Commit Graph

721 Commits

Author SHA1 Message Date
Petru Conduraru 4776bd7639 Atomic IFC file writes to prevent corruption on interrupted save (#4797)
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>
2026-07-07 14:22:44 +02:00
Petru Conduraru fa597536e1 IfcParse: drop ostringstream from format_double per review #7696
std::to_chars is locale-independent, so the ostringstream and imbue(locale)
are no longer needed. Build the REAL string with plain std::string operations.
Output is unchanged (verified in standalone compile: same shortest values, all
round-trip). Addresses review feedback on #8309.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-06 13:14:24 +02:00
Petru Conduraru ee2b357d74 IfcParse: serialize REALs with shortest round-trip form #7696
format_double formatted doubles with setprecision(max_digits10) (17 digits),
which padded clean values with noise: 0.0174532925199433 was rewritten as
0.017453292519943299 and 1.E-05 as 1.0000000000000001E-05. Every REAL in a file
changed on save, producing enormous diffs for anyone version-controlling IFC.
Use std::to_chars, which emits the shortest string that round-trips exactly
(like Python's repr), then keep the existing mantissa/exponent formatting.

Verified in a standalone compile of the exact function logic: the reporter's
values become 0.0174532925199433 and 1.E-05, 0.1 stays 0.1, and every tested
value (including a denormal) round-trips back to the identical double.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-06 13:14:24 +02:00
Petru Conduraru d2381ad6c6 IfcParse: don't strip delimiters from a single-character token #5683
asStringRef removes the first and last characters of a string, enumeration
or binary token to drop the delimiters, guarded only by !str.empty(). A
malformed single-character token (e.g. a bare '.' left when a fuzzer turns
'.PHYSICAL.' into '.)HYSICAL.') has length 1, so the first erase empties the
string and the second erase(str.begin()) runs on an empty string. That is
undefined behaviour: benign on a normal build, but it aborts (or throws
std::length_error from a later append) under a hardened libstdc++ with
_GLIBCXX_ASSERTIONS, which is why this file only segfaulted on the Fedora
build. Require at least two characters before stripping.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-06 12:28:17 +02:00
Petru Conduraru 5e539890f1 buildinfo: report the release version instead of a hardcoded fallback #8164
When ADD_COMMIT_SHA is off (the default for release tarballs), buildinfo.cpp
fell back to a hardcoded "0.8.0", so a 0.8.5/0.8.6 build reported 0.8.0 from
IfcConvert --version and in written file headers. Pass CMake's RELEASE_VERSION
(read from the VERSION file) to IfcParse as IFCOPENSHELL_VERSION_STRING and use
it as the fallback, mirroring how the branch/commit defines are handled. The
commit-sha build and the last-resort literal are unchanged.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-06 11:30:26 +02:00
Thomas Krijnen f3dcf0b539 Attempt at less double counting of inverses #7973 2026-06-28 13:42:36 +02:00
Thomas Krijnen 59383f5010 constexpr more cases to prevent gcc calling non-existing template overloads 2026-06-24 11:34:04 +02:00
Thomas Krijnen a9d6776875 header.assign() helper 2026-06-24 11:12:21 +02:00
Thomas Krijnen 7c9df9f980 Don't erronously terminate on [SYN004] Non-entity type messages 2026-06-20 21:18:23 +02:00
Thomas Krijnen 3e7b739d8d Don't rely on typeid() naming in VariantArray 2026-06-14 20:27:49 +02:00
Thomas Krijnen ca99ef3af7 More changes to pass around logger to parse-related calls 2026-06-14 14:49:14 +02:00
Thomas Krijnen dcebf23af8 Workaround for header construction order 2026-06-12 11:41:11 +02:00
Thomas Krijnen 4d22a3fdb9 Enable retargeting of example schema 2026-06-12 11:04:18 +02:00
Thomas Krijnen 347a3c80bb More logger changes 2026-06-11 21:09:56 +02:00
Thomas Krijnen a7738eeb64 Pass around non-static logger instances and programmatic access to messages in-memory 2026-06-10 18:40:17 +02:00
Thomas Krijnen a751fb956d Introduce unique error codes 2026-06-10 18:40:17 +02:00
Bruno Postle 24a241addc Use version preprocessor guards for RocksDB unique_ptr API, retain unique_ptr internally 2026-06-05 14:22:07 +02:00
Bruno Postle 365be8fb52 Support RocksDB shared library and new unique_ptr DB::Open API
Some distributions (e.g. Fedora) ship only a shared RocksDB that exports
RocksDB::rocksdb-shared rather than RocksDB::rocksdb. The CMake target
selection now falls back to the shared target when the static one is absent.

