Remove obsolete DIRECTC_EXPECTED_FAILURES entries

[krystophny]

User description

Summary

• Remove 19 examples from DIRECTC_EXPECTED_FAILURES that now pass in Direct-C mode
• Only issue261_array_shapes and keep_single_interface actually fail
• Add missing issue306_allocatable_realloc to CMakeLists.txt

Testing

Manually tested all 21 examples in DIRECTC_EXPECTED_FAILURES with DIRECTC=yes:

• 19 examples now PASS
• 2 examples still FAIL (kept as expected failures)

Base

Stacked on jameskermode#346

---

PR Type

Enhancement, Bug fix, Tests

---

Description

• Fix segfault with -fdefault-integer-8 by using integer(c_int) for handles

  • Ensures consistent size between wrapper generation and runtime
  • Applies to all handle arrays in derived type wrappers

• Add support for pointer arrays in Direct-C mode

  • Removes skip logic that silently ignored pointer arrays
  • Adds pointer guard checks alongside allocatable checks

• Improve test coverage and fix test infrastructure

  • Add missing tests.py files for errorbinding example
  • Fix memory leak detection test in issue227_allocatable
  • Add tests for complex types and pointer arrays
  • Improve callback test documentation and fix segfault issue

• Enhance Direct-C code generation

  • Add character type setter support for module variables
  • Fix polymorphic type assignment with select type in Direct-C mode
  • Add numpy array API symbol for fortranobject.c compatibility
  • Fix deferred binding handling for abstract types

• Reorganize and sort example tests in build system

  • Sort examples alphabetically in CMakeLists.txt and Makefile
  • Add DIRECTC_EXPECTED_FAILURES list for known limitations
  • Improve test reporting with pass/fail/expected-fail tracking

---

Diagram Walkthrough

flowchart LR
  A["Integer Scalar Args<br/>in Direct-C Mode"] -->|"Convert to<br/>integer(c_int)"| B["Handle Arrays"]
  B -->|"Immune to<br/>-fdefault-integer-8"| C["Consistent Size"]
  D["Pointer Arrays<br/>in Derived Types"] -->|"Add Guard Check<br/>associated()"| E["Direct-C Support"]
  E -->|"Wrap with<br/>select type"| F["Polymorphic Assignment"]
  G["Test Infrastructure"] -->|"Add Missing Tests<br/>& Fix Bugs"| H["Better Coverage"]
  C -->|"Fixes"| I["Segfault Fix"]
  F -->|"Enables"| I
  H -->|"Validates"| I

Loading

File Walkthrough

	Relevant files
Bug fix	12 files f90wrapgen.py Convert integer scalars to integer(c_int) in Direct-C        +142/-25 transform.py Set wrapper_type to integer(c_int) for derived types          +22/-3    pywrapgen.py Add _setup_finalizer for abstract classes and factory returns +19/-1    main.py Fix PyInit function name derivation for Direct-C modules  +13/-6    run.py Fix assertion bug comparing wrong variables                            +1/-1      tests.py Fix callback test and document f2py callback bug                  +13/-9    Makefile Clean build directory and add fault handler                            +4/-3      Makefile Fix test target to use correct f2py build                                +1/-1      Makefile Fix clean target for build directory                                          +2/-1      Makefile Add f90wrap*.f90 to clean target                                                  +1/-1      Makefile Use F90 instead of FC for compilation                                        +1/-1      Makefile Fix types.py shadowing issue in test                                          +3/-2
Enhancement	3 files __init__.py Remove pointer array skip logic and fix deferred bindings +7/-15    module_helpers.py Add character type setter for module variables                      +110/-0  module_helpers_array.py Return handle in array helper tuple for Direct-C                  +2/-1
Configuration changes	7 files CMakeLists.txt Sort and reorganize example test list                                        +44/-25  Makefile Sort examples and add DIRECTC_EXPECTED_FAILURES tracking  +82/-34  Makefile.meson Delegate to regular Makefile for callback build                    +6/-13    Makefile.meson Add Makefile.meson for f2py_string_input example                  +8/-0      Makefile.meson Add Makefile.meson for array_shapes example                            +6/-0      Makefile.meson Add Makefile.meson for nested functions example                    +9/-0      Makefile.meson Add Makefile.meson for pointer_warning example                      +6/-0
Tests	8 files tests.py Add minimal test for errorbinding example                                +15/-0    tests_pkg.py Add package mode test for errorbinding example                      +15/-0    run.py Improve memory leak detection with multi-round testing      +32/-9    run.py Add gc.collect() calls to ensure object destruction            +3/-0      tests.py Add test for complex type wrapping                                              +19/-0    tests.py Add test for pointer array handling                                            +28/-0    Makefile Update test to verify pointer array wrapping                          +12/-10  test_directc.py Add test for character setter code generation                        +13/-0
Additional files	2 files cback.f90 +1/-1      Makefile.meson +8/-0

