backend/sdoc: allows slashes in the node UID and MultipleChoice fields

Closes #2197

Author: stanislaw

strictdoc/backend/sdoc/grammar/grammar.py

@@ -1,4 +1,4 @@
-REGEX_UID = r"([\w]+[\w()\-. ]*)"
+REGEX_UID = r"([\w]+[\w()\-\/. ]*)"
 
 NEGATIVE_MULTILINE_STRING_START = "(?!>>>\n)"
 NEGATIVE_MULTILINE_STRING_END = "(?!^<<<)"
@@ -44,7 +44,7 @@
 ;
 
 FieldName[noskipws]:
-  /{NEGATIVE_UID}{NEGATIVE_RELATIONS}[A-Z]+[A-Z_]*/
+  /{NEGATIVE_UID}{NEGATIVE_RELATIONS}[A-Z]+[A-Z_0-9]*/
 ;
 """
 

strictdoc/backend/sdoc/validations/sdoc_validator.py

@@ -1,4 +1,4 @@
-# mypy: disable-error-code="arg-type,no-redef,no-untyped-call,no-untyped-def,union-attr"
+# mypy: disable-error-code="arg-type,no-untyped-call,no-untyped-def,union-attr"
 import re
 from typing import Iterator, Optional, Set
 
@@ -19,11 +19,20 @@
     RequirementFieldName,
 )
 
+MULTIPLE_CHOICE_FIELD_KW = r"[a-zA-Z0-9\/_]+"
+MULTIPLE_CHOICE_FIELD_REGEX = re.compile(
+    rf"^{MULTIPLE_CHOICE_FIELD_KW}(, {MULTIPLE_CHOICE_FIELD_KW})*$"
+)
+
+
+def multi_choice_regex_match(value: str) -> bool:
+    """
+    Validate MultipleChoice field value.
+    Bug: Forward slashes in field values cause errors #2197,
+    https://github.com/strictdoc-project/strictdoc/issues/2197
+    """
 
-def multi_choice_regex_match(value):
-    keyword = "[a-zA-Z0-9_]+"
-    regex = re.compile(rf"^{keyword}(, {keyword})*$")
-    match = regex.match(value)
+    match = MULTIPLE_CHOICE_FIELD_REGEX.match(value)
     return match is not None
 
 
@@ -37,7 +46,7 @@ class as well. This would remove the need in the ParseContext class
     """
 
     @staticmethod
-    def validate_document(document: SDocDocument):
+    def validate_document(document: SDocDocument) -> None:
         assert isinstance(document, SDocDocument), document
         SDocValidator._validate_document_config(document)
         SDocValidator._validate_document_view(document)
@@ -46,23 +55,25 @@ def validate_document(document: SDocDocument):
     @staticmethod
     def validate_grammar_from_file(
         path_to_grammar: str, grammar_from_file: DocumentGrammar
-    ):
+    ) -> None:
         for grammar_element_ in grammar_from_file.elements:
             SDocValidator.validate_grammar_element(
                 path_to_grammar, grammar_element_
             )
 
     @staticmethod
-    def _validate_grammar(document: SDocDocument):
+    def _validate_grammar(document: SDocDocument) -> None:
+        assert document.meta is not None
+        assert document.grammar is not None
         for grammar_element_ in document.grammar.elements:
             SDocValidator.validate_grammar_element(
                 document.meta.input_doc_full_path, grammar_element_
             )
 
     @staticmethod
     def validate_grammar_element(
-        path_to_grammar, grammar_element: GrammarElement
-    ):
+        path_to_grammar: str, grammar_element: GrammarElement
+    ) -> None:
         if grammar_element.content_field[0] not in grammar_element.fields_map:
             raise StrictDocSemanticError.grammar_missing_reserved_statement(
                 grammar_element,
@@ -101,7 +112,7 @@ def validate_grammar_element(
             )
 
     @staticmethod
-    def _validate_document_config(document: SDocDocument):
+    def _validate_document_config(document: SDocDocument) -> None:
         document_config: DocumentConfig = document.config
         if document_config.default_view is not None:
             if document.view is None:
@@ -122,7 +133,7 @@ def _validate_document_config(document: SDocDocument):
                     )
 