Newer RocksDB also changed DB::Open and DB::OpenForReadOnly to take
std::unique_ptr<DB>* instead of DB**. IfcFile.cpp uses SFINAE tag dispatch
to build against both old and new APIs without version detection.
2026-06-05 14:22:07 +02:00
Bruno Postle bd264f1d85 Add missing standard library includes for self-sufficient headers
Fixes builds with newer GCC/libstdc++ that no longer provide <cstdint>,
<cstring>, <cfloat>, <memory>, <algorithm> etc. transitively. Also
disambiguates visit<> calls in taxonomy.h with the full namespace and
casts the character value in IfcCharacterDecoder to uint32_t to silence
ambiguous overload warnings.
2026-06-05 08:54:27 +02:00
Richard Brice c39fe6e8a3 Fixes bug in addRelatedObject<> for IfcRelReferencedInSpatialStructure 2026-04-23 08:40:05 -07:00
Thomas Krijnen e7db239647 inverse access in schema 2026-04-10 21:46:39 +02:00
Thomas Krijnen 3fbf01f446 partial revert of 24acfea 2026-04-08 13:48:23 +02:00
Thomas Krijnen 24acfeaf45 Thinking indicator under chat 2026-04-03 10:59:09 +02:00
ssg3d 7b4889d2ec Update IfcParse.cpp
IfcOpenshell read file, and write file without changes. This round trip introduces truncation noise. It should not hurt to increase the precision to keep this clean.
2026-02-13 10:03:05 +01:00
Andrej730 d41985f23a format cmake files 2025-12-22 18:44:42 +05:00
Andrej730 deb6243ce5 cmake - use INCLUDES DESTINATION for includes paths export
As it's more idiomatic.
2025-12-15 19:02:06 +05:00
Andrej730 d5210e0d75 cmake export - propagate HAS_SCHEMA definitions to exported IfcParse target 2025-12-12 20:16:14 +05:00
Andrej730 d303acd660 cmake export - propagate IFOPSH_WITH_ROCKSDB to cmake config
Apparently, cmake isn't exporting definitions defined globally `add_definitions`, so creating interface target to combine rocksdb and our definitions.
2025-12-12 20:16:13 +05:00
Andrej730 01f97a5860 IfcParse - export include path 2025-12-12 19:30:23 +05:00
Andrej730 5a0f8d30cf cmake - install headers using PUBLIC_HEADER, deprecate INCLUDEDIR 2025-12-12 19:30:21 +05:00
Andrej730 92964fb5ef cmake - option to build only common schemas 2025-12-12 19:30:21 +05:00
Andrej730 2627052b55 cmake - export IfcOpenShellConfig.cmake 2025-12-11 19:55:41 +05:00
Andrej730 6a0caa7989 cmake - fix missing xml.lib in debug builds #7413
After 5d56025 we were using `LibXml2::LibXml2` instead of `LIBXML2_LIBRARIES` for linking, but this target was missing debug variants leading to issues during linking.
2025-12-04 17:44:13 +05:00
Andrej730 1468bf3c97 parse_ifcxml - remove redundant ifdef WITH_IFCXML
The entire file is already wrapped in `WITH_IFCXML`, so no need to do it again for some block.
2025-12-02 19:55:56 +05:00
Andrej730 a5d0187343 ifcparse cmake - fix missing ifcxml symbols in python wrapper #7406 2025-12-02 19:55:56 +05:00
Andrej730 67421c533c ifcparse cmake - make compile definition more explicit #7406 2025-12-02 19:55:56 +05:00
Esteban DUGUEPEROUX 083fab5255 cmake: Use standard install way
For specific install customization, use CMAKE_INSTALL_BINDIR, CMAKE_INSTALL_LIBDIR and CMAKE_INSTALL_INCLUDEDIR variables.

See-Also: https://cmake.org/cmake/help/latest/guide/tutorial/Installation%20Commands%20and%20Concepts.html
2025-11-30 11:11:51 +01:00
Esteban DUGUEPEROUX 81bf187ee6 cmake: put IFCXML_SUPPORT related config in IfcParse cmake config file 2025-11-30 11:05:46 +01:00
Thomas Krijnen 9d437b5bc3 Mark simple type instance file ownership #7226 2025-11-25 09:19:28 +01:00
Thomas Krijnen 461404cd45 Don't terminate parsing on invalid entity names 2025-11-23 11:16:47 +01:00
Esteban DUGUEPEROUX 5d56025014 cmake: Separate ifcparse build in its own CMakeLists.txt file 2025-11-10 09:42:51 +01:00
Thomas Krijnen 7cb1f07a0b hacky way to retain attribute too many messages on header entities in validate.py 2025-10-25 21:12:12 +02:00
Thomas Krijnen bdb52dcffd Pass header entity types during pass to resolve empty list type 2025-10-25 20:29:41 +02:00
Thomas Krijnen 1594b9505c Don't emit Parsing terminated on benign files by not reading past stream in skipComment() 2025-10-25 20:14:02 +02:00
Thomas Krijnen fa46fe0cf0 Typos 2025-10-25 15:30:36 +02:00
Thomas Krijnen e82ca0e932 Less back and forths by returning semi count in page 2025-10-25 15:08:21 +02:00
Thomas Krijnen 2e951eb952 Whitespace handling now that token strings are built on the fly 2025-10-25 13:55:15 +02:00
Thomas Krijnen bcb5fd7810 Missing return type 2025-10-24 13:32:18 +02:00
Thomas Krijnen 40c5145a13 Address some omissions in the FileReader rewrite regarding unicode paths on windows 2025-10-24 13:26:27 +02:00
Thomas Krijnen 0494bd9677 Option to bypass storing types when opening model 2025-10-24 12:06:59 +02:00