Resume XML parsing after reporting error

CHANGELOG.md

@@ -5,6 +5,13 @@ The first name in a bullet point below is the Python class name while the second
 Elements marked as `collectable` means that they can be added directly to a package.
 Non-collectable elements are various sub-elements to collectable elements.
 
+## [Unreleased]
+
+### Changed
+
+* Reader class attempts to resume parsing at next element if an error occurs.
+  * To stop parsing on first error, give the option `stop_on_error=True` to method `Reader.read_file`.
+
 ## [v0.5.3] - 2024-03-31
 
 ### Fixed

examples/xml/reader/data/xml_with_errors.arxml

@@ -0,0 +1,55 @@
+<?xml version="1.0" encoding="utf-8"?>
+<AUTOSAR xsi:schemaLocation="http://autosar.org/schema/r4.0 AUTOSAR_00051.xsd" xmlns="http://autosar.org/schema/r4.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
+  <AR-PACKAGES>
+    <AR-PACKAGE>
+      <SHORT-NAME>Constants</SHORT-NAME>
+      <ELEMENTS>
+        <CONSTANT-SPECIFICATION>
+            <SHORT-NAME>C_HazardSwitchIllumination_IV</SHORT-NAME>
+            <VALUE-SPEC>
+            <NUMERICAL-VALUE-SPECIFICATION>
+                <SHORT-LABEL>C_HazardSwitchIllumination_IV</SHORT-LABEL>
+                <VALUE>0</VALUE>
+            </NUMERICAL-VALUE-SPECIFICATION>
+            </VALUE-SPEC>
+        </CONSTANT-SPECIFICATION>
+        <CONSTANT-SPECIFICATION>
+            <SHORT-NAME>C_TelltaleBrightness_IV</SHORT-NAME>
+            <VALUE-SPEC>
+            <NUMERICAL-VALUE-SPECIFICATION>
+                <SHORT-LABEL>C_TelltaleBrightness_IV</SHORT-LABEL>
+                <VALUE>FiftyFive</VALUE>
+            </NUMERICAL-VALUE-SPECIFICATION>
+            </VALUE-SPEC>
+        </CONSTANT-SPECIFICATION>
+        <CONSTANT-SPECIFICATION>
+            <SHORT-NAME>C_HazardSwitchIllumination_IV</SHORT-NAME>
+            <VALUE-SPEC>
+            <NUMERICAL-VALUE-SPECIFICATION>
+                <SHORT-LABEL>C_HazardSwitchIllumination_IV</SHORT-LABEL>
+                <VALUE>0</VALUE>
+            </NUMERICAL-VALUE-SPECIFICATION>
+            </VALUE-SPEC>
+        </CONSTANT-SPECIFICATION>
+        <CONSTANT-SPECIFICATION>
+            <SHORT-NAME>C_LightSensor_IV</SHORT-NAME>
+            <VALUE-SPEC>
+            <NUMERICAL-VALUE-SPECIFICATION>
+                <SHORT-LABEL>C_LightSensor_IV</SHORT-LABEL>
+                <VALUE>65535</VALUE>
+            </NUMERICAL-VALUE-SPECIFICATION>
+            </VALUE-SPEC>
+        </CONSTANT-SPECIFICATION>
+        <CONSTANT-SPECIFICATION>
+            <SHORT-NAME>C_HazardSwitchIllumination_IV</SHORT-NAME>
+            <VALUE-SPEC>
+            <NUMERICAL-VALUE-SPECIFICATION>
+                <SHORT-LABEL>C_HazardSwitchIllumination_IV</SHORT-LABEL>
+                <VALUE>0</VALUE>
+            </NUMERICAL-VALUE-SPECIFICATION>
+            </VALUE-SPEC>
+        </CONSTANT-SPECIFICATION>
+      </ELEMENTS>
+    </AR-PACKAGE>
+  </AR-PACKAGES>
+</AUTOSAR>

examples/xml/reader/print_errors.py

@@ -0,0 +1,27 @@
+"""
+SwAddrMethod example
+"""
+import os
+import autosar.xml
+import autosar.xml.element as ar_element
+
+
+def print_package_stats(package: ar_element.Package, parent: str = "/"):
+    """
+    Prints number of elements in package and sub-packages
+    """
+    print(f'Number of elements in "{parent}{package.name}": {len(package.elements)}')
+    parent_path = parent + package.name + "/"
+    for sub_package in package.packages:
+        print_package_stats(sub_package, parent_path)
+
+
+if __name__ == "__main__":
+
+    file_path = os.path.join(os.path.dirname(__file__), 'data', 'xml_with_errors.arxml')
+    # This file contains an invalid numerical constants as well as two duplicates.
+    # From the 5 elements in the XML only 2 is successfully placed into the resulting document.
+    reader = autosar.xml.Reader()
+    document = reader.read_file(file_path)
+    for pkg in document.packages:
+        print_package_stats(pkg)

run_examples.cmd

@@ -13,6 +13,7 @@ python examples\xml\port_interface\client_server_interface.py
 python examples\xml\port_interface\mode_switch_interface.py
 python examples\xml\component\application_component.py
 python examples\xml\component\composition_component.py
+python examples\xml\reader\print_errors.py
 python examples\template\generate_xml_using_config.py
 python examples\template\generate_xml_without_config.py
 python examples\generator\data_types\gen_type_defs_scalar.py

src/autosar/xml/element.py

@@ -3411,15 +3411,15 @@ def append(self, item: CollectableElement):
         if isinstance(item, Package):
             package: Package = item
             if package.name in self._collection_map:
-                raise ValueError(
+                raise ar_except.DuplicateElement(
                     f"Package with SHORT-NAME '{package.name}' already exists in package '{self.name}")
             package.parent = self
             self.packages.append(package)
             self._collection_map[package.name] = package
         elif isinstance(item, ARElement):
             elem: ARElement = item
             if elem.name in self._collection_map:
-                raise ValueError(
+                raise ar_except.DuplicateElement(
                     f"Element with SHORT-NAME '{elem.name}' already exists in package '{self.name}'")
             elem.parent = self
             self.elements.append(elem)

src/autosar/xml/exception.py

@@ -11,6 +11,12 @@ class ParseError(RuntimeError):
     """
 
