Data

Tools

Indexes

Applications

Experiments

Open Sustainable Technology

Recent Releases of PowerIO

PowerIO - v0.8.0

The BMOPF schema has been released: https://github.com/distribution-system-opt/bmopf-report This release aligns powerio with BMOPF schema 0.1.0, introduces one version number for .pio.json, and distribution JSON reader validation. Note the following migration protocols:

  • .pio.json files written by 0.7.x and earlier are rejected with an error that says to regenerate them from their source case (convert with a 0.7.x install to migrate an orphaned file),
  • the BMOPF writer targets the published schema 0.1.0 $id, so consumers that key on the old $schema URI should update their accepted list.

Details

  • .pio.json carries one version number. schema_version (now 0.2.0)
    covers the whole document, model JSON included: while the major is 0 an
    incompatible change bumps the minor, and the reader accepts exactly its own
    major.minor lineage, rejecting anything else with an error that says to
    regenerate the package from its source case. The schema,
    payload_schema, and payload_schema_version fields and the
    PIO_PACKAGE_SCHEMA_URL / PIO_PAYLOAD_* constants are gone; the payload
    schema documents under docs/schema/pio-payload-* are no longer published
    (the pio-package/0.2 document embeds every model type). Files written by
    0.7.x and earlier are rejected with the regenerate error. schema_version
    is required: it used to default to the current version when absent, which
    let a document skip the lineage check by leaving the field out.
  • powerio-dist JSON reader validation (#262):
    • PMD bound arrays keep their finite entries when another entry is a
      null-derived infinity (an unbounded phase): generator
      pg_lb/pg_ub/qg_lb/qg_ub and linecode cm_ub/sm_ub no longer
      vanish whole. The PMD writer spells nonfinite entries back as null; the
      BMOPF writer drops such a field with a warning instead of coercing to 0,
      and the dss writer skips emergamps with a warning when the first
      i_max entry is nonfinite.
    • A PMD matrix column that is not an array warns and stays zero; the
      parseable columns survive instead of the whole matrix dropping to
      nothing silently. A matrix field that is not an array of columns has no
      shape to keep, so it drops, but it now names itself in a warning.
    • Dangling cross-references warn: a BMOPF or PMD element referencing an
      undefined bus, or a line referencing an undefined linecode, names the
      reference instead of parsing silently into a phantom bus.
    • parse_file/parse_str no longer route arbitrary JSON to the BMOPF
      reader: a .json without the PMD data_model marker, and without a
      BMOPF bus table beside another BMOPF table, errors (a PowerModels
      document used to parse into a bogus near-empty network). A pre-0.1.0
      feeder fragment with no voltage_source still classifies, because the
      reader accepts it. An explicit format override still forces the reader.
  • BMOPF schema 0.1.0 (bmopf-report#16). The writer targets the published
    schema $id and the reader keeps accepting the pre-0.1.0 spellings:
    • meta carries case_study_generator (was generator) and the system
      frequency; the reader also still accepts the legacy top-level
      frequency/base_frequency.
    • Load model strings are uppercase (CONSTANT_POWER, ZIP, ...).
    • Bus symmetrical component bounds are the per-sequence scalars
      vpos_min/vpos_max/vneg_max/vzero_max/vn_max; legacy
      vsym_min/vsym_max arrays map on read assuming zero/positive/negative
      order. DistBus renames the fields, part of the .pio.json 0.2.0 bump
      above.
    • Three phase transformers emit one lumped r_series/x_series pair on
      the wye base, with each winding's percent resistance referred to its own
      rating before the sum; the split _from/_to fields lost their slots.
    • Transformer taps, neutral impedance, and no load admittance relocate to
      extras.transformer.<subtype>.<name> (the schema's escape hatch); the
      IBR, control profile, DC, and time series tables relocate under extras
      the same way. The reader folds all of them back.
    • A typed capacitor element (DistCapacitor: bus, terminal_map,
      configuration, q_rated, v_nom). The DSS converter still lowers
      OpenDSS capacitors to shunt matrices; the dss and PMD writers drop typed
      capacitors with a warning.
    • Lines accept the inline impedance alternative to linecode + length
      (read into a synthesized linecode) and carry i_max/s_max. The line
      ratings map to the ENGINEERING line's own cm_ub/sm_ub (both
      directions; an inline line's ratings stay on the synthesized linecode),
      and the dss writer warns when it drops them (the normamps/emergamps
      mapping decision is #266).
    • One-triangle matrix spellings mirror on read, the shorthand
      BMOPFTools' reader also accepts; a spelled cell always wins, and both
      writers emit full matrices.
    • Generator s_max/i_max and linecode source are typed fields, read
      and written; the dss and PMD writers warn when they drop them.
    • The meta provenance fields (title, description, license, authors,
      data_sources, created, modified, provenance, version) survive a BMOPF
      round trip; the writer keeps owning $schema, frequency, and
      case_study_generator.
    • A grounded terminal counts as referenced, so the unused-terminal prune
      no longer silently drops standalone grounding.
    • The vendored schema and example networks track bmopf-report@f2e3684;
      cargo run -p powerio-dist --example regen_bmopf_examples regenerates
      the checked-in example outputs.
  • Python: Network.write_file(path, to) / DistNetwork.write_file(path, to)
    and a convert_file(..., out=...) output path write the serialized case to
    disk exactly as produced. Writing to_format text through
    open(path, "w") corrupts a CRLF source echo on Windows (text mode turns
    each \r\n into \r\r\n, which PSS/E family tools reject as malformed
    records); the new paths bypass Python's newline translation.
    DistNetwork.write_file also writes any sidecar the writer produced beside
    the case, so a dss write that emits a Buscoords directive no longer names
    a file that does not exist.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.7.3...v0.8.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 7 days ago

