create_shape() returns a Python-owned Element (SWIG_POINTER_OWN in the
boost::variant out typemap). Its .geometry property calls Element::geometry(),
which returns a reference into the element's boost::shared_ptr<Representation>
_geometry member. SWIG wraps that reference as a non-owning pointer, so the
returned Triangulation/BRep/Serialization proxy does not keep the element alive.
When a caller keeps only .geometry (e.g. create_shape(s, e).geometry) and drops
the parent element, Python garbage-collects the element, destroying its
shared_ptr and freeing the underlying representation. Subsequent reads of
verts/faces then return freed memory: empty or implausible float/int garbage,
non-deterministically depending on GC and allocator timing. This is silent data
corruption, not a crash, and has bitten users since 2020.
Fix: in the TriangulationElement/SerializedElement/BRepElement pythoncode, wrap
the geometry getter so the returned geometry stores a backreference to its
owning element (result._parent = self). This makes the parent's lifetime at
least as long as the geometry's, automatically and transparently, so no caller
has to remember to hold the element. This is aothms's suggested backreference,
applied generically in the binding rather than left as a workaround.
Reproduced deterministically (washBasin fixture): before, verts len 0 vs 133500
across repeated GC-pressure runs; after, 133500 every run for all three element
types. test_create_shape passes; no regressions.
Note: tree.select_ray()'s ray_intersection_result (2024 follow-up in #1124) is a
separate ownership mechanism (std::vector element reference + std::array member
pointer) and is left as follow-up scope.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
(cherry picked from commit 824c1fc280)
Follow-up to the scalar-only fix in #8754, per aothms's direct request on
that PR ("Please do make all int types consistent") and his own original
2023 design intent on issue #3058 ("make all integers (incl. schema
namespaces) an int64_t"). Widens the remaining inconsistent spots now that
compatibility isn't a constraint on this v0.9 branch:
- Integer aggregates (IfcTriangulatedFaceSet.CoordIndex and similar
List<int> attributes), including the SWIG to_vec_int/to_vec_vec_int
helpers, which previously silently truncated via static_cast<int> on the
Python-set path - the same bug class as the original scalar issue.
- The schema code generator (express/mapping.py's integer type mapping),
and all 12 generated schema header/source pairs regenerated to match, so
every schema-typed getter/setter (e.g. IfcOwnerHistory::CreationDate) is
int64_t end to end, not just the dynamic attribute-value path.
Instance/reference identifiers (STEP #123 ids) are deliberately left at
32-bit: they're a file-local index into internal maps, not an EXPRESS
domain value an application chooses, and no realistic STEP file has
billions of entities. The lexer's Token_IDENTIFIER parsing still funnels
through a 32-bit int for this reason - flagged as a known, low-risk gap
rather than fixed, since fixing it would mean touching indexing/hashing
code for no realistic benefit.
Verified: original PR's round-trip tests extended with aggregate cases
(IfcTriangulatedFaceSet.CoordIndex, InnerCoordIndices) at 64-bit boundary
values, in memory and through STEP text, IFC2X3 and IFC4. A standalone C++
program exercising the generated schema API directly (Ifc4::IfcOwnerHistory
::setCreationDate/CreationDate, IfcTriangulatedFaceSet::setCoordIndex/
CoordIndex) confirms int64_t end to end, bypassing SWIG. Full build
(BUILD_IFCGEOM, WITH_OPENCASCADE, BUILD_IFCPYTHON, IFC2X3+IFC4) clean.
test/util/test_attribute.py and test_file.py pass unchanged.
This contribution was produced with the assistance of an AI coding tool.
Setting an IfcInteger/IfcTimeStamp typed attribute (e.g. IfcOwnerHistory.CreationDate)
outside the signed 32-bit range corrupted the value instead of raising, since the
Python wrapper's set_attribute_value_py() truncated it with a plain static_cast<int>
before handing it to the C++ storage. Unix timestamps before 1901-12-13 or after
2038-01-19 silently wrapped around (e.g. 3000000000 became -1294967296) rather than
being rejected or stored correctly. Fixes#3058, equivalent to PR #8683 but ported to
this branch's rewritten ifcparse (snake_case files, variant_array/instance_data
storage, SWIG PyObject-based attribute setter) instead of the old IfcEntityInstanceData
sources, which no longer exist here.
The scalar slot of the attribute variant (Argument_INT) becomes int64_t. Integer
aggregates (Argument_AGGREGATE_OF_INT, e.g. CoordIndex) and instance/reference
identifiers stay 32-bit, since neither is the value that overflows here; this narrow
scope is kept on its own technical merits (aggregates and identifiers were never the
source of the bug, and widening them would be a much larger, riskier change for no
benefit) even though aothms said compatibility isn't a concern on this v0.9-track
branch. express::Base::set_attribute_value promotes the schema-generated int to
int64_t at a single choke point, so the generated setters keep compiling unchanged.
The STEP lexer, writer, and SWIG wrapper (set_attribute_value_py, pythonize) are all
widened together, since widening only the Python-facing setter would have silently
wrapped the value on file write instead of raising.
Verified in a build (IFC2X3 and IFC4, BUILD_IFCGEOM off, no kernels): pre-1901,
post-2038, both 32-bit boundaries, and a 9e12 value all round trip exactly both in
memory and through STEP text serialization (write then reopen). A value outside the
64-bit range now raises a clean exception instead of corrupting data. Ordinary
in-range integers and integer aggregates (e.g. IfcTriangulatedFaceSet.CoordIndex) are
unaffected. The existing util/test_attribute.py and test_file.py suites pass
unchanged; test_entity_instance.py has 5 pre-existing failures unrelated to this
change (confirmed identical on an unfixed build of this branch, caused by a missing
get_info_2 binding and _patch_swig_comparisons never being implemented here).
