Addresses Rick Brice's PR review feedback on #7785.
Root cause: the AddElement dialog was manually stitching together
IfcAlignment layout containers and calling create_representation(),
bypassing the validated construction sequence in align_api.create().
This left IfcGradientCurve.BaseCurve potentially None and skipped
zero-length segment and stationing referent setup.
Changes:
- root/operator.py: alignment templates now call align_api.create()
directly (HORIZONTAL, GRADIENT, CANT) or _create_polyline_representation
(POLYLINE_2D/3D), then manually link the result to the Blender object via
tool.Ifc.link + tool.Collector.assign. The old generic core.assign_class
path is retained for all non-alignment templates unchanged.
- tool/alignment.py: remove create_representation_structure() — superseded
by the operator changes above.
- create_representation.py: revert the if layout_nest: guard; with
align_api.create() as the entry point the zero-length segment always
exists before create_representation is called.
- add_zero_length_segment.py: revert the BaseCurve None guard; the root
cause (gradient curve created without a base curve) no longer occurs.
- alignment/operator.py: guard _create_geometric_representation call so it
only runs when no curve representation exists yet; auto-invoke
create_alignment_by_pi after PI picker finishes if >=2 PIs are defined.
- util/file.py: fix StopIteration on short IFC template files
(next(ifc_file) → next(ifc_file, None) with break).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Two bugs prevented IfcAlignmentVertical segments from being added when
a geometric representation exists.
Bug 1 — _add_segment_to_curve.py:
Removed an unconditional `if not curve.is_a("IfcCompositeCurve")` guard
that was left over from when the function only supported horizontal
segments. The preceding if/elif/elif chain already validates the correct
curve type for each segment type; the redundant check always raised
TypeError for vertical (IfcGradientCurve) and cant
(IfcSegmentedReferenceCurve) segments.
Bug 2 — add_zero_length_segment.py:
Added a None guard before the recursive `add_zero_length_segment(file,
layout.BaseCurve)` call for IfcGradientCurve. When an IfcGradientCurve
is created without a BaseCurve (e.g. before a horizontal representation
exists), the recursive call previously crashed with AttributeError.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Wire IfcAlignment creation into Bonsai's Add IFC Element flow with
representation template options (Horizontal, Gradient, Cant, 3D/2D
Polyline). Adds create_representation_structure() to tool.Alignment
which creates layout containers, geometric representation, and
polyline placeholders. Guards create_representation against empty
layout nests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Ignore CLAUDE.md, CLAUDE.local.md, and .mcp.json so personal
Claude Code configuration (managed via private dotfiles repo)
doesn't pollute the shared repository.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
get_mapped_segments now looks for representations attached to alignment segments before using the more complex method of computing the index of segments in the composite curve.
updates segment_vertices to use get_mapped_segments
Wrap SAIKEI_OT_enter_pi_edit_mode with Bonsai's IfcStore transaction
system so IFC segment changes from applying PI edits are tracked and
undoable via Ctrl+Z.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Uses Rick Brice's new ifcopenshell.api.alignment.segment_vertices()
to extract PI positions from alignment segments via the C++ geometry
engine, replacing ~300 lines of hand-coded trig that only handled
LINE and CIRCULARARC. Now supports all segment types (CLOTHOID,
Helmert curves, etc.). Includes backward-compatible fallback for
IFC files without Axis/Segment representations.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Renamed PI and CC to TI and NI
Fixes handling of units (now works correctly with US and SI units)
Changed tests to use US feet units
Fixed documentation
Guidelines for external contributors using AI coding tools,
covering licensing, AI disclosure requirements, PR scope,
commit style, code formatting, and testing expectations.
Generated with the assistance of an AI coding tool.
Replace custom SAIKEI_OT_pick_pi_from_viewport modal with a
PolylineOperator subclass, reusing Bonsai's proven polyline
infrastructure (same base class as wall/slab/profile drawing).
Gains: snapping, numeric D/A/X/Y input, axis locking, angle
locking, measurement display, undo-last-point, status bar hints.
Remove PIPickerDecorator (replaced by PolylineDecorator).
Fix EN string formatting in UI list display rows.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Previously, there was a dance between invoke, execute, and draw. This
can probably be resolved, but is a high-risk for undo bugs. This
simplifies the logic flow to just a traditional _invoke -> _execute.
I add a new feature test to at least make sure it does something, and
this also fixes the segfault in tool tests as it no longer requires the
launching of the file browser.
In my previous commit, I mistakenly believed that there was an API change from
snap_angle_increment to snap_angle_increment_3d
But since the feature was introduced in blender 4.2 the setting is called
snap_angle_increment_3d.
Previously it was implemented inline. This now implements it as a
tool.Blender function with tests. Also the previous tests didn't
actually run and weren't actually testing any tools despite being in a
tool tests.
IfcSectionedSurface and IfcSectionedSolidHorizontal both of CrossSectionPositions attributes which are lists of IfcAxis2PlacementLinear. The implementation of each class used its own bespoke mapping of IfcAxis2PlacementLinear, which were identical to each other and slightly different than IfcAxis2PlacementLinear. Now the two sectioned classes use the one and only mapping for IfcAxis2PlacementLinear
* Linked IFC projects enhancement (multiple links to same project file)
- Implement link management system using UUIDs as identifiers to support multiple links to the same IFC file
- Add georeferencing compatibility detection and UI display (NONE, NOT_COMPATIBLE, PARTIAL_COMPATIBLE, FULL_COMPATIBLE)
- Support for duplicate link creation with Shift+D shortcut and automatic position offset
- Add false origin and project north calculation from 3D cursor for MANUAL mode
- Only store one cache per file, regardless of the amount of links
- Prevent duplicate links based on filepath and position comparison
- Improve error handling for missing files and loading failures
- Update tests
* Remove duplicate georef UI
I try to avoid duplicate UI (especially for one that can be as
sophisticated as georef - e.g. missing is WCS) as it means double the
code, double the tests, potential user confusion. BTW the note about
vertical datum isn't quite accurate as it may be included in the CRS
definition so vertical datum is optional.
* Remove depsgraph_update_post handler for update_link_ui_on_transform as per core developer feedback
* Move get_projected_crs to geolocation module
* Refactor get_projected_crs to simplify as per core developer feedback
* Remove unused import of bonsai.tool from project module
* Use IfcDocumentInformation per linked file and IfcDocumentReference for locaiton information
* Refactor SaveBlendMetadataFile operator to remove try-except blocks and remove linked projects collections since they are recreated by bonsai
* Cleanup removing empty collection instances for linked models in metadata.blend file and call determine_georeferencing_compatibility on link reload
* Add locking mechanism for linked models and update UI to reflect lock status
* Update logic that track IFC to execute_ifc_duplicate_operator instead of having it in execute() which does not track IFC undo/redo
* Refactor link handling to use get_link_empty_handle and set_link_empty_handle methods which in turn use the standard blender-ifc integrations patters (tool.Ifc.get_object(doc_reference) and tool.Ifc.link(doc_reference, empty_handle)
* remove operator.DuplicateLink and move it to tool.Project.duplicate_link()
* Refactor link handling to use sequential identifiers (no need for STEP ID DocRef)
* Refactor IFC linking logic to handle cases without a parent IFC file loaded. Firts link flase origin becomes parent origin
* Lock should not affect selection.
This makes it consistent with grid / spatial lock, and also toggle
selectability is already implemented.
* Remove unnecessary check for loaded library as Blender seems to do this internally already
* Rename util to get_crs because in IFC4X3 you can also have geographic CRS not just projected
* Remove unnecessary call to determine_georeferencing_compatibility
This function is already always called prior to calculate_link_position
so shouldn't be called here. It's also a very expensive function: as it
currently stands, just to link a single IFC, ifcopenshell.open() is
called 3 times. This reduces it to 2.
* Store CRS as metadata for linked models, and compare metadata when indicating georeferencing compatibility
Previously, to check georeferencing compatibility, ifcopenshell.open()
was used. When linking large models, this adds considerable time and
memory usage. This instead captures the georef as standard metadata in
our .cache.json. This now reduces the ifcopenshell.open() calls back
down to only 1 as necessary (see previous commit).
