dict(results_dir=results_dir)
+        test_files = self.create_test_file_dict(results_dir=results_dir, stub=stub)
 
         # If test was repeated, compare the md5 sums
         if results_dir_repeat:
-            test_files_repeat = self.create_test_file_dict(results_dir=results_dir_repeat, is_repeat=True)
+            test_files_repeat = self.create_test_file_dict(results_dir=results_dir_repeat, is_repeat=True, stub=stub)
 
             # Compare both test.yml files
             for i in range(len(test_files)):

tests/lint/nextflow_config.py

@@ -1,3 +1,6 @@
+import os
+import re
+
 import nf_core.create
 import nf_core.lint
 
@@ -33,3 +36,20 @@ def test_nextflow_config_dev_in_release_mode_failed(self):
     result = lint_obj.nextflow_config()
     assert len(result["failed"]) > 0
     assert len(result["warned"]) == 0
+
+
+def test_nextflow_config_missing_test_profile_failed(self):
+    """Test failure if config file does not contain `test` profile."""
+    new_pipeline = self._make_pipeline_copy()
+    # Change the name of the test profile so there is no such profile
+    nf_conf_file = os.path.join(new_pipeline, "nextflow.config")
+    with open(nf_conf_file, "r") as f:
+        content = f.read()
+        fail_content = re.sub(r"\btest\b", "testfail", content)
+    with open(nf_conf_file, "w") as f:
+        f.write(fail_content)
+    lint_obj = nf_core.lint.PipelineLint(new_pipeline)
+    lint_obj._load_pipeline_config()
+    result = lint_obj.nextflow_config()
+    assert len(result["failed"]) > 0
+    assert len(result["warned"]) == 0

tests/test_lint.py

@@ -213,6 +213,7 @@ def test_sphinx_md_files(self):
         test_nextflow_config_bad_name_fail,
         test_nextflow_config_dev_in_release_mode_failed,
         test_nextflow_config_example_pass,
+        test_nextflow_config_missing_test_profile_failed,
     )
     from .lint.version_consistency import test_version_consistency