Generated with the assistance of an AI coding tool.
The data-model branch's spf_header::set_file_description/name/schema take a
const shared_pointer_type& (an internal instance_data* storage handle). SWIG
wraps them and emits the alias unqualified into the global-scope wrapper,
which MSVC rejects (C2065 'shared_pointer_type': undeclared identifier). The
matching getters are already %ignore'd and re-exposed via %extend; the raw
setters are not a usable Python API, so ignore them the same way.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
file.write() streamed directly onto the target path, so a crash mid-write
left a truncated file with dangling STEP references. Serialize to a temp
file in the same directory, then atomically rename it onto the target.
- New IfcUtil::path::atomic_rename_file: std::rename on POSIX, MoveFileExW
with MOVEFILE_REPLACE_EXISTING on Windows. Unlike rename_file it never
unlinks the destination first, so there is no window where it goes missing.
- Fully in C++/swig (per aothms), so the FILE_NAME header is untouched: it
comes from the model header, not the output path (verified empirically).
- Temp lives next to the target so the rename stays on one filesystem.
- Stream is closed before the rename (Windows cannot move an open file).
- On any write error the temp is removed and the original target is intact.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
check_aggregate_of_type used an exact type comparison (element->ob_type ==
type_obj), so a numpy array was rejected because its elements are numpy scalars
(numpy.float64) rather than direct float instances. For the numeric types,
accept subclasses: PyFloat_Check for double (numpy.float64 subclasses float) and
PyLong_Check (excluding bool) for int. The SPF REAL vs INTEGER distinction is
kept, so a float is not accepted where an int is expected and vice versa.
This replaces the earlier Python-side walk() approach, which the maintainer
preferred not to take since walk() is removed in v0.9. Verified with a runtime
red-green (built as a shared lib, called via ctypes): the old check rejects
np.array([3.0, 4.0]) and the new one accepts it, plain lists still work, an int
list is still rejected where a REAL is expected, and bool is rejected for INTEGER.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Three coupled fixes that close the macOS bring-up loop:
## 1. ifcwrap: fix INSTALL_RPATH on Apple
The ifcopenshell_wrapper Python module had `INSTALL_RPATH "$ORIGIN"` set
for "NOT WIN32 AND NOT WASM_BUILD" — but `$ORIGIN` is a Linux ld.so
placeholder, not a macOS dyld one. macOS dyld doesn't expand it; it
bakes the literal string `$ORIGIN` into LC_RPATH, which resolves to
nothing at runtime. The wrapper's hard-link `@rpath/ifcopenshell
.document.rdb.dylib` then fails to load even though INSTALL(TARGETS …
LIBRARY DESTINATION "${python_package_dir}/ifcopenshell") above had
already dropped the plug-in dylib right next to the wrapper.
Split the rpath assignment: `@loader_path` on Apple (the dyld
equivalent of `$ORIGIN`), `$ORIGIN` elsewhere.
This is what b0ef47819 (the build_osx IFCOS_BUILD_PYTHON_WRAPPER=off
gate) was working around. The gate is removed below.
## 2. ifcviewer-minimal: stage IfcOpenShell + wgpu_native into the .app
IfcViewerMinimal.app was building on macOS via the cmake
`BUILD_BONSAIVIEWER → BUILD_BONSAIVIEWER_WGPU` promotion, but had no
bundle staging — Contents/Frameworks/ only contained the Qt
frameworks macdeployqt deposited, so the .app would refuse to start
("Library not loaded: @rpath/libwgpu_native.dylib").
Mirror what src/bonsaiviewer/CMakeLists.txt does for BonsaiViewer.app:
* Set INSTALL_RPATH to `@executable_path/../Frameworks` so the exe
knows where to look for @rpath/* deps.
* install(FILES) libwgpu_native.dylib into the bundle's Frameworks/
(globbed from WGPU_NATIVE_LIB_DIR rather than hard-coded so it
covers any future versioned name).
* install(CODE) staging block that copies every `*.dylib` from
<prefix>/lib/ into the bundle's Frameworks/, excluding the
geometry-writer plug-ins (same EXCLUDE regex as BonsaiViewer.app —
viewer doesn't need OBJ/glTF/DAE/STP/IGS/SVG/TTL export converters).
Same long-form rationale + caveats apply (macdeployqt doesn't follow
non-Qt @rpath deps, lib-prefixed core libs vs ifcopenshell.* plug-in
naming split, Linux's equivalent lives in build_rocky.yml workflow
bash via patchelf + stage_runtime_payload). See src/bonsaiviewer/
CMakeLists.txt for the full version.
## 3. build_osx.yml: drop IFCOS_BUILD_PYTHON_WRAPPER=off
With (1) fixed, the Python wrapper smoke test should pass again. The
gate goes away; the comment block in build_osx.yml is replaced with a
short note pointing at the ifcwrap rpath fix as the underlying change
that re-enables this.
Together, (1)+(2)+(3) close the standalone IfcViewerMinimal-on-macOS
gap (task #43) and re-enable IfcOpenShell-Python on the macOS arm64 CI.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Petru Conduraru