* Use link index instead of link name to fetch link collection item
Link name runs into issues with name uniqueness. This is why you created
a function for "get next link ID". After this refactoring, we can no
longer worry about uniqueness and that function may be removed.
* Simplify reloadlink into just unload and reload (with cache disabled)
This function should not be responsible for editing any data.
* Remove unnecessary get_next_link_id as names no longer need uniqueness
This now frees up the name variable to track a more meaningful, human
name like IfcDocumentInformation's Name attribute.
* Rewrite get / set link_empty_handle to just use the link directly
This prevents needless logic to fetch the link and also removes issues
related to duplicate names.
* Temporarily remove logic in prop callback
Right now, pretty much all the logic is done in a prop callback. In
general logic in prop callbacks should be minimised, since it's hard to
test and easily triggered as a domino effect of another change, and may
also impact undo/redo.
* Remove code that unnecessarily removes cache
This code removes cache, which means any project unlinking an IFC auto
clears the cache for any other project which doesn't make sense, and
also breaks the ability to readd it quickly.
* Rewrite link, unlink, load, and unload IFC
There were a few issues tackled here:
- Operators that change any IFC data must use tool.Ifc.Operator and
_execute, otherwise undo/redo will break. That's one of the risks of
using prop callbacks, as it is not explicit when an IFC edit happens.
- The usage of IfcDocumentReference was not correct. The Location
should store the URL, _not_ the position. The position should be in the
Identification attribute.
- The URL was stored in IfcDocumentInformation location, which does not
work in IFC2X3. There are a few changes here to make it IFC2X3
compatible.
- Generally move logic in operators, not prop callback.
* Remove restriction around manual mode.
Users should be able to use manual mode if they want.
* Restore AUTOMATIC mode to identical behaviour to file open
This is the first step to reusing cache files agnostic of the host.
* Revert tests for a fresh start for updating tests
* Revert "test_feature - clean up .ifc.cache. files after test was executed"
This reverts commit 99ae768ddf.
* Update tests and reimplement calculations for matrix of empty handle
Previously, the empty would always be placed at the origin, unless a
"position" offset was present. This is a problem, because the "position"
is simply a local offset relative to the Blender cache! If the cache was
regenerated, the offsets would be outdated. Also, the cache appeared in
different locations depending on the false origin mode, so the offset
would mean different things to different people.
Instead, a more robust method is:
1. When you link a file, a Blender cache is generated. The Blender
origin of this cache is arbitrary! It depends on the user's false origin
mode and is purely a Blender session specific thing.
2. When you load a link, a link is _always_ loaded into the correct
location with regards to IFC global coordinates. All math is done from
the perspective of IFC.
3. If you choose to transform (move / rotate / scale!?) this link from
its correct location, that gets recorded as a 4x4 transformation matrix.
Note: I haven't implemented this properly yet.
Tests all pass, with a minor modification to the new behaviour that
false origin mode now won't affect the location it ends up in, only the
generation of the cache.
* Remove arbitrary convention around display name
Not needed anymore now that A/M/D is a detail and not significant on
actual coordinates, and also that the UUID is no longer needed.
* Simplify implementation of loading linked models when opening an IFC
* Move link matrix calculation from operator to tool for reuse
* Implement editing link location and calculation of transformation matrix
I changed my mind on the is_locked thing, since it isn't clear to the
user that locking need to be done to save changes.
* Remove old is_locked, prop update callback no longer needed (dedicated operator instead), remove old calculation code
* Simplify code related to placed_as_per_georef
* For now, simple skip for duplicate / delete
IMO duplicate / delete / move a link are very rare and explicit
operations.
* Update tests
* Remove host_model coordinate data as cache is no longer host model dependent
* Move icons outside list because there are too many
* Minor tweaks
---------
Co-authored-by: Dion Moult <dionmoult@gmail.com>
Co-authored-by: Dion Moult <dion@thinkmoult.com>
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.
Replace hardcoded 5-degree angle snapping with Blender's
snap_angle_increment setting in create_wall_from_2_points()
and create_profile_from_2_points().
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Replace hardcoded 5-degree angle snapping with Blender's
snap_angle_increment setting in handle_lock_axis() for:
- Initial angle rounding when locking axis (A key)
- Angle rounding and increments on Shift+Wheel scroll
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Replace hardcoded 5-degree angle snapping with Blender's
snap_angle_increment setting in calculate_distance_and_angle().
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This function retrieves the angle snap increment from Blender's
tool_settings.snap_angle_increment property, which was added in
Blender 4.2. This allows users to configure the angle snap value
through Blender's native UI instead of using hardcoded values.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add clear_layout_segments API to remove segments while preserving alignment ID
- Modify exit_pi_edit_mode to edit segments in-place instead of delete+recreate
- Fix curve visualization by using create_shape for segment vertices
- Fix evaluate_segment validation to handle negative-length curve segments
This prevents "Active alignment no longer exists" errors when editing PIs
and properly renders circular arcs regardless of turn direction.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Previously, sort, reverse list, and join functionality was implemented
as special cases in Bonsai itself. Given that it has usecases
(especially in material lists, but any sort of list applies) I've moved
this function into the IOS formatting language.
The IOS formatting language previously wasn't capable of this, but the
awesome addition by @falken10vdl made the formatting language accept
queries inline, so that means it can handle lists. I also added tests
for all the new functions and expression syntax (+-*/ operators).
I simplified the code that gets the evaluated text literal - previously
it seems to call format() multiple times.
Previously, copy attribution was coupled with text editing. This meant
that you couldn't just do something like change the font or alignment
without also affecting literals. Now like most apps you can just select
bunch of text and change font size etc, using the same UI look and feel
that copying attribute has when editing attributes.
This refactor also removes the need for explicit props tracking each
possible attribute to copy, and the settings collection group. Bulk
applying is now done in core with no calls to UI.
Turned out `aud` module we had in our makefile had nothing to do with Blender built-in `uad` module 🫣
So no need to install anything from PyPI since this module is generally available in Blender
This isn't complete yet, but it hopefully demonstrates a preferred
implementation:
* Logic in core, not operator
* Loop done in core, without needing to call other core functions, so
the overhead of enabling and disabling editing per object is removed. No
more Blender logic, just straight editing in IFC.
* Reuse existing function to grab text attributes instead of
reimplementing it twice.
* Remove dead code, there seems to be a function
apply_to_selected_objects which was completely unused and duplicated
code twice.
When using G key to move PIs, Blender's transform modal consumes
the Enter key. Adding Space as an alternative lets users apply
changes more easily. Also adds Numpad Enter support.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Implements the ability to edit alignment PI (Point of Intersection)
positions after creation using Blender's standard transform tools:
- Back-calculate PI positions from existing IFC alignment segments
- Create temporary EMPTY objects at PI locations for editing
- Visual feedback via PIEditDecorator (yellow tangent lines, HUD)
- Modal operator handles G key movement, Enter to apply, Escape to cancel
- Regenerates alignment with new PI positions on apply
- Handles edge cases: single-segment, tangent-only, undo during edit
Architecture follows Bonsai patterns:
- Core layer: Business logic orchestration (enter/exit_pi_edit_mode)
- Tool layer: Math, IFC, and Blender implementations
- UI layer: Modal operator with PASS_THROUGH for standard transforms
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This file contains project-specific context for Claude Code sessions.
It should NOT be pushed to origin (protected by pre-push hook).
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add visual feedback during PI placement with PIPickerDecorator:
- Yellow tangent lines connecting placed PIs
- Rubber band line from last PI to cursor position
- Green circle markers at each PI location
- HUD text showing instructions and PI count
Follows Bonsai's established decorator pattern with GPU draw handlers.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Replace segment empty objects with actual curve geometry so that selecting
a segment in the Outliner highlights the corresponding line/curve in the
viewport.
