add --ignore_validation_errors (#161)

Co-authored-by: Jake Fennick <jake.fennick@axleinfo.com>

src/wic/ast.py

@@ -23,7 +23,8 @@ def read_ast_from_disk(homedir: str,
                        yaml_tree_tuple: YamlTree,
                        yml_paths: Dict[str, Dict[str, Path]],
                        tools: Tools,
-                       validator: Draft202012Validator) -> YamlTree:
+                       validator: Draft202012Validator,
+                       ignore_validation_errors: bool) -> YamlTree:
     """Reads the yml workflow definition files from disk (recursively) and inlines them into an AST
 
     Args:
@@ -32,6 +33,7 @@ def read_ast_from_disk(homedir: str,
         yml_paths (Dict[str, Dict[str, Path]]): The yml workflow definitions found using get_yml_paths()
         tools (Tools): The CWL CommandLineTool definitions found using get_tools_cwl()
         validator (Draft202012Validator): Used to validate the yml files against the autogenerated schema.
+        ignore_validation_errors (bool): Temporarily ignore validation errors. Do not use this permanently!
 
     Raises:
         Exception: If the yml file(s) do not exist
@@ -42,7 +44,8 @@ def read_ast_from_disk(homedir: str,
     (step_id, yaml_tree) = yaml_tree_tuple
 
     try:
-        validator.validate(yaml_tree)
+        if not ignore_validation_errors:
+            validator.validate(yaml_tree)
     except Exception as e:
         yaml_path = Path(step_id.stem)
         print('Failed to validate', yaml_path)
@@ -64,7 +67,8 @@ def read_ast_from_disk(homedir: str,
         for back_name, back in wic['wic']['backends'].items():
             plugin_ns = wic['wic'].get('namespace', 'global')
             stepid = StepId(back_name, plugin_ns)
-            backends_tree = read_ast_from_disk(homedir, YamlTree(stepid, back), yml_paths, tools, validator)
+            backends_tree = read_ast_from_disk(homedir, YamlTree(stepid, back), yml_paths, tools, validator,
+                                               ignore_validation_errors)
             backends_trees.append(backends_tree)
         yaml_tree['wic']['backends'] = dict(backends_trees)
         return YamlTree(step_id, yaml_tree)
@@ -109,7 +113,8 @@ def read_ast_from_disk(homedir: str,
                 sub_yaml_tree_raw: Yaml = yaml.safe_load(y.read())
 
             y_t = YamlTree(StepId(step_key, plugin_ns), sub_yaml_tree_raw)
-            (step_id_, sub_yml_tree) = read_ast_from_disk(homedir, y_t, yml_paths, tools, validator)
+            (step_id_, sub_yml_tree) = read_ast_from_disk(homedir, y_t, yml_paths, tools, validator,
+                                                          ignore_validation_errors)
 
             step_i_dict = {} if steps[i][step_key] is None else steps[i][step_key]
             # Do not merge these two dicts; use subtree and parentargs so we can

src/wic/cli.py

@@ -61,6 +61,8 @@
                     help='Enables the use of naming conventions in the inference algorithm')
 parser.add_argument('--validate_plugins', default=False, action="store_true",
                     help='Validate all CWL CommandLineTools')
+parser.add_argument('--ignore_validation_errors', default=False, action="store_true",
+                    help='Temporarily ignore validation errors. Do not use this permanently!')
 parser.add_argument('--no_skip_dollar_schemas', default=False, action="store_true",
                     help='''Does not skip processing $schemas tags in CWL files for performance.
                     Skipping significantly improves initial validation performance, but is not always desired.

src/wic/cwl_watcher.py

@@ -94,7 +94,7 @@ def rerun_cwltool(homedir: str, _directory_realtime: Path, cachedir_path: Path,
             plugin_ns = 'global'  # wic['wic'].get('namespace', 'global')
             step_id = StepId(yaml_path, plugin_ns)
             y_t = YamlTree(step_id, root_yaml_tree)
-            yaml_tree_raw = ast.read_ast_from_disk(homedir, y_t, yml_paths, tools_cwl, validator)
+            yaml_tree_raw = ast.read_ast_from_disk(homedir, y_t, yml_paths, tools_cwl, validator, True)
             yaml_tree = ast.merge_yml_trees(yaml_tree_raw, {}, tools_cwl)
             yml = yaml_tree.yml
         else:

src/wic/main.py

@@ -46,7 +46,8 @@ def main() -> None:
 
         for yml_path_str, yml_path in yml_paths_tuples:
             schema = wic_schema.compile_workflow_generate_schema(args.homedir, yml_path_str, yml_path,
-                                                                 tools_cwl, yml_paths, validator)
+                                                                 tools_cwl, yml_paths, validator,
+                                                                 args.ignore_validation_errors)
             # overwrite placeholders in schema_store. See comment in get_validator()
             schema_store[schema['$id']] = schema
 
@@ -69,7 +70,8 @@ def main() -> None:
     plugin_ns = wic['wic'].get('namespace', 'global')
     step_id = StepId(yaml_path, plugin_ns)
     y_t = YamlTree(step_id, root_yaml_tree)
-    yaml_tree_raw = ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator)
+    yaml_tree_raw = ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator,
+                                           args.ignore_validation_errors)
     # Write the combined workflow (with all subworkflows as children) to disk.
     with open(f'autogenerated/{Path(yaml_path).stem}_tree_raw.yml', mode='w', encoding='utf-8') as f:
         f.write(yaml.dump(yaml_tree_raw.yml))

src/wic/schemas/wic_schema.py

@@ -496,7 +496,8 @@ def compile_workflow_generate_schema(homedir: str,
                                      yml_path_str: str, yml_path: Path,
                                      tools_cwl: Tools,
                                      yml_paths: Dict[str, Dict[str, Path]],
-                                     validator: Draft202012Validator) -> Json:
+                                     validator: Draft202012Validator,
+                                     ignore_validation_errors: bool) -> Json:
     """Compiles a workflow and generates a schema which (recursively) includes the inputs/outputs from subworkflows.
 
