dataset-schema.yaml

meta:
  version: 1
  description: >-
    Structure of UppASD datasets, combined DFT and spindynamics inputs and outputs.
    Only files/directories listed in the schema are allowed in the dataset.
    File content is only validated for selected files.
filesystem-schema:
  metadata.yaml: {type: file, required: true}
  DOSCAR: {type: file, required: false}
  README.md: {type: file, required: false}
  intrinsic_properties.yaml: {type: file, required: true}
  structure.cif: {type: file, required: true}
  RSPt:
    type: directory
    required: true
    consistent_gs_mode: true  # all gs_* must either have hist OR out_MF
    schema:
      README.md: {type: file, required: false}
      common_input:
        type: directory
        required: true
        schema:
          atomdens: {type: file, required: true}
          kmap: {type: file, required: true}
          spts: {type: file, required: true}
          symcof: {type: file, required: true}
          symt.inp: {type: file, required: true}
      gs_x: &gs_dir
        type: directory
        required: true
        schema:
          data: {type: file, required: true}
          out_last: {type: file, required: true}  # output of self-consistent run
          hist: {type: file, excludes: out_MF, required: true}
          out_MF: {type: file, excludes: hist, required: true}
      # re-use definition of gs_x for gs_y (not required) and gs_z
      gs_y: {<<: *gs_dir, required: false}
      gs_z: *gs_dir
      Jij:
        type: directory
        required: true
        # The Jij directory contains data, out_last, and pairs of green.inp-* and out-* files with arbitrary suffix.
        # The following rules enforce that green.inp-* and out-* come in pairs and that no other unknown files are present.
        schema:
          data: {type: file, required: true}
          out_last: {type: file, required: true}
        allow_unknown:
          type: file
        keysrules:
          regex: '^(data|out_last|green\.inp-.+|out-.+)$'
        paired_prefix: ['green.inp-', 'out-']
  UppASD:
    type: directory
    required: true
    schema:
      MC_1: &MC_dir
        type: directory
        required: true
        schema:
          jfile: {type: file, required: true}
          momfile: {type: file, required: true}
          posfile: {type: file, required: true}
          inpsd.dat: {type: file, required: true}
          thermal.csv: {type: file, required: true}
          thermal.dat: {type: file, required: true}
      # reuse definition of MC_1 but make the following two directories optional
      MC_2: {<<: *MC_dir, required: false}
      MC_3: {<<: *MC_dir, required: false}
      README.md: {type: file, required: false}
file-schemas:
  - validator: mc_order
    files:
      - UppASD  # operate on base directory to access all MC subdirectories
    schema: {}  # the check is fully defined in mc_order, no schema is required
  - validator: mammos_entity_file
    files:
      - intrinsic_properties.yaml
    schema:
      Js: {type: mammos_entity.Entity, ontology_label: SpontaneousMagneticPolarization, unit: T}
      Ms: {type: mammos_entity.Entity, ontology_label: SpontaneousMagnetization, unit: kA/m}
      Ku: {type: mammos_entity.Entity, ontology_label: UniaxialAnisotropyConstant, unit: MJ/m3}
      Tc: {type: mammos_entity.Entity, ontology_label: CurieTemperature, unit: K}
  - validator: mammos_entity_file
    files:
      - UppASD/MC_1/thermal.csv
      - UppASD/MC_2/thermal.csv
      - UppASD/MC_3/thermal.csv
    schema:
      T: {type: mammos_entity.Entity, ontology_label: ThermodynamicTemperature, unit: K}
      Ms: {type: mammos_entity.Entity, ontology_label: SpontaneousMagnetization, unit: kA/m}
      Js: {type: mammos_entity.Entity, ontology_label: SpontaneousMagneticPolarization, unit: T}
      U_L: {type: mammos_entity.Entity, ontology_label: BinderCumulant, unit: ''}
      chi: {type: mammos_entity.Entity, ontology_label: MagneticSusceptibility, unit: ''}
      E: {type: mammos_entity.Entity, ontology_label: HelmholtzEnergy, unit: eV}
      Cv: {type: mammos_entity.Entity, ontology_label: IsochoricHeatCapacity, unit: eV/K}  # dE/dT
  - validator: csv_file
    files:
      - UppASD/MC_1/thermal.dat
      - UppASD/MC_2/thermal.dat
      - UppASD/MC_3/thermal.dat
    schema:
      sep: '\s+'
      columns: ['T', '<M>', '<M^2>', '<M^4>', 'U_{Binder}', '\chi', 'C_v(tot)', '<E>', '<E_{exc}>', '<E_{lsf}>']
  - validator: yaml_file
    files:
      - metadata.yaml
    schema:
      dataset_schema_version: {type: integer, allowed: [1], required: true}
      mammos_parser_version: {type: string, required: true}