---

[qodo-code-review]

You are nearing your monthly Qodo Merge usage quota. For more information, please visit here.

PR Compliance Guide 🔍

Below is a summary of compliance checks for this PR:

Security Compliance
⚪	Integer truncation overflow Description: The new generated C character setter truncates PyBytes_GET_SIZE() (a Py_ssize_t) into an int (value_len), which can overflow for large inputs and lead to incorrect allocation size followed by memcpy, creating a potential heap buffer overflow / memory corruption condition when extremely large bytes/str values are passed. module_helpers.py [200-263] Referred Code def _write_character_setter(gen: DirectCGenerator, helper: ModuleHelper, helper_symbol: str) -> None: """Helper for character type setters.""" gen.write("PyObject* py_value;") gen.write(f"static char *kwlist[] = {{\"{helper.element.name}\", NULL}};") gen.write( "if (!PyArg_ParseTupleAndKeywords(args, kwargs, \"O\", kwlist, &py_value)) {" ) gen.indent() gen.write("return NULL;") gen.dedent() gen.write("}") gen.write("if (py_value == Py_None) {") gen.indent() gen.write( f'PyErr_SetString(PyExc_TypeError, "Argument {helper.element.name} must be str or bytes");' ) gen.write("return NULL;") gen.dedent() gen.write("}") gen.write("PyObject* value_bytes = NULL;") gen.write("if (PyBytes_Check(py_value)) {") ... (clipped 43 lines)
Ticket Compliance
⚪	🎫 No ticket provided Create ticket/issue
Codebase Duplication Compliance
⚪	Codebase context is not defined Follow the guide to enable codebase context checks.
Custom Compliance
🟢	Generic: Comprehensive Audit Trails Objective: To create a detailed and reliable record of critical system actions for security analysis and compliance. Status: Passed Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Secure Error Handling Objective: To prevent the leakage of sensitive system information through error messages while providing sufficient detail for internal debugging. Status: Passed Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Secure Logging Practices Objective: To ensure logs are useful for debugging and auditing without exposing sensitive information like PII, PHI, or cardholder data. Status: Passed Learn more about managing compliance generic rules or creating your own custom rules
⚪	Generic: Meaningful Naming and Self-Documenting Code Objective: Ensure all identifiers clearly express their purpose and intent, making code self-documenting Status: Unused import: The newly added import sys appears unused, which may reduce clarity and self-documenting quality. Referred Code import sys Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Robust Error Handling and Edge Case Management Objective: Ensure comprehensive error handling that provides meaningful context and graceful degradation Status: Potential length truncation: The character setter converts PyBytes_GET_SIZE() to int, which could truncate very large inputs and lead to incorrect allocation/behavior. Referred Code gen.write("int value_len = (int)PyBytes_GET_SIZE(value_bytes);") gen.write("char* value = (char*)malloc((size_t)value_len + 1);") gen.write("if (value == NULL) {") Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Security-First Input Validation and Data Handling Objective: Ensure all data inputs are validated, sanitized, and handled securely to prevent vulnerabilities Status: Large input handling: The new character setter accepts arbitrary-sized str/bytes and allocates via malloc using a length stored in int, which may be unsafe without explicit maximum-length checks. Referred Code gen.write("if (py_value == Py_None) {") gen.indent() gen.write( f'PyErr_SetString(PyExc_TypeError, "Argument {helper.element.name} must be str or bytes");' ) gen.write("return NULL;") gen.dedent() gen.write("}") gen.write("PyObject* value_bytes = NULL;") gen.write("if (PyBytes_Check(py_value)) {") gen.indent() gen.write("value_bytes = py_value;") gen.write("Py_INCREF(value_bytes);") gen.dedent() gen.write("} else if (PyUnicode_Check(py_value)) {") gen.indent() gen.write("value_bytes = PyUnicode_AsUTF8String(py_value);") gen.write("if (value_bytes == NULL) {") gen.indent() gen.write("return NULL;") gen.dedent() ... (clipped 24 lines) Learn more about managing compliance generic rules or creating your own custom rules

