ORCA parser & MolecularOrbitals as HDF5Dataset

[EBB2675]

Summary

Initialize the ORCA output parser and map representative ORCA electronic-structure calculations to the schema

Implemented

• Parse program metadata

• Parse ModelSystem

  • Atomic species and positions
  • Total charge and spin

• Parse electronic-structure methods

  • DFT and exchange-correlation functionals
  • Hartree-Fock reference form
  • MP2 and local MP2
  • Coupled-cluster methods and perturbative triples
  • Local-correlation methods, spaces, and screening thresholds
  • Orbital localization
  • CASSCF and CASCI active-space methods
  • NEVPT2 multireference perturbation methods
  • State multiplicities, roots, and state weights
  • Scalar relativistic Hamiltonians, including DKH order

• Parse basis sets

  • Orbital basis sets
  • SCF auxiliary basis sets
  • Post-HF auxiliary basis sets
  • Basis-set roles and numbers of basis functions

• Parse MolecularOrbitals

  • Orbital energies and occupations
  • AO-to-MO coefficients
  • Numbers of atomic and molecular orbitals

• Build representative outputs and workflows

  • Link molecular orbitals to the model system
  • Build serial local-correlation workflows

Testing

• Added representative ORCA outputs covering:

  • DFT
  • RI-MP2
  • DLPNO-CCSD(T)
  • CASSCF
  • CASCI and QD-NEVPT2
  • Molecular-orbital coefficients

• Added integration tests for:

  • Model systems
  • Electronic-structure methods
  • Basis sets
  • Relativistic treatments
  • Molecular orbitals

[coveralls]

Coverage Report for CI Build 27400558775

Coverage increased (+1.3%) to 75.474%

Details

• Coverage increased (+1.3%) from the base build.
• Patch coverage: 44 uncovered changes across 2 files (681 of 725 lines covered, 93.93%).
• No coverage regressions found.

Uncovered Changes

File	Changed	Covered	%
src/nomad_simulation_parsers/parsers/orca/parser.py	461	419	90.89%
src/nomad_simulation_parsers/schema_packages/orca.py	97	95	97.94%
Total (7 files)	725	681	93.93%

Coverage Regressions

No coverage regressions found.

---

Coverage Stats

Relevant Lines:	11070
Covered Lines:	8355
Line Coverage:	75.47%
Coverage Strength:	0.75 hits per line

---

💛 - Coveralls

[Copilot]

Pull request overview

This PR adds an initial ORCA output parser integration to the NOMAD simulation parser plugin, including schema-package mappings, a text parser for ORCA .out files, and tests + test data to validate parsed model systems, methods (DFT/MP2/CC/CAS), and molecular orbitals (including HDF5-backed coefficients).

Changes:

• Introduces a new ORCA parser (OrcaParser + OutReader/OutParser) capable of extracting key simulation, method, and molecular-orbital data.
• Adds an ORCA schema package with mapping annotations and registers both parser + schema package as plugin entry points.
• Adds ORCA test fixtures/data and new tests covering model systems, DFT settings, orbital energies/occupations, and MO coefficient handling.

Reviewed changes

Copilot reviewed 8 out of 15 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
tests/parsers/test_orca_parser.py	Adds unit tests for ORCA parsing, including HDF5-backed MO coefficients.
tests/data/orca/RI_MP2_water.out	Adds ORCA MP2 output fixture used by parsing tests.
src/nomad_simulation_parsers/schema_packages/orca.py	Introduces ORCA schema package mapping annotations (simulation, methods, outputs, molecular orbitals).
src/nomad_simulation_parsers/schema_packages/init.py	Registers the ORCA schema package entry point.
src/nomad_simulation_parsers/parsers/orca/text_parser.py	Adds the ORCA text extraction layer (quantities/regex blocks).
src/nomad_simulation_parsers/parsers/orca/parser.py	Adds ORCA mapping parser + post-processing/enrichment for methods, basis sets, and molecular orbitals.
src/nomad_simulation_parsers/parsers/init.py	Registers the ORCA parser entry point.
pyproject.toml	Exposes the ORCA parser and schema package via plugin entry points.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[ladinesa]

for loop?

[EBB2675]

because the MO coefficient blocks have variable column and row counts. The next block start can only be determined after consuming the current block, so the loop needs to update the token index dynamically. I added a comment to make this explicit

[ladinesa]

As promised, I put up some simplifications in #185