     Args:
@@ -506,6 +507,7 @@ def compile_workflow_generate_schema(homedir: str,
         tools_cwl (Tools): The CWL CommandLineTool definitions found using get_tools_cwl()
         yml_paths (Dict[str, Dict[str, Path]]): The yml workflow definitions found using get_yml_paths()
         validator (Draft202012Validator): Used to validate the yml files against the autogenerated schema.
+        ignore_validation_errors (bool): Temporarily ignore validation errors. Do not use this permanently!
 
     Returns:
         Json: An autogenerated, documented schema based on the inputs and outputs of the Workflow.
@@ -519,7 +521,8 @@ def compile_workflow_generate_schema(homedir: str,
     plugin_ns = wic_tag['wic'].get('namespace', 'global')
     step_id = StepId(yml_path_str, plugin_ns)
     y_t = YamlTree(step_id, root_yaml_tree)
-    yaml_tree_raw = wic.ast.read_ast_from_disk(homedir, y_t, yml_paths, tools_cwl, validator)
+    yaml_tree_raw = wic.ast.read_ast_from_disk(homedir, y_t, yml_paths, tools_cwl, validator,
+                                               ignore_validation_errors)
     # with open(f'autogenerated/{Path(yml_path).stem}_tree_raw.yml', mode='w', encoding='utf-8') as f:
     #    f.write(yaml.dump(yaml_tree_raw.yml))
     yaml_tree = wic.ast.merge_yml_trees(yaml_tree_raw, {}, tools_cwl)

tests/test_examples.py

@@ -153,7 +153,8 @@ def run_workflows(yml_path_str: str, yml_path: Path, cwl_runner: str, args: argp
     plugin_ns = wic_tag['wic'].get('namespace', 'global')
     step_id = StepId(yml_path_str, plugin_ns)
     y_t = YamlTree(step_id, root_yaml_tree)
-    yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator)
+    yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator,
+                                               args.ignore_validation_errors)
     with open(f'autogenerated/{Path(yml_path).stem}_tree_raw.yml', mode='w', encoding='utf-8') as f:
         f.write(yaml.dump(yaml_tree_raw.yml))
     yaml_tree = wic.ast.merge_yml_trees(yaml_tree_raw, {}, tools_cwl)
@@ -212,7 +213,8 @@ def test_cwl_embedding_independence(yml_path_str: str, yml_path: Path) -> None:
     plugin_ns = wic_tag['wic'].get('namespace', 'global')
     step_id = StepId(yml_path_str + '.yml', plugin_ns)
     y_t = YamlTree(step_id, root_yaml_tree)
-    yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator)
+    yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator,
+                                               args.ignore_validation_errors)
     with open(f'autogenerated/{yml_path.stem}_tree_raw.yml', mode='w', encoding='utf-8') as f:
         f.write(yaml.dump(yaml_tree_raw.yml))
     yaml_tree = wic.ast.merge_yml_trees(yaml_tree_raw, {}, tools_cwl)
@@ -310,7 +312,8 @@ def test_inline_subworkflows(yml_path_str: str, yml_path: Path) -> None:
     plugin_ns = wic_tag['wic'].get('namespace', 'global')
     step_id = StepId(yml_path_str, plugin_ns)
     y_t = YamlTree(step_id, root_yaml_tree)
-    yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator)
+    yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator,
+                                               args.ignore_validation_errors)
     with open(f'autogenerated/{Path(yml_path).stem}_tree_raw.yml', mode='w', encoding='utf-8') as f:
         f.write(yaml.dump(yaml_tree_raw.yml))
     yaml_tree = wic.ast.merge_yml_trees(yaml_tree_raw, {}, tools_cwl)

tests/test_fuzzy_compile.py

@@ -51,7 +51,8 @@ def test_fuzzy_compile(self, yml: Yaml) -> None:
         args = get_args(str(yml_path))
 
         y_t = YamlTree(StepId('random_stepid', plugin_ns), yml)
-        yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator)
+        yaml_tree_raw = wic.ast.read_ast_from_disk(args.homedir, y_t, yml_paths, tools_cwl, validator,
+                                                   args.ignore_validation_errors)
         yaml_tree = wic.ast.merge_yml_trees(yaml_tree_raw, {}, tools_cwl)
 
         graph_gv = graphviz.Digraph(name=f'cluster_{yml_path}')

tests/test_setup.py

@@ -46,7 +46,8 @@ def get_args(yaml_path: str = '', cwl_inline_subworkflows: bool = False) -> argp
 
 for yml_path_str, yml_path in yml_paths_tuples:
     schema = wic.schemas.wic_schema.compile_workflow_generate_schema(args.homedir, yml_path_str, yml_path,
-                                                                     tools_cwl, yml_paths, validator)
+                                                                     tools_cwl, yml_paths, validator,
+                                                                     args.ignore_validation_errors)
     # overwrite placeholders in schema_store. See comment in get_validator()
     schema_store[schema['$id']] = schema