     @staticmethod
-    def _validate_document_view(document: SDocDocument):
+    def _validate_document_view(document: SDocDocument) -> None:
         document_view: Optional[DocumentView] = document.view
         if document_view is not None:
             for view in document_view.views:
@@ -156,14 +167,14 @@ def validate_node(
         requirement: SDocNode,
         document_grammar: DocumentGrammar,
         path_to_sdoc_file: str,
-        auto_uid_mode: bool = True,
-    ):
+        auto_uid_mode: bool = False,
+    ) -> None:
         if requirement.node_type not in document_grammar.registered_elements:
             raise StrictDocSemanticError.unknown_node_type(
                 requirement, path_to_sdoc_file
             )
 
-        grammar_element = document_grammar.elements_by_type[
+        grammar_element: GrammarElement = document_grammar.elements_by_type[
             requirement.node_type
         ]
         registered_fields: Set[str] = set(grammar_element.get_field_titles())
@@ -177,9 +188,6 @@ def validate_node(
                     path_to_sdoc_file=path_to_sdoc_file,
                 )
 
-        grammar_element: GrammarElement = document_grammar.elements_by_type[
-            requirement.node_type
-        ]
         grammar_fields_iterator: Iterator[GrammarElementField] = iter(
             grammar_element.fields
         )
@@ -250,6 +258,9 @@ def validate_requirement_field(
         path_to_sdoc_file: str,
         auto_uid_mode: bool = True,
     ) -> bool:
+        """
+        Validate a single node field using its grammar element field definition.
+        """
         if grammar_field is None:
             if requirement_field is None:
                 # Both grammar and requirements fields are over.
@@ -282,11 +293,6 @@ def validate_requirement_field(
                     grammar_field.title
                     not in requirement.ordered_fields_lookup.keys()
                 ):
-                    # A special case: The Manage UID command auto-generates the UID,
-                    # so the field presence validation has to be relaxed.
-                    # The GitHub issue report:
-                    # manage auto-uid: UID field REQUIRED True leads to an error
-                    # https://github.com/strictdoc-project/strictdoc/issues/1896
                     if grammar_field.title == "UID" and auto_uid_mode:
                         return False
 

strictdoc/cli/main.py

@@ -129,6 +129,7 @@ def _main(parallelizer: Parallelizer) -> None:
         # FIXME: This must be improved.
         project_config.input_paths = [manage_config.input_path]
         # FIXME: This must be improved.
+        project_config.auto_uid_mode = True
         project_config.autouuid_include_sections = (
             manage_config.include_sections
         )

strictdoc/commands/manage_autouid_command.py

@@ -28,7 +28,6 @@ def execute(*, project_config: ProjectConfig, parallelizer: Parallelizer):
                 TraceabilityIndexBuilder.create(
                     project_config=project_config,
                     parallelizer=parallelizer,
-                    auto_uid_mode=True,
                 )
             )
         except DocumentTreeError as exc:

strictdoc/core/document_meta.py

@@ -12,16 +12,16 @@ class DocumentMeta:
     def __init__(
         self,
         level: int,
-        file_tree_mount_folder,
+        file_tree_mount_folder: str,
         document_filename: str,
-        document_filename_base,
-        input_doc_full_path,
+        document_filename_base: str,
+        input_doc_full_path: str,
         input_doc_rel_path: SDocRelativePath,
         input_doc_dir_rel_path: SDocRelativePath,
         input_doc_assets_dir_rel_path: SDocRelativePath,
         output_document_dir_full_path: str,
         output_document_dir_rel_path: SDocRelativePath,
-    ):
+    ) -> None:
         """
         Example explaining meta data stored by this class:
 

strictdoc/core/project_config.py

@@ -173,6 +173,18 @@ def __init__(
         self.reqif_enable_mid: bool = reqif_enable_mid
         self.reqif_import_markup: Optional[str] = reqif_import_markup
 