PowerIO - v0.7.3

Security fixes for parsing untrusted case files. The parsers are written to reject malformed input with an error, never to crash, exhaust memory, or read or write files outside the ones named on the command line; this release closes several gaps in that model.

  • A PowerWorld .aux legacy DATA header whose field list closes before it opens (] before [) no longer panics on an inverted slice. It returns a read error, matching the guard the parenthesized form already carried.
  • OpenDSS Redirect/Compile/Buscoords no longer read arbitrary local files. Parsing from a string (parse_dss_str, and the C ABI and Python string entry points) disables filesystem includes entirely, so untrusted .dss text cannot pull in a file such as /etc/passwd and echo its contents back through bus names. Parsing from a file (parse_dss_file) still follows includes, now confined to the case directory: an include that resolves outside that directory is refused with a warning, whether it climbs out with .. or is an absolute path outside the directory.
  • OpenDSS phases, windings, and wdg counts are capped. A single small property could otherwise size a dense n by n conductor matrix or a per-winding vector into the gigabytes; an oversized value clamps to the supported maximum with a warning.
  • A PMD ENGINEERING impedance or admittance matrix dimension is capped the same way, so an array of thousands of empty rows no longer demands an n by n allocation.
  • The matrix pipeline and the CLI sensitivities command sanitize the case name before it forms an output filename. A name like ../../x or an absolute path can no longer steer a write outside the chosen output directory.
  • The PowerWorld .pwd reader groups drawing records by tag in logarithmic time and bounds its substation identity search with a probe budget, so a crafted file cannot force quadratic work.
  • The OpenDSS include confinement holds for a case file parsed by bare filename: an empty case directory no longer passes every path as a prefix match, so Redirect /etc/passwd in parse_dss_file("master.dss") is refused. A case directory that itself starts with .. now confines rather than refusing everything under it.
  • OpenDSS includes are also checked after symlink resolution: a lexically contained include that is really a symlink out of the case directory is refused. parse_raw_file now confines includes exactly like parse_dss_file instead of following them anywhere on disk.
  • sanitize_stem follows Windows filename rules (trailing dots trimmed, reserved device names like con prefixed, length capped) and appends a short hash of the original name whenever sanitization changed it, so two distinct case names that sanitize alike (a/b and a_b) cannot silently overwrite each other's files in a multi case export. The gridfm dataset writer routes its output directory through the same sanitizer.
  • A DistNetwork arriving without reader caps (the model JSON C entry point deserializes one unchecked) can no longer force quadratic allocation out of linear-size input: the BMOPF writer caps the dense zero fill for an absent linecode/shunt matrix and the transformer x_sc pair expansion at 64 conductors/windings, with a warning.
  • A Network JSON bus id near usize::MAX is rejected when 3-winding transformers are present, closing an integer overflow in the synthetic star bus allocation that could alias an existing bus. An oversized piecewise cost ncost clamps instead of overflowing during normalization and Surge export.