Changes:
- Add get_segment_vertices() using IfcOpenShell's evaluate_segment() to
sample points along individual segments via the geometry engine
- Replace _create_segment_empty with _create_segment_curve that creates
Blender CURVE objects with actual geometry
- Remove single HorizontalCurve in favor of per-segment curves
- Supports all segment types (LINE, CIRCULARARC, CLOTHOID, spirals, etc.)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Zero-length segments are required by IFC to mark alignment ends but should
be invisible to users. This change:
- Adds helper methods to detect zero-length and empty layouts
- Silently skips geometry generation for empty alignments (no error messages)
- Excludes zero-length segments from Outliner display
- Uses separate visible segment counter for consistent naming
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Fix modal operator to use absolute mouse coordinates converted to 3D
viewport region space, instead of event.mouse_region_x/y which are
relative to whichever region received the event
- Store 3D viewport area, region, and region_data references in invoke()
for consistent raycasting throughout modal operation
- Add coordinate transformation methods (blender_to_ifc_coordinates and
ifc_to_blender_coordinates) for projects with geospatial Blender offsets
- Transform alignment curve vertices from IFC global to Blender local
coordinates when has_blender_offset is enabled
- Add try/except for piecewise-step-size geometry setting in util.py
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Simplifies the Horizontal Alignment panel by removing the Import
Alignment CSV button and the PI Details submenu that displayed
when selecting rows in the PI Editor list.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Fix issue where selected text annotations remained in editing mode after
applying changes. Now properly restores original editing state for each
selected object.
Add RuntimeError handling for IfcOpenShell versions that don't support
the piecewise-step-type setting in geometry generation.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Git checkout with glob patterns (*.ifc) doesn't work on Windows.
This change:
- Expands glob via git ls-files and checks out files individually
- Skips symlink recreation if they already exist and are valid
- Refreshes git index before checkout to recognize deleted files
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Replace manual per-segment geometry creation with IfcOpenShell's
built-in generate_vertices() utility. This provides automatic support
for all curve types (CLOTHOID, spirals, etc.) and removes ~75 lines
of manual geometry code.
Changes:
- Add create_curve_from_representation() using IfcOpenShell geometry engine
- Add _create_segment_empty() for segment selection without geometry
- Update create_objects_for_layout_segments() to use new methods
- Delete manual geometry methods: create_object_for_segment(),
_create_line_segment(), _create_arc_segment()
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Sync children was a bit odd because it's not actually an "array
parameter" per se, just a way to regenerate. It's now an operator.
There was a deeper issue I encountered where the way arrays work is that
they duplicate the parent element. (first encountered in e51d2d )
However, the duplication code has special array handling too. To avoid
issues with this cyclical coupling the previous solution was to
reimplement object duplication (with all sorts of pitfalls that has).
Now, I've tried to decouple it further by clearing all array psets prior
to any change, and readding the pset after everything has been
regenerated.
This can be improved upon but I don't feel confident until there is more
comprehensive test coverage for the duplicate operator.
This has a few advantages:
- The collection logic is centralised
- The collection logic is configurable based on the collection mode
- The name is not hardcoded
This reimplements @theoryshaw 's commit 9adbd4 but has a few upgrades:
- Considers all parent / child relationships, not just aggregates
- Puts business logic in core where it belongs and tool code in tool
- Uses existing utils where possible like get_decomposition
- Does not use name based collection checking which is fragile
- Reuses tool.Collector
- Makes container assignment handle the API's capability to do things
in bulk instead of one by one in a loop, so it's faster
- Tests
Operators removed (not called in any UI):
- SAIKEI_OT_create_alignment_polyline
- SAIKEI_OT_create_alignment_offset
- SAIKEI_OT_add_vertical_layout
- SAIKEI_OT_add_layout_segment
- SAIKEI_OT_layout_horizontal_by_pi
- SAIKEI_OT_layout_vertical_by_pi
- SAIKEI_OT_create_representation
- SAIKEI_OT_create_segment_representations
- SAIKEI_OT_update_fallback_position
- SAIKEI_OT_validate_segments
- SAIKEI_OT_refresh_alignment_data
Also fixed poll_ifc4x3() and replaced all tool.Alignment.get_ifc_file()
calls with tool.Ifc.get() after previous refactoring removed that method.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This is a cool idea, but users have all sorts of Blender collection
strategies and I think it's a good idea for Bonsai code to just touch
Bonsai's stuff and leave everything else. Separate functionality can be
built for non-Bonsai workflows and preferrably in a more discoverable
way than in individual Bonsai features.
When joining objects with mismatched representation types, deselect and
skip incompatible objects rather than canceling the entire operation.
This allows users to join compatible objects even when the selection
includes incompatible ones.
The previous fix by @theoryshaw had the correct conclusion but had some
issues:
1. Don't implement anything that fetches IFC data in the UI draw calls.
This historically has lead to race conditions and crashes. The data
classes are much preferred.
2. Rather than load the inapplicable attribute then selectively omit it
in the UI, it's preferred to just never load it in the first place.
Reuse tool.Geometry.duplicate_ifc_objects because that's where all
duplication logic should stay instead of half reimplementing it every
time which introduces subtle bugs.
I generally like to minimise knobs and dials, so this is now part of the
wizard but now defaults to a "NONE" unit. You also now have the option
to choose "NONE" for area / volume units which makes things consistent.
Enum names also match the IFC lowercase convention for conversion based
units. This also simplifies the core logic and treats all units the same
way instead of special cases for each unit type. (length is still
special and required in Bonsai as we are inherently graphical)
- is_skipped is calculated upstream instead of reimplementing logic
- the viewer now shows cardinality (not just in edit mode) to make it
clear what the requirements are
- prohibited specs are not allowed to have any requirements, so that
section is hidden
- failed prohibited specs now shows list of failed (applicable)
entities
The report_specification() method in the Json reporter class was raising
an UnboundLocalError when processing IDS specifications with certain
minOccurs/maxOccurs combinations that weren't explicitly handled.
Problem:
The cardinality variable was only assigned for three specific cases:
- minOccurs=1, maxOccurs="unbounded" → "required"
- minOccurs=0, maxOccurs="unbounded" → "optional"
- minOccurs=0, maxOccurs=0 → "prohibited"
However, the IDS schema allows other valid combinations such as:
- minOccurs=0, maxOccurs=1 (commonly used for optional specifications)
- minOccurs=1, maxOccurs=1 (exactly one occurrence required)
- Any other valid XML Schema cardinality values
When processing IDS files with these combinations, the cardinality
variable remained unassigned, causing an UnboundLocalError at line 382
when attempting to use it in ResultsSpecification().
Solution:
Added fallback logic to handle all valid IDS cardinality combinations:
- If minOccurs >= 1: cardinality = "required" (must occur at least once)
- Otherwise (minOccurs == 0): cardinality = "optional" (may occur)
This maintains semantic compatibility with the existing codebase, which
expects cardinality to be one of the semantic strings ("required",
"optional", "prohibited") rather than numeric ranges. This is critical
for:
- HTML template rendering (line 457: .capitalize())
- Conditional logic for skipped specs (line 454)
- UI rendering for prohibited specs (line 456)
Testing:
- Tested with IDS file containing minOccurs=0 without explicit maxOccurs
(defaults to 1 per XML Schema specification)
- Validation now completes successfully without UnboundLocalError
- HTML report generation works correctly with semantic cardinality labels
- Maintains backward compatibility with existing IDS files
Fixes: Validation failure when using valid IDS cardinality combinations
* black .
* add typing
* move function
* add get_class_relations
* add test for class relation
---------
Co-authored-by: Dion Moult <dionmoult@gmail.com>
Demo - https://files.catbox.moe/extsxf.mp4
But now it requires ifcsverchok graph to be nested inside a subgroup because we need a way to indicate which graph parameters are important, so user can modify them from Parametric Geometry tab without actually explicitly opening sverchok graph.
This reverts commit 96fe9b5398.