+        #
+        # auto_uid_mode: default is False. The True-case is used by the
+        # manage/auto_uid command: the SDocNodeValidator will
+        # not raise an exception if it sees a node with a missing UID.
+        # Important for a special case:
+        # The Manage UID command auto-generates the UID, so the field presence
+        # validation has to be relaxed.
+        # The GitHub issue report:
+        # manage auto-uid: UID field REQUIRED True leads to an error
+        # https://github.com/strictdoc-project/strictdoc/issues/1896
+        #
+        self.auto_uid_mode = False
         self.autouuid_include_sections: bool = False
 
         self.config_last_update: Optional[datetime.datetime] = (

strictdoc/core/traceability_index_builder.py

@@ -66,7 +66,6 @@ def create(
         project_config: ProjectConfig,
         parallelizer,
         skip_source_files: bool = False,
-        auto_uid_mode: bool = False,
     ) -> TraceabilityIndex:
         # TODO: It would be great to hide this code behind --development flag.
         # There is no need for this to be activated in the Pip-released builds.
@@ -109,7 +108,6 @@ def create(
             TraceabilityIndexBuilder.create_from_document_tree(
                 document_tree,
                 project_config,
-                auto_uid_mode,
             )
         )
         traceability_index.document_tree = document_tree
@@ -208,7 +206,6 @@ def create(
     def create_from_document_tree(
         document_tree: DocumentTree,
         project_config: ProjectConfig,
-        auto_uid_mode: bool = True,
     ) -> TraceabilityIndex:
         # FIXME: Too many things going on below. Would be great to simplify this
         # workflow.
@@ -386,7 +383,7 @@ def create_from_document_tree(
                             node,
                             document_grammar=document.grammar,
                             path_to_sdoc_file=document.meta.input_doc_full_path,
-                            auto_uid_mode=auto_uid_mode,
+                            auto_uid_mode=project_config.auto_uid_mode,
                         )
                     except StrictDocSemanticError as exc:
                         print(exc.to_print_message())  # noqa: T201

strictdoc/export/spdx/spdx_generator.py

@@ -235,8 +235,8 @@ def export_tree(
         lookup_file_name_to_spdx_file: Dict[str, File] = {}
 
         for document_ in traceability_index.document_tree.document_list:
-            with open(document_.meta.input_doc_full_path, "rb") as file_:
-                document_bytes = file_.read()
+            with open(document_.meta.input_doc_full_path, "rb") as input_file_:
+                document_bytes = input_file_.read()
 
             """
             Create SPDX File from SDoc Document.
@@ -318,8 +318,8 @@ def export_tree(
                         if node_link_path_ in lookup_file_name_to_spdx_file:
                             continue
 
-                        with open(node_link_path_, "rb") as file_:
-                            file_bytes = file_.read()
+                        with open(node_link_path_, "rb") as node_link_file_:
+                            file_bytes = node_link_file_.read()
 
                         source_spdx_file = (
                             sdoc_spdx_converter.convert_file_to_file(
@@ -435,10 +435,10 @@ def export_tree(
         if True:
             sdoc_document = SPDXToSDocConverter.convert(spdx_container)
 
-            sdoc_output = SDWriter(project_config).write(sdoc_document)
+            sdoc_output: str = SDWriter(project_config).write(sdoc_document)
 
             sdoc_output_path = os.path.join(
                 output_spdx_root, "output.spdx.sdoc"
             )
-            with open(sdoc_output_path, "w", encoding="utf8") as file_:
-                file_.write(sdoc_output)
+            with open(sdoc_output_path, "w", encoding="utf8") as output_file_:
+                output_file_.write(sdoc_output)

tests/unit/helpers/test_document_builder.py renamed to tests/unit/helpers/document_builder.py

@@ -0,0 +1,17 @@
+from strictdoc.core.document_meta import DocumentMeta
+from strictdoc.helpers.paths import SDocRelativePath
+
+
+def create_fake_document_meta():
+    return DocumentMeta(
+        level=1,
+        file_tree_mount_folder="root_folder",
+        document_filename="fake_doc.sdoc",
+        document_filename_base="fake_doc",
+        input_doc_full_path="/tmp/root_folder/fake_doc.sdoc",
+        input_doc_rel_path=SDocRelativePath("fake_doc.sdoc"),
+        input_doc_dir_rel_path=SDocRelativePath(""),
+        input_doc_assets_dir_rel_path=SDocRelativePath("_assets"),
+        output_document_dir_full_path="/tmp/output/root_folder/",
+        output_document_dir_rel_path=SDocRelativePath(""),
+    )