  • The Python mcp extra pins the SDK below 2.0, which removed the mcp.server.fastmcp module the server imports.
  • OpenDSS like= splicing is capped per object. A self-referencing or mutually-referencing chain (Edit Load.a like=a repeated) otherwise doubled an object's property count each edit, so a few hundred bytes could exhaust memory; the splice is now refused with a warning past the cap.
  • The PMD writer no longer panics on a transformer with no windings, reachable from a PMD or BMOPF document with the winding array absent. The emergency rating default derives from the first winding only when one exists.
  • A PSS/E transformer COD field of an extreme magnitude (which saturates to i64::MIN) no longer overflows when the control mode is decoded; it reads as a fixed ratio.
  • The PowerWorld .aux reader routes a bus's own BusNum/Number through the same validation as every bus reference, so a fractional or out-of-range value is a read error instead of a silently truncated or saturated id.
  • A GOC3 bus uid of bus_<usize::MAX> no longer overflows when its suffix is mapped to a bus id; it falls back to the 1-based document position.
  • The MCP server's allowed-roots write check follows a symbolic link in the output filename before deciding containment. A dangling link named as the output file, sitting inside an allowed root but pointing outside it, could otherwise pass the check while the write escaped the sandbox.
  • Lifting a GOC3 time series into a package operating-point series binds the declared time_periods to the interval_duration array length (the same equality the SCOPF loader enforces). An oversized time_periods that does not match the data no longer drives an unbounded per-period allocation; the series is refused with a diagnostic and the package stays static.
  • The PMD writer caps the conductor count it expands into square matrices. A switch or voltage source terminal map is a linear model array, and a 35 KB case file naming thousands of terminals drove a 360 MB document and 1.5 GB of resident memory; the matrices now emit at 64 conductors with a warning while the terminal list itself stays faithful. Reachable from any distribution case file, not only the unchecked C entry point.
  • The dist graph builder caps the transformer winding count feeding its pair expansion, so a DistNetwork deserialized without reader caps cannot turn a linear winding array into a quadratic edge list.
  • A degenerate model matrix (rows shorter than the row count) reads as zero in the PMD writer instead of panicking on an out of range index, matching the DSS writer.
  • sanitize_stem hashes a name that already carries the disambiguating suffix shape, so the suffixed and unsuffixed name spaces stay disjoint: a case cannot be named to impersonate another case's disambiguated stem, which took no search to construct because the suffix derives from a published hash. The suffix widens to 64 bits, and the DC OPF bundle writer routes its output directory through the same sanitizer instead of a weaker local copy.
  • The CLI refuses a conversion sidecar whose path is absolute or climbs out of the output directory, closing the join before a writer can reach it.
  • Fuzz harnesses cover the distribution family (dss, pmd_json). Both parse and then write the network back, since a reader cap that does not hold shows up in the consumer that sizes an allocation from it.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.7.2...v0.7.3