This commit contains a logical error where hide_viewport is checked in
the if statement but hide_set() is used as the effect. This causes a
regression in tests about project visibility. `hide_viewport` is the
correct hiding setting to use, because hide_set is more ephemeral and
affected by Shift-H, Alt-H etc.
Previously, when the add occurrence modal operator was executed, on
every modal loop (i.e. every mouse movement) it would fetch the mesh
geometry to be previewed, store the verts / edges / faces in mesh
collections, then the decorator would fetch that geometry, the clear the
collections, in a loop.
I've removed the Blender collections. Instead the same strategy is used
as in ItemDecorator i.e. the mesh and verts are fetched once during
decorator installation, then on each draw call only a single vertex loop
to multiply by the transformation matrix for snapping and mouse
position.
You can test with the LOD400 model in #7566. On my machine it would
cause lag on anything with >500 faces. Now it seems to work without lag
on a 26k polygon mesh.
Although it is a very cool trick, I feel this essentially duplicates the
UI in two spots, and is an outlier in UX. I'd prefer for all IFC data
and relationships to be in one location only (the panels). I think there
are a better unified solutions (e.g. favourite bookmark panels) for
quick access for things like this.
To be consistent with all other settings, I've moved the visibility
config UI from inline into the add-on settings. This restores the
previous tab layout and no longer needs the "settings" icons to be
there. This also removes the need for a "enable UI config" checkbox.
Most of the code previously had dedicated operators to toggle booleans.
This has been removed. This new approach also means helpers aren't
needed.
Previously calculation of visibility was done on every draw (3x 10tabs x
10 collection items). The data class is intended to calculate UI data
once only which is more efficient. This also removes all helper calls
from the UI.
When assigning a container to an aggregated element, automatically promote
the operation to the root aggregate and move all nested parts to the
container's collection in the Blender outliner.
Changes:
- AssignContainer now traverses the aggregate hierarchy to find the root
aggregate when a user selects any nested part
- All parts and sub-aggregates are moved to the container's collection in
the outliner while preserving IFC aggregate relationships
- Parts remain aggregated in IFC (not directly contained), only their
Blender collection membership changes
- RefreshLinkedAggregate now also moves all parts to the correct container
collection when restoring original data
This provides a more intuitive UX - users can select any part and the entire
assembly moves together, properly organized under the spatial container.
Fixes the previous behavior where:
- Aggregated elements were skipped with a warning
- Parts weren't organized under the container in the outliner
- Aggregate nesting was broken after container assignment
- Auto-include all parts when duplicating aggregates
- Preserve nested aggregate relationships during duplication
- Select all duplicated objects for immediate moving
- Add dual-rotation support for AXIS3 slabs (IFC angle + object rotation)
- Fix profile editing to display horizontal projection for tilted slabs
- Fix AXIS2 layer slicing to use local extrusion direction for walls
- Fix ChangeExtrusionDepth to refresh geometry after depth changes
- Remove rotation lock on slabs to allow free rotation
- Fix undefined variable bug in add_slab_representation.py
This commit introduces a new BBIM_MaterialLayer property set to persist custom
material layer offset settings in IFC files, along with significant UI improvements
for material editing.
Features Added:
- New BBIM_MaterialLayer pset with properties:
- UseCustomOffset (bool): Toggle for custom offset
- CustomOffset (float): Offset value in SI units
- CustomWallReference (str): Wall reference point (EXTERIOR/CENTER/INTERIOR)
- CustomSlabReference (str): Slab reference point (TOP/MIDDLE/BOTTOM)
Tool Updates (tool.py):
- Added save_custom_offset_to_pset(): Saves custom offset from UI props to pset
- Added load_custom_offset_from_pset(): Loads custom offset from pset to UI props
- Updated get_material_layer_custom_offset(): Reads from pset when props unavailable
Operator Updates (operator.py):
- EnableEditingAssignedMaterial: Loads custom offset from pset on edit start
- EditAssignedMaterial: Saves custom offset to pset on edit completion
- Fixed KeyError for CardinalPoint in material constituent sets
Data Layer (data.py):
- Added bbim_material_layer_pset() to ObjectMaterialData for caching pset data
- Improves performance by avoiding repeated IFC queries during UI drawing
UI Improvements (ui.py):
- Added custom offset display in both editable and read-only material UIs
- Added OffsetFromReferenceLine display in read-only UI
- Implemented dynamic headers based on material type (Layers/Profiles/Constituents)
- Improved visual hierarchy with consistent boxing and indentation
- Aligned editable and read-only UI layouts for consistency
- Fixed layer set boundary labels (Top/Bottom for slabs, Interior/Exterior for walls)
- Reorganized "Add Material" section into material layers box
Bug Fixes:
- Fixed format_distance() to correctly handle negative imperial values
(e.g., -0.5' now displays as "-0' - 6"" instead of "0' - -6"")
This allows users to set custom material layer offsets that persist in the IFC
file and remain available across sessions, with clear visual feedback in both
editing and viewing modes.
When appending a wall type and a slab type in turn, if their material
layer sets have the same name then the slab type would have a wall
construction. Now the material sets are compared before reusing an
existing material set.
Fixes typo introduced in b4740b6 where `element in MATERIAL_SETS` should
have been `element.is_a() in MATERIAL_SETS`. This resulted in
deduplication of layersets, but not of the layers themselves.
Enable multiple CSS files in stylesheet_path using comma separation.
Files are loaded in order with natural CSS cascading behavior.
Example: "base.css, overrides.css"
"In some versions of Python, instances of classes may have an
__annotations__ attribute. However, this is not supported functionality.
If you need the annotations of an instance, you can use type() to access
its class" https://docs.python.org/3/howto/annotations.html
When editing representation items for elements with IfcMaterialLayerSetUsage
(LAYER2/LAYER3), the depth attribute is now hidden from the UI as it should
not be modified at the item level for these parametric elements.
The check is performed by accessing the parent element through the
representation_obj property in geometry props and checking its material
usage type.
The copy_node_graph() method checks for a "window" key in the context
override, but get_shader_editor_context() wasn't providing it. This
caused the function to return early, preventing external styles from
loading.
Updated get_shader_editor_context() to include the window when the
screen differs from the current context, following the same pattern
used in get_viewport_context().
Changed tolerance from 1e-5 to 1e-4 to account for floating-point errors
introduced by matrix transformations. Fixes section annotations being
incorrectly excluded from drawings when they should be visible.
Moves IfcSpace hiding from collection assignment to after scene
addition, allowing hide_set() to work properly once objects are
in the view layer. Fixes RuntimeError during IFC import.
When duplicating elements, skip assign_body_styles if the element's
constituent materials already have style representations. This prevents
creating redundant IfcStyledItem entities on the geometry when styles
should be inherited from the material definition.
To speed up the process. Previously we were using `/m` which build different targets in parallel, but not .cpp files. So rocksdb could have been building for 30 mins because of this.
sync_references was syncing spatial element placements FROM Blender TO IFC,
corrupting their correct positions. Spatial elements (storeys, spaces,
buildings) often have Blender objects at Z=0 for modeling convenience, but
their IFC placements store absolute positions.
Solution: Skip syncing placements for IfcSpatialElement and IfcGrid types,
as their IFC placement is the source of truth.
Fixes storey elevation corruption during SECTION_LEVEL annotation generation.
- Removed interface.rst (was added in this PR)
- Removed all PNG documentation images that were added in this PR:
- bonsai_customization1.png
- bonsai_customization2.png
- bonsai_interface.png
- bonsai_interface2.png
- bookmarks.png
- gear_tabs.png
These files were not present in the upstream v0.8.0 branch and have been removed to restore the original documentation state.
Force UI redraw to flush pending Blender operations, as Blender was crashing during bulk printing. Not 100% this fixed everything, as there are a few crashes that spring up still, but is better.
Force viewport to use scene camera before rendering underlay. When
print_all was used, viewport remained locked to local camera from
previous drawing, causing OpenGL render to capture wrong view.
Enable importing multiple .svg reference files at once using
standard Blender multi-select (Shift/Ctrl+click).
Includes backward compatibility and test coverage.