 
+class DuplicateElement(ValueError):
+    """
+    Element with this name already in exist in current context
+    """
+
+
 class VersionError(ValueError):
     """
     Invalid/Unsupported XML version

src/autosar/xml/reader.py

@@ -110,6 +110,7 @@ def __init__(self,
         self.schema_file: str = ''
         self.schema_version = schema_version
         self.document: ar_document.Document = None
+        self.stop_on_error = False
         self.switcher_collectable = {  # Collectable elements
             # CompuMethod
             'COMPU-METHOD': self._read_compu_method,
@@ -277,7 +278,7 @@ def __init__(self,
             "VariableDataPrototype": self._read_variable_data_prototype,
         }
 
-    def read_file(self, file_path: str) -> ar_document.Document:
+    def read_file(self, file_path: str, stop_on_error: bool = False) -> ar_document.Document:
         """
         Reads ARXML document file
         """
@@ -286,12 +287,13 @@ def read_file(self, file_path: str) -> ar_document.Document:
         self.file_path = file_path
         self.file_base_name = os.path.basename(file_path)
         self.observed_unsupported_elements = set()
+        self.stop_on_error = stop_on_error
         self._clean_namespace('http://autosar.org/schema/r4.0')
         self._read_root_element()
         self._read_packages()
         return self.document
 
-    def read_str(self, xml: str) -> None | ar_document.Document:
+    def read_str(self, xml: str, stop_on_error: bool = False) -> None | ar_document.Document:
         """
         Reads ARXML document from string.
         """
@@ -300,6 +302,7 @@ def read_str(self, xml: str) -> None | ar_document.Document:
         self.xml_root = ElementTree.fromstring(bytes(xml, encoding="utf-8"))
         self.file_path = ""
         self.file_base_name = ""
+        self.stop_on_error = stop_on_error
         self._clean_namespace('http://autosar.org/schema/r4.0')
         self._read_root_element()
         self._read_packages()
@@ -344,13 +347,19 @@ def _report_unprocessed_element(self, xml_elem: ElementTree.Element):
                 else:
                     print(f"Unprocessed element <{xml_elem.tag}>", file=sys.stderr)
 
-    def _raise_parse_error(self, element: ElementTree.Element, message: str):
+    def _element_error_message(self, element: ElementTree.Element, message: str) -> str:
         """
-        Raises an ARXML parse error with message
+        Generates an error message with file-name and source-line
         """
         file = self.file_path if self.use_full_path_on_warning else self.file_base_name
         header = f"{file}({element.sourceline}): "
-        raise ar_exception.ParseError(header + message)
+        return header + message
+
+    def _raise_parse_error(self, element: ElementTree.Element, message: str):
+        """
+        Raises an ARXML parse error with message
+        """
+        raise ar_exception.ParseError(self._element_error_message(element, message))
 
     def _clean_namespace(self, namespace: str) -> None:
         """
@@ -574,9 +583,23 @@ def _read_package_elements(self, package: ar_element.Package, xml_elements: Elem
             read_method = self.switcher_collectable.get(
                 xml_child_elem.tag, None)
             if read_method is not None:
-                element = read_method(xml_child_elem)
-                assert isinstance(element, ar_element.ARElement)
-                package.append(element)
+                try:
+                    element = read_method(xml_child_elem)
+                    assert isinstance(element, ar_element.ARElement)
+                    package.append(element)
+                except ar_exception.ParseError as exc:
+                    msg = "Parse error encountered while reading element starting on this line"
+                    message = self._element_error_message(xml_child_elem, msg)
+                    if self.stop_on_error:
+                        raise ar_exception.ParseError(message) from exc
+                    print(message + ":")
+                    print("    " + str(exc))
+                except ar_exception.DuplicateElement as exc:
+                    message = self._element_error_message(xml_child_elem,
+                                                          str(exc))
+                    if self.stop_on_error:
+                        raise ar_exception.DuplicateElement(message) from exc
+                    print(message)
             else:
                 self._report_unprocessed_element(xml_child_elem)
 
@@ -2608,7 +2631,7 @@ def _read_numerical_value_specification_group(self,
         """
         xml_child = child_elements.get("VALUE")
         if xml_child is not None:
-            data["value"] = ar_element.NumericalValue(xml_child.text).value
+            data["value"] = self._read_number(xml_child.text)
 
     def _read_not_available_value_specification(self,
                                                 xml_element: ElementTree.Element

tests/xml/test_document.py

@@ -28,7 +28,7 @@ def test_read_write_document_with_multi_level_package_structure(self):
         writer = autosar.xml.Writer()
         xml = writer.write_str(document1, False)
         reader = autosar.xml.Reader()
-        document2 = reader.read_str(xml)
+        document2 = reader.read_str(xml, stop_on_error=True)
         self.assertEqual(len(document2.packages), 1)
         datatype_package: ar_element.Package = document2.find("/DataTypes")
         self.assertEqual(len(datatype_package.packages), 2)