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 13 days ago

PowerIO - v0.7.2

  • CLI case discovery is recursive and covers every supported format (#260):
    powerio batch and the TUI walk the input directory for .m, .raw,
    .aux, .epc, .pwb, .json, and .dss files, matched case
    insensitively, pruning hidden directories and the output directory. A file
    that fails to load during a scan is skipped with a warning instead of
    aborting the run, and the no-cases error names the supported extensions.
    Distribution cases (.dss, BMOPF/PMD JSON) parse to the multiconductor
    model and go through the explicit lower_multiconductor_to_balanced pass,
    whose approximations and dropped fields surface as warnings.
  • A leading UTF-8 byte order mark no longer defeats the JSON classifier or
    any text reader. classify_json_text, the transmission read funnel, the
    distribution parsers (including .dss and redirected files), the PMD/BMOPF
    JSON split, Network::from_json, and Package::from_json all strip it,
    and the parse warnings itemize the removal (a same-format echo returns the
    text without the mark).
  • parse_str_with_name: parse_str plus the name hint role the file stem
    plays in parse_file. The CLI now reads and classifies a .json case
    once and hands the text straight to the typed parser; a batch scan
    previously read each .json twice and DOM-parsed it three times.
  • A nameless distribution case loaded by the CLI takes its file stem as the
    network name, so batch exports no longer collide on the lowering's
    lowered-multiconductor fallback.
  • JSON reader hardening from a review pass over the family:
    • PowerModels: a status field written as a JSON boolean ("br_status": false) reads out of service instead of in service; baseMVA must be
      finite and positive; gencost rows padded to the MATPOWER matrix width are
      trimmed to the declared ncost before the per-unit unscale; an
      out-of-range cost model number passes through unscaled instead of
      wrapping into the piecewise/polynomial rescale.
    • pandapower: a trafo sn_mva of zero or below falls back to the system
      base instead of dividing the impedance by zero; string cells can no
      longer smuggle "inf"/"nan" into numeric columns.
    • egret: a polynomial cost exponent key is bounded, so a few bytes of JSON
      can no longer demand an arbitrarily large allocation or index out of
      bounds.
    • GOC3: a duplicate simple_dispatchable_device uid is rejected instead of
      silently taking another device's time series bounds and cost.
    • Surge: a float bus reference too large for an index is rejected like the
      equivalent integer instead of saturating to usize::MAX.
    • BMOPF: matrix key indices and winding counts are bounded (a crafted
      document could previously demand gigabytes or quadratic work), a missing
      line length warns instead of propagating NaN silently, and the meta
      block is kept in extras instead of dropped.
    • PMD: a data_model other than ENGINEERING (the index based MATHEMATICAL
      model) is rejected instead of being misread as ENGINEERING.
    • .pio.json: a semver build tag containing a hyphen (1.0.0+build-x) no
      longer fails the schema version check.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.7.1...v0.7.2

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 23 days ago

PowerIO - v0.7.1

  • The SCOPF Julia wire conversion is structural (#252): every struct reaching
    the wire classifies its fields (index, renamed, value) through an exhaustive
    destructure, so a new field fails to compile until classified and a value
    field reusing an index name is never renumbered. Wire output is unchanged.
  • GOC3 parses once per read (#250): the reader hands its parsed document
    forward as Parsed::document, and the package boundary derives the
    operating point series from it instead of reparsing the retained text.
  • DcOpfInstance carries the constant cost term c0 (generator and nodal
    data), and the DC OPF bundle writes c0.mtx/c0_gen.mtx (bundle schema
    0.3.0).
  • Balanced formats harvest and emit coordinates (#183): PowerWorld aux
    Latitude:1/Longitude:1, pandapower bus geo Point strings, PyPSA
    buses.csv x/y, each in both directions. Writers with no geometry concept
    warn that locations were dropped.
  • The standalone geographic document (#184): GeoLayer/ElementKey with
    tolerant reads (headerless buscoords CSV, aliased CSV/JSON records, GeoJSON)
    and canonical GeoJSON writes (.geo.json, the powerio_geo foreign
    member); DisplayFormat::GeoJson/DisplayData::Geo;
    Network::geo_layer()/apply_geo_layer() with multiconductor equivalents
    in powerio-pkg; Branch.route/DistLine.route polyline routing (payload
    schemas 1.2.0); .pwd promotion (geo_layer_from_pwd,
    apply_substation_points, pwd_mercator_to_lonlat); and
    powerio geo extract | apply | convert in the CLI.
  • C ABI and Python bindings for the new surfaces (#249, #185), all additive
    (PIO_ABI_VERSION stays 4): pio_acopf_from_network / pio_acopf_to_json
    / pio_acopf_instance_free, pio_geo_parse / pio_geo_extract /
    pio_geo_apply, and pio_dist_geo_extract / pio_dist_geo_apply. Python
    gains parse_geo, Network.geo_layer()/apply_geo_layer()/acopf_instance(),
    and the distribution equivalents.
  • DeepMind OPFData reader (#258): SourceFormat::DeepMindOpfDataJson reads
    one raw OPFData JSON document as a solved snapshot, echoes the same format
    byte for byte, and converts through the standard surface (CLI, C ABI,
    Python).

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.7.0...v0.7.1

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 1 month ago

PowerIO - v0.7.0