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.
Added more gizmos for multi-panel windows and for the door transom.
Support negative dimension values (lining offset for door and window)
Fix railing, stair, and roof being regenerated during UI panel draw instead of on property change
Various code quality changes and DRY improvements
Major changes:
- Relocate gizmo infrastructure from bim/gizmo.py to module/drawing/gizmos.py
- Replace arrow-based property gizmos with dimension line gizmos throughout
- Add view-dependent positioning: gizmos automatically reposition based on camera angle to stay visible and avoid overlapping geometry
- Add special icons for stair to deal with edge cases
Code quality:
- Add DRY helper methods for gizmo positioning across stair, door, and window
- Remove redundant visibility logic
- Extract integer input handling to dedicated module for reuse
- Add comprehensive documentation for gizmo architecture
Add dialog to DuplicateType operator allowing users to set name, description,
and optionally assign the active object to the duplicated type. Add duplicate
button to product UI in type panel with auto-assignment enabled by default.
- You can now input values using the keyboard once you clicked on a gizmo
- Gizmos now support click > move > click in addition to drag and drop (yay carpal tunnel !)
- Optimize snapping performance
- Consolidate door and window type in model.py
- You can now cycle through window and door types using the cycle gizmo
- Pen, validate, cancel, lock, and cycle gizmos are now billboards and follow view direction
- Draggable gizmos are now billboard 2D arrows instead of 3D cones
Implement visual parameter manipulation in 3D viewport with reusable
gizmo components. Features snap-to-mesh (Ctrl), precision mode (Shift),
and real-time feedback. Includes cone gizmos for dimensions,
arc gizmo for swing direction, and icon gizmos for edit controls. Gizmos are globally are individually toggleable in the addon preferences.
May be expanded upon later to add gizmos to other paramaterized elements (window, railing, roof, stairs, array, ...).
Unfortunately gizmos can't tap into the builtin snap system or bonsai snap system which is built for modal operators so gizmos implement yet another custom rather naive snap system.
When importing spatial decomposition, get_storey_elevation() returns values in the project's unit system (e.g., feet), but format_distance() expects values in meters. Added unit_scale conversion (elevation * unit_scale) before formatting to ensure elevations display correctly regardless of project units.
Level annotations were incorrectly converting feet to meters and back,
causing values like 3ft to display as 9'10". Updated format_value() to
accept in_unit_length parameter and pass it to format_distance(), since
Blender's Z coordinates are already in project units.
- Add in_unit_length parameter to BaseDecorator.format_value()
- Set in_unit_length=True in PlanLevelDecorator.draw_labels()
- Set in_unit_length=True in SectionLevelDecorator.draw_labels()
Replaced duplicate format_distance in unit.py with comprehensive MeasureIt-ARCH version from helper.py. Fixed critical floating point errors by changing inPerFoot from 11.9999 to 12.0, eliminating spurious fractions on whole numbers. Implemented auto-precision detection to simplify fractions (64/256 → 1/4) while preserving precise values like 3/256 when needed. Enhanced imperial parsing to support flexible input formats (optional symbols, mixed fractions, leading decimals) and fixed suppress_zero_inches functionality in dimension annotations.
Files saved in previous Blender versions stored position checksums as float64
(24 bytes for location, 72 bytes for rotation), but the current code only
expected float32 format. This caused a shape mismatch error (6,) vs (3,) when
comparing checksums during save operations.
Added byte size detection in is_moved() methods to automatically handle both
old float64 and new float32 checksum formats, ensuring backward compatibility
with files from previous versions.
Fixes ValueError when saving IFC files loaded from older Blender versions.
Unexpected consequence of #7365 - previously `OPENCASCADE_LIBRARIES` supposed to be a list of libraries filepaths, but after refactor `OpenCASCADE_LIBRARIES` end up being both list of library names and list of filepaths.
And since link directories for the names was not provided, it failed to find the libraries to link.
If using OpenCASCADEconfig it wouldn't fail as cmake would evaluate library names as imported target names.
To fix it and simplify it even further, in case of manually provided paths, it now creates `OpenCASCADE_INTERFACE` target and stores it in `OpenCASCADE_LIBRARIES` - target will automatically have all the includes, library names and library dirs.
Since this if-block happens in parentheses, then all `%VAR%` are expanded at the start and then `%ERRORLEVEL%` and `%RET%` values have no relation to the actual result of `git clone`. So if user were (un)lucky, then `%RET%` would be not yet set to "0" (by the previous function call), then it would skip checking out revision.
`GitCloneOrPullRepository` had similar issue, but it was never used anywhere.
`EIgen::Eigen` target used in #7340 for linking exists only if there's a cmake config, while there are cases when there's just include directory. Added same name interface target to mimic config.
It doesn't really matter in this case - it matters really only for FindLibXml2, because `PROJECT_SOURCE_DIR` point to svgfill, when it's config executed, instead of `CMAKE_SOURCE_DIR` - so `CMAKE_SOURCE_DIR ` is never removed from the path, leading to cyclic calls. Adding for HDF5 too just for consistency.
Not sure when it was introduced, but it was still using previous dependency path and leading to confusing error:
```
Boost 1.86.0 already downloaded. Skipping.
Extracting Boost 1.86.0 into L:\Projects\Github\IfcOpenShell\_deps from boost-1.86.0-b2-nodocs.7z
An error occurred
```
Note: rv3d.perspective_matrix should be avoided in these context. Use a numpy version of np.array(window_matrix) @ np.array(view_matrix). See https://github.com/IfcOpenShell/IfcOpenShell/issues/7046
Note2: region_2d_to_origin_3d_np clamp should be used for the snap to work in orthographic view.
Add target view detection to generate_section_reference_points() to create
appropriate geometry for each view type. Plan views get horizontal lines
(clip_segment), section/elevation views get vertical lines (elevate_segment).
Added steps to set 'is_visible' to TRUE and 'is_locked' to FALSE in the 'Spatial Decomposition' panel for the 'Override paste buffer - with active IFC data' scenario. This ensures the test covers visibility and lock state handling for IFC objects.
Example error:
```
TypeError: bpy_struct: item.attr = val: enum "Default" not found in ('city.exr', 'courtyard.exr', 'forest.exr', 'interior.exr', 'night.exr', 'studio.exr', 'sunrise.exr', 'sunset.exr')
```
There was a bug in shading_styles.json where we were using `Default` for `studio_light`, which is only available for `SOLID` shading type. Changed value to `forest.exr`, added code to handle old bugged value since users already copied and possibly modified shading_styles.json
The edit_assigned_product() function now removes the annotation from all
existing IfcRelAssignsToProduct relationships instead of just one, preventing
annotations from being incorrectly assigned to multiple products simultaneously.
This fixes issues where leader lines wouldn't update correctly because the
annotation retained old product assignments.
Previously, when multiple CSS selectors were comma-separated in a rule,
only the last selector received the unique drawing ID suffix. This caused
style conflicts when multiple drawings were placed on the same sheet.
Now all selectors in comma-separated lists receive the unique suffix,
ensuring proper style isolation between drawings.
Example:
Before: .cut.Status-DEMOLISH, .cut.Option-D.d2156 { ... }
After: .cut.Status-DEMOLISH.d2156, .cut.Option-D.d2156 { ... }
In Blender 5.0 some theme attributes were removed and moved to other places and code broke.
In the commit I've also switched the code to less dynamic approach, so static type checker could caught potential errors.
`get_active_stair_calculated_params` wasn't updated when zero tread runs support was added and it's still was relying on a convention that `0.0` custom tread run = no custom tread run.
`s0 = V_(custom_first_last_tread_run[0] or tread_run, tread_rise) + td_vector` was assuming `0` means "no custom first tread run", not it means actually "zero first tread run", so could should be adjusted
before - https://files.catbox.moe/5ovqnh.mp4
after - https://files.catbox.moe/srorl2.mp4
Before f52aafd default `(0,0)` meant that there are no custom tread runs, but after the change it actually means that runs for first and last treads are 0. So default value should be adjusted.
Added explicit `None` value for clarity.
Since it's not illegal in Blender and can be helpful - e.g. when you were assigning class to ifc space, you would lose selection completely and would need to reselect object to continue working on it.
After 3aa5ddb07 `select_products` adds to current selection, instead of resetting it.
So couple tests started to fail since they create new objects and `view_layer` is not yet updated (previously `bpy.ops.object.select_all(action="DESELECT")` was also implicitly updating view layer).
```
FAILED test/tool/test_spatial.py::TestSelectProducts::test_select_products - AssertionError: assert bpy.data.objects['Object'] in []
FAILED test/tool/test_system.py::TestSelectSystemProducts::test_run - AssertionError: assert bpy.data.objects['Object'] in []
```
Long story short - since 5.0 `np.array(Vector())` is now producing `np.float32` instead of `np.float64`. So we have to provide `dtype` explicitly to support both <5.0 and >= 5.0.
See https://projects.blender.org/blender/blender/issues/149283
It's probably will be very popular bug, when someone would try to migrate to Blender 5.0 using old Bonsai.
Example traceback:
```
File "Blender\5.0\extensions\.local\lib\python3.11\site-packages\bonsai\bim\prop.py", line 34, in <module>
import bonsai.bim.schema
File "Blender\5.0\extensions\.local\lib\python3.11\site-packages\bonsai\bim\schema.py", line 23, in <module>
import bpy_restrict_state
ModuleNotFoundError: No module named 'bpy_restrict_state'
```
Noticed testing command from #7277. I believe Serializers it not actually needed to build IfcGeomServer.
Example error:
```
/usr/bin/ld: cannot find -lSerializers: No such file or directory
```
`-Wl,-undefined,suppress` have no meaning in emscripten and they are just ignored. And emscripten is ignoring undefined symbols by default without any flags.
Next to `EXPORTED_FUNCTIONS` and avoid possible issues with pyodide in the future (pyodide is using `SIDE_MODULE=1`, but currently not providing it libraries of type `MODULE`, only for shared ones).
Adding extension, but avoiding adding sources as we're not actually building.
As a result build will have a correct suffix `pyodide_2025_0_wasm32.whl` that's pyodide is currently using and this is pyodide is plan to use when it's going to release to PyPI, see https://peps.python.org/pep-0783/
Documentation claimed that ODS/XLSX output was to a given filename, but
the code actually writes files with generated names in a folder.
(also create the folder if necessary for convenience)
This also deprecates one more instance of the add_representation API
call. As this is slowly cleaned up we can move towards "regenerating"
annotations, not just generating from scratch.
* first attempt to do annotation rotation when dong bulk tags
* Enhance annotation rotation handling based on camera orientation
* handle properly S_G
* black .
* reorder property in UI
* cleanup imports
Remove validate_sheet_files() check in regenerate_sheet() that prevented
regenerating missing LAYOUT files. The validation was blocking the
exact scenario that regeneration was designed to handle.
This allows users to open a bare IFC file with configured drawings and
sheets and have the SVG layouts automatically recreated with drawings
placed at default positions.
Note: now we are using hide_viewport not hide_set anymore. There might
be pending issues from this. Also, now because we use foreach_set, we
can quickly also apply hide_render, which saves time during underlay
generation.
The issue was with ci.yml - it was using system boost shared libraries (since Boost_USE_STATIC_LIBS by default is OFF), but CGALConfig was overriding leading to compilation issues.
Text annotations can now reverse the order of list values (e.g., materials) and use custom separators instead of the default comma-space. Two new properties have been added to EPset_Annotation: Reverse_List (boolean) and List_Separator (string). When text literals contain IFC list/tuple values, they are now detected during variable replacement and can be reversed and joined with the specified separator. The list separator supports escape sequences like \n for newlines, enabling multi-line list displays. These properties are exposed in the text editing UI and properly persist to the IFC file through the new update_text_annotation_properties method, which consolidates all annotation property updates into a single efficient operation.
Not sure why it occurred only now, maybe after recent changes with shared libraries this problem just got more apparent and previously it was masked by some other library after mappings also using IfcGeom.
But the issue was that mapping were not linked against IfcGeom, though it's symbols were used in them. E.g. IfcSectionedSurface.cpp was using `has_loft` from `infra_sweep_helper`.
Issue occurred only on Linux - msvc and clang are less strict and probably do multiple passes to fetch symbols.
Example errors:
```
libgeometry_mapping_ifc4x3.a(IfcSectionedSolidHorizontal.cpp.o): In function `ifcopenshell::geometry::mapping_Ifc4x3::map_impl(Ifc4x3::IfcSectionedSolidHorizontal const*)':
IfcSectionedSolidHorizontal.cpp:(.text._ZN12ifcopenshell8geometry14mapping_Ifc4x38map_implEPKN6Ifc4x327IfcSectionedSolidHorizontalE+0x19a4): undefined reference to `ifcopenshell::geometry::make_loft(ifcopenshell::geometry::Settings const&, IfcUtil::IfcBaseClass const*, std::shared_ptr<ifcopenshell::geometry::taxonomy::function_item> const&, std::vector<ifcopenshell::geometry::cross_section, std::allocator<ifcopenshell::geometry::cross_section> >&)'
libgeometry_mapping_ifc4x3.a(IfcSectionedSurface.cpp.o): In function `ifcopenshell::geometry::mapping_Ifc4x3::map_impl(Ifc4x3::IfcSectionedSurface const*)':
IfcSectionedSurface.cpp:(.text._ZN12ifcopenshell8geometry14mapping_Ifc4x38map_implEPKN6Ifc4x319IfcSectionedSurfaceE+0x19c4): undefined reference to `ifcopenshell::geometry::make_loft(ifcopenshell::geometry::Settings const&, IfcUtil::IfcBaseClass const*, std::shared_ptr<ifcopenshell::geometry::taxonomy::function_item> const&, std::vector<ifcopenshell::geometry::cross_section, std::allocator<ifcopenshell::geometry::cross_section> >&)'
libgeometry_mapping_ifc4x3_tc1.a(IfcSectionedSolidHorizontal.cpp.o): In function `ifcopenshell::geometry::mapping_Ifc4x3_tc1::map_impl(Ifc4x3_tc1::IfcSectionedSolidHorizontal const*)':
IfcSectionedSolidHorizontal.cpp:(.text._ZN12ifcopenshell8geometry18mapping_Ifc4x3_tc18map_implEPKN10Ifc4x3_tc127IfcSectionedSolidHorizontalE+0x19a4): undefined reference to `ifcopenshell::geometry::make_loft(ifcopenshell::geometry::Settings const&, IfcUtil::IfcBaseClass const*, std::shared_ptr<ifcopenshell::geometry::taxonomy::function_item> const&, std::vector<ifcopenshell::geometry::cross_section, std::allocator<ifcopenshell::geometry::cross_section> >&)'
libgeometry_mapping_ifc4x3_tc1.a(IfcSectionedSurface.cpp.o): In function `ifcopenshell::geometry::mapping_Ifc4x3_tc1::map_impl(Ifc4x3_tc1::IfcSectionedSurface const*)':
IfcSectionedSurface.cpp:(.text._ZN12ifcopenshell8geometry18mapping_Ifc4x3_tc18map_implEPKN10Ifc4x3_tc119IfcSectionedSurfaceE+0x19c4): undefined reference to `ifcopenshell::geometry::make_loft(ifcopenshell::geometry::Settings const&, IfcUtil::IfcBaseClass const*, std::shared_ptr<ifcopenshell::geometry::taxonomy::function_item> const&, std::vector<ifcopenshell::geometry::cross_section, std::allocator<ifcopenshell::geometry::cross_section> >&)'
libgeometry_mapping_ifc4x3_add1.a(IfcSectionedSolidHorizontal.cpp.o): In function `ifcopenshell::geometry::mapping_Ifc4x3_add1::map_impl(Ifc4x3_add1::IfcSectionedSolidHorizontal const*)':
IfcSectionedSolidHorizontal.cpp:(.text._ZN12ifcopenshell8geometry19mapping_Ifc4x3_add18map_implEPKN11Ifc4x3_add127IfcSectionedSolidHorizontalE+0x19a4): undefined reference to `ifcopenshell::geometry::make_loft(ifcopenshell::geometry::Settings const&, IfcUtil::IfcBaseClass const*, std::shared_ptr<ifcopenshell::geometry::taxonomy::function_item> const&, std::vector<ifcopenshell::geometry::cross_section, std::allocator<ifcopenshell::geometry::cross_section> >&)'
libgeometry_mapping_ifc4x3_add1.a(IfcSectionedSurface.cpp.o):IfcSectionedSurface.cpp:(.text._ZN12ifcopenshell8geometry19mapping_Ifc4x3_add18map_implEPKN11Ifc4x3_add119IfcSectionedSurfaceE+0x19c4): more undefined references to `ifcopenshell::geometry::make_loft(ifcopenshell::geometry::Settings const&, IfcUtil::IfcBaseClass const*, std::shared_ptr<ifcopenshell::geometry::taxonomy::function_item> const&, std::vector<ifcopenshell::geometry::cross_section, std::allocator<ifcopenshell::geometry::cross_section> >&)' follow
collect2: error: ld returned 1 exit status
```
Otherwise it can be overlooked if you build IfcOpenShell-Python or any other target specifically, it was only built if you run build-all.py without any target arguments.
This will give you additional styling granularity when working with material layers, allowing you to target elements not just by their material name but also by their material category in your SVG styling.
cmake configs seems to be provided with the default installation, so `find_package` should do the job that `find_path` and `find_library` did previously.
Previously we were setting `HDF5_LIBRARIES` explicitly, but I'm not sure if really worked, since we never set `HDF5_INCLUDE_DIR` and user would have to provide it manually either way.
`find_package(HDF5)` will search for default paths and will set `HDF5_LIBRARIES` and `HDF5_INCLUDE_DIR` automatically.
As it was needed for more recent swig version. Bumped swig in ae771cb without knowing this, so essentially building swig with build-all.py was broken for a bit.
After version 1.13.0 it seems hdf5 stopped publishing builds on the ftp and they also stopped publishing `CMake-hdf5-*` builds.
Switching to cmake, it will also make it consistent with build-deps.cmd
Browser was stuck trying to initialize the environment, because of the error below:
[worker] Received message: {type: 'init', payload: {…}, id: '9dLz/0eBQDa0xTQORXfdUg/0'} :5000/pyodide/pyodide.mjs:1 Failed to load module script: Expected a JavaScript-or-Wasm module script but the server responded with a MIME type of "text/plain". Strict MIME type checking is enforced for module scripts per HTML spec.
They were hardcoded and `:download:` wasn't used, so sphinx never copied the scripts to be available to download.
They also were always pointing to main docs website, so they would show up on unstable documentation too, and I guess because they were hardcoded, there were no warnings when those files were recently moved.
Noticed compiling against libxml2 2.13.8 package that it's trying to use dynamic libs by default if `LIBXML_STATIC` is not defined. Adding `LIBXML2_DEFINITIONS` helps.
Actually after 2ec2ac0 there's no need for json serialization as we support working with string subitems natively.
For some reason previously it was using get_georeference_props to add attributes, I guess it a typo that happened awhile ago.
To avoid crashes and either way we stored the same data twice - once bim text props (and they were updated on each data refresh) and then again in data.py. Now it's going to use just data.py
Resolves issue #6099: PythonOCC >=7.8.0 changed WriteToString() method
signature, causing TypeError in ifcopenshell.geom.serialise().
This fix uses signature inspection to detect the method signature:
- For PythonOCC < 7.8.0: Use WriteToString() (no parameters)
- For PythonOCC >= 7.8.0: Fall back to Write() method
- Graceful handling of signature inspection failures
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Simplify path definitions in `test_write.py` and `test_open.py`. Replace `typing` with `collections.abc` for `Sequence` usage in `test_sweeps.py` and adjust type hints to use Python's generic collections. Add optional verbosity to `load_ifc_occ_shape` and refine geometry test assertions. Update `simple_sweep_2.ifc` fixtures for consistency with test expectations.
Introduces a new `lint` feature with `ruff` and `black` dependencies. Extends test setup with additional dependencies and tasks for building, testing, and installation. Cleans up `.gitignore` and removes unused entries from `pixi.lock`.
Improves formatting and readability in `test_sweeps.py`, adjusts imports and comments, and adds assertions for geometry validation in `simple_sweep_1.ifc` and `simple_sweep_2.ifc` tests. Updates code comments in `IfcFixedReferenceSweptAreaSolid`.
Includes `simple_sweep_2.ifc` test input, geometry validations in `test_sweeps.py`, and a new constructor for `sweep_along_curve` to handle directrix-based sweeps.
Enhances geometry processing for swept area solids by integrating matrix transformations. Includes new utility `_is_swept_shape` for topology analysis, extends testing with `load_ifc_occ_shape`, and updates schema versions in CMake presets.
Currently it's in Bonsai docs as this next to previously added formatter docs. Probably, should move it to ifcopenshell docs later.
C++ code style part about braces seems to be outdated.
Our minimum cmake version is 3.21, so policies below use `NEW` by default already:
- CMP0048 - VERSION option support for `project`, introduced in 3.0
- CMP0074 - find_package() uses <PackageName>_ROOT variables, introduced in 3.12
- CMP0078 - UseSWIG generates standard target names, introduced in 3.13
- CMP0086 - UseSWIG honors SWIG_MODULE_NAME via -module flag, introduced in 3.14
If you try to append an asset which has a relationship such as
IfcRelAssociatesClassification to a whole bunch of opening elements, all
those openings will also (incorrectly) be pulled in.
To fix 1k warnings running in cmake 3.30+
The new policy is that cmake is normalizing destination paths in `install` and warns if normalization has any effect on them. In our case paths after normalization should be fine, so switching it to `NEW`.
* Enhance AddElement operator to orient the extruded ifcopening along the AXIS orientation
* Adjust extrusion direction based on DirectionSense for AXIS2 & AXIS3 IfcOpeningElement
* Remove transform orientation settings in BoundingBoxDecorator
* Remove unused transform orientation handling in update_measure_xyz and black formatting
It adds this option to open directory in system browser using alt+click and we don't need anymore a special operator `bim.select_dir` to fill these props.
Example - https://files.catbox.moe/t8ejrs.png
The problem was with the line `read_surface_style(surface_styles[i], surface_style_ptrs[i]);` since `surface_style_ptrs[i]` was a nullptr.
Changed the signature to `style&` to make it more clear that it's expecting already created style struct.
It wasn't written, but HdfSerializer was still requiring and throwing errors if it couldn' t find it, so cache wasn't working...
Honestly, me when I realized what happened - 🫣🫣
Though there are no particular known issues with https://github.com/ortega2247/pypi-upload-action we've used before, but it wasn't updating for 5 years now, so it's better to switch to official PyPA workflow.
Args for workflow seem to be exactly the same, so it's just a drop-in replacement.
E.g. when user opened .blend file and ifc filepath was pointing to a wrong destination, after they changed the filepath pset templates wouldn't reload.
There was a contradiction - modal operators were treated as special and were always creating new transactions (`is_top_level_operator = method == "MODAL"`), but those transactions were only closed if modal returned `FINISHED` or `CANCELLED`, so in most case previous transactions were just overridden, so IFC undo data was lost.
Ping @Moult just in case
verbose logging was setting log level only for `logger`, never setting for the `logger`'s handler, so `-v` option, it seems, never worked. For simplicity now handler will just inherit log level from the `logger`.
See #6831. Warning: this is a breaking change.
Linked Aggregate names are numbered to keep them consistent when reopening
the file instead of relying on Blender auto numbering objects with the
same name. However, for annotation purposes, the numbering is irrelevant,
and the user might want to use just the aggregate name. So it was added a
"Name" to the "BBIM_Linked_Aggregate" Pset that can be used in these cases.
When refreshing the linked aggregates, their names will be updated to reflect
the "Name" value and the product assignment will be preserved to keep
annotations in sync.
In order to adapt existing files to this new Pset structure see this gist:
https://gist.github.com/brunoperdigao/37a5b838935b15a91aa70b2bc4de325f
E.g. IfcWorkSchedule.Duration wasn't recognized as such or IfcWorkSchedule.TotalFloat wasn't imported properly and was uneditable.
Now we also check IfcDuration type on attributes explicitly, instead of relying on hardcoded list
Location - https://files.catbox.moe/sxfxv4.png
To mirror the similar UI we have in 4d, this way feature will much easier to find than checking File->Import
A feature allowing duplicate existing IfcCostSchedule - either from API or from Bonsai UI
Button location in Bonsai - https://files.catbox.moe/ct058q.png
After ba2456a if main IFC project had an offset and linked object had more than 333 faces, then it was imported at it's original location instead, missing an offset.
Also processing occurrences should be now more optimized, since we reuse the same verts buffer instead of recreating it 10 times.
* Export_IfcCostSchedule_to_PDF_improvements
* IfcCostSchedule CSV export: Added ItemIsASum column
New column in the ifc export that tracks if IfcCostItem is a sum, also added a new static method to the IfcDataGetter class.
* IfcCostSchedule CSV export: Added cost quantities column
Cost quantities are a serialsed list containing the name of the quantity and the quantity value.
* IfcCostScheduel PDF export: add options to fine tune export
New options include nested_structure_depth, should_print_cover, should_print_description, should_print_rates, should_print_summary, should_print_cost_ids.
Also pass project currency to typst (still not used).
Added footer with "proudly created with IfcOpenShell".
Updated Cover with formatting and IfcCostSchedule Description
Now we store in the IFC file list of active bsdd user was using in Bonsai.
So when you open IFC file it automatically marks dictionaries as active and there's no need to go to bsdd UI, unless there are new dictionaries to add/remove.
Example - https://files.catbox.moe/kzwa2n.mp4
Since we added lookup for multiple paths, it was failing with confusing "Could not find Bonsai path in expected locations" even if the problem was with BLENDER_PATH.
Previously some of them were saved only for current .blend file, moving them to preferences will make it more consistent with usual Blender UX. If needed we'll be able to add some way to fine-grain them later.
There's also a temporary patch that's going to migrate old .blend-props to new preferences-props to make process less disruptive.
`settings` wasn't defined, so `create_shape` was always resulting in an error. Also, `create_shape` is returning `Triangulation` in this kind of cases, so there's no `.geometry`.
Also removed IfcCircleProfileDef if-check since it's covered by IfcParameterizedProfileDef.
It's probably was just an overlook in the original commit - it was referring to `settings["organisation"]`, though `"organization"` was never passed, leading to an error. And probably by accident remove_role tests were using the same tests as remove_organization, so the issue went under the radar for a while.
| bsdd | Library to query the bSDD API | LGPL-3.0-or-later | [](https://pypi.org/project/bsdd/) |
| ifc2ca | Utility to convert IFC structural analysis models to Code_Aster | LGPL-3.0-or-later |
| ifc4d | Convert to and from IFC and project management software | LGPL-3.0-or-later | [](https://pypi.org/project/ifc4d/) |
| ifc5d | Report and optimise cost information from IFC | LGPL-3.0-or-later | [](https://pypi.org/project/ifc5d/) |
| ifcbimtester | Wrapper for Gherkin based unit testing for IFC models | LGPL-3.0-or-later |
| [bcf](https://docs.ifcopenshell.org/bcf.html) | Library to read and write BCF-XML and query OpenCDE BCF-API modules | LGPL-3.0-or-later | [](https://pypi.org/project/bcf-client/) [](https://anaconda.org/conda-forge/bcf-client) |
| [bsdd](https://docs.ifcopenshell.org/bsdd.html) | Library to query the bSDD API | LGPL-3.0-or-later | [](https://pypi.org/project/bsdd/) |
| [ifc2ca](https://docs.ifcopenshell.org/ifc2ca.html) | Utility to convert IFC structural analysis models to Code_Aster | LGPL-3.0-or-later |
| [ifc4d](https://docs.ifcopenshell.org/ifc4d.html) | Convert to and from IFC and project management software | LGPL-3.0-or-later | [](https://pypi.org/project/ifc4d/) |
| [ifc5d](https://docs.ifcopenshell.org/ifc5d.html) | Report and optimise cost information from IFC | LGPL-3.0-or-later | [](https://pypi.org/project/ifc5d/) |
| [ifcbimtester](https://docs.ifcopenshell.org/bimtester.html) | Wrapper for Gherkin based unit testing for IFC models | LGPL-3.0-or-later |
| ifcconvert | CLI app to convert IFC to many other formats | LGPL-3.0-or-later\* | [](https://docs.ifcopenshell.org/ifcconvert/installation.html) [](https://github.com/IfcOpenShell/IfcOpenShell/releases?q=ifcconvert&expanded=true)
| ifccsv | Library and CLI app to export and import schedules from IFC | LGPL-3.0-or-later | [](https://pypi.org/project/ifccsv/) |
| ifcfm | Extract IFC data for FM handover requirements | LGPL-3.0-or-later | [](https://pypi.org/project/ifcfm/) |
| ifcmax | Historic extension for IFC support in 3DS Max | LGPL-3.0-or-later\* | [](https://docs.ifcopenshell.org/ifcmax.html)
| [ifcconvert](https://docs.ifcopenshell.org/ifcconvert.html) | CLI app to convert IFC to many other formats | LGPL-3.0-or-later\* | [](https://docs.ifcopenshell.org/ifcconvert/installation.html) [](https://github.com/IfcOpenShell/IfcOpenShell/releases?q=ifcconvert&expanded=true)
| [ifccsv](https://docs.ifcopenshell.org/ifccsv.html) | Library and CLI app to export and import schedules from IFC | LGPL-3.0-or-later | [](https://pypi.org/project/ifccsv/) |
| [ifcfm](https://docs.ifcopenshell.org/ifcfm.html) | Extract IFC data for FM handover requirements | LGPL-3.0-or-later | [](https://pypi.org/project/ifcfm/) |
| [ifcmax](https://docs.ifcopenshell.org/ifcmax.html) | Historic extension for IFC support in 3DS Max | LGPL-3.0-or-later\* | [](https://docs.ifcopenshell.org/ifcmax.html)
configure-release={cmd=["cmake","--preset","win-release","-B","build/win-release","cmake"],description="Configure the project",depends-on=[{task="init-submodules",environment="common"}]}
build-release={cmd=["cmake","--build","build/win-release","--config","Release"],description="Build the project"}
[feature.dev.target.win-64.tasks]
configure-debug={cmd=["cmake","--preset","win-debug","-B","build/win-debug","cmake"],description="Configure the project",depends-on=[{task="init-submodules",environment="common"}],outputs=["build/win-debug/CMakeCache.txt"]}
build-debug={cmd=["cmake","--build","build/win-debug","--config","Debug"],description="Build the project",depends-on=["configure-debug"]}
install-debug={cmd=["cmake","--install","build/win-debug","--config","Debug"],description="Install the project"}# Optionally Install files to your desired env using --prefix
vsdebug={cmd=["python"],description="Run a python script with vs debugger attached"}
[feature.tests.tasks]
configure-test={cmd=["cmake","--preset","win-test","-B","build/win-test","cmake"],description="Configure the project",depends-on=[{task="init-submodules",environment="common"}],outputs=["build/win-test/CMakeCache.txt"]}
build-test={cmd=["cmake","--build","build/win-test","--config","Release"],description="Build the project",depends-on=["configure-test"],outputs=["build/win-test/IfcGeom.lib"]}
install-test={cmd=["cmake","--install","build/win-test","--config","Release"],description="Install the project",depends-on=["build-test"]}# Optionally Install files to your desired env using --prefix
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
DesertSpringsCivil