Compliance status legend

🟢 - Fully Compliant
🟡 - Partial Compliant
🔴 - Not Compliant
⚪ - Requires Further Human Verification
🏷️ - Compliance label

[qodo-code-review]

You are nearing your monthly Qodo Merge usage quota. For more information, please visit here.

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
General	Add error checking for PyLong allocations In _write_array_helper_return, add error checking after PyLong_FromLong and PyLong_FromLongLong calls to handle potential allocation failures and prevent crashes. f90wrap/directc_cgen/module_helpers_array.py [125-136] gen.write("PyObject* result = PyTuple_New(4);") gen.write("if (result == NULL) {") gen.indent() gen.write("Py_DECREF(shape_tuple);") gen.write("return NULL;") gen.dedent() gen.write("}") -gen.write("PyTuple_SET_ITEM(result, 0, PyLong_FromLong((long)nd));") -gen.write("PyTuple_SET_ITEM(result, 1, PyLong_FromLong((long)dtype));") +gen.write("PyObject* nd_obj = PyLong_FromLong((long)nd);") +gen.write("PyObject* dtype_obj = PyLong_FromLong((long)dtype);") +gen.write("PyObject* handle_obj = PyLong_FromLongLong(handle);") +gen.write("if (nd_obj == NULL || dtype_obj == NULL || handle_obj == NULL) {") +gen.indent() +gen.write("Py_XDECREF(nd_obj); Py_XDECREF(dtype_obj); Py_XDECREF(handle_obj);") +gen.write("Py_DECREF(shape_tuple); Py_DECREF(result);") +gen.write("return NULL;") +gen.dedent() +gen.write("}") +gen.write("PyTuple_SET_ITEM(result, 0, nd_obj);") +gen.write("PyTuple_SET_ITEM(result, 1, dtype_obj);") gen.write("PyTuple_SET_ITEM(result, 2, shape_tuple);") -gen.write("PyTuple_SET_ITEM(result, 3, PyLong_FromLongLong(handle));") +gen.write("PyTuple_SET_ITEM(result, 3, handle_obj);") gen.write("return result;") Apply / Chat Suggestion importance[1-10]: 8 __ Why: This is a valid and important error handling improvement for the C-API code, preventing potential segfaults from memory allocation failures.	Medium
	Ensure consistent attribute checking pattern In write_arg_decl_lines, modify the check for the "optional" attribute to be a string-based check for consistency and to prevent potential failures if attributes are not simple strings. f90wrap/f90wrapgen.py [440-447] if self.direct_c_interop: arg_type = str(arg_dict["arg_type"]).strip().lower() is_integer_scalar = arg_type == "integer" and not any( str(attr).startswith("dimension(") for attr in getattr(arg, "attributes", []) ) - if is_integer_scalar and "optional" not in arg.attributes: + if is_integer_scalar and not any(str(attr).startswith("optional") for attr in getattr(arg, "attributes", [])): arg_dict["arg_type"] = "integer(c_int)" direct_c_integer_scalars.append(arg.name) Apply / Chat Suggestion importance[1-10]: 6 __ Why: The suggestion improves code robustness and consistency by using a safer method to check for the 'optional' attribute, preventing potential bugs.	Low
More