The hand-written MJCF[] table in xml_native_reader.cc is replaced by
mjcf_table.inc, emitted from mjcf.schema by generate_mjcf_table.py and
checked for freshness by doc_test. nMJCF is now self-sizing. The two
tables are identical as trees of (tag, cardinality, attribute-set);
within-row attribute order changes where the schema factors shared
groups and projects default-context rows, and top-level rows follow
the schema's dependency order -- neither affects validation, which is
set-based, nor XMLschema.rst, whose generator orders sections itself
(regenerated here, reading the .inc instead of the reader source).
The schema's constraint declarations become enforcement: the emitter
writes a companion MJCF_constraints[] array (row-indexed into MJCF[]),
and mjXSchema::Check evaluates each element's constraints after its
attribute check, with uniform messages derived from the declaration:
"at most one of 'fovy', 'sensorsize' can be specified", "attributes
'reftype', 'refname' must be specified together", and so on.
Multi-attribute bundles render as ('site1', 'site2').
Fifteen hand-written co-occurrence checks across fourteen elements are
deleted -- connect/weld semantics mixing and completeness, the actuator
transmission mutex, camera fovy/sensorsize, light directional/type,
inertial fullinertia-versus-orientation, rangefinder and the distance
family, contact's matching criteria, user-sensor pairing, the frame
family's reftype/refname, size memory exclusivities, mesh builtin
exclusions, and attach body/frame (newly declared). Tests assert the
uniform messages.
Two findings along the way: sensorsize-requires-resolution is a
value-level compiler rule (positive resolution), not a presence rule --
a presence constraint would be wrong and is not declared; and Size()'s
nstack/njmax range checks tested the spec value before assignment, so
they never validated the parsed value -- now they do.
Verified by compiling all 81 models in the model/ corpus.
PiperOrigin-RevId: 958064622
Change-Id: I802cf5c0aee08a62926e36a281320ff9e34c0668
Moving forward, the compiler objects should have encapsulated data, therefore the old ReadAttr* methods are being deprecated as they require raw pointers to read into.
No external changes other than minor improvements to ReadAttr and ReadVector.
PiperOrigin-RevId: 550880401
Change-Id: I632f40322ccd935f9af9b69e66b056c5cda6face
This lets the compiler catch the mistake of creating an exception without throwing it.
PiperOrigin-RevId: 474536776
Change-Id: I5bd18258ce6ddd266adb1c852b1ba9e3063cd0b2
- Mesh inertias can now be computed exactly for well-formed (no holes) non-convex meshes.
- To activate this feature, set `<compiler exactmeshinertia="true">` (defaults to `false`). This default may change in the future.
- Added `<geom shellinertia="true/false">` (defaults to `false`). When true, geom inertia is computed assuming all the mass is concentrated on the surface, and `density` is interpreted as surface density (mass/area). Currently only mesh geoms are supported.
PiperOrigin-RevId: 464368395
Change-Id: I17afd99b9b221c9d24ae951f6e63d5c61ff89820
Somehow tests load these libs twice, and construct/destruct globals twice.
Can be preproduced even without sanitizers with logging from GlobalModel::~GlobalModel.
themodel does not need to be visible. If it's static each instance will have own a copy.
_mjMap does not need to use string and avoid lifecycle issues at all.
PiperOrigin-RevId: 454678814
Change-Id: I8fca5e1f06cbc2b413beabcece1219445482c0ed
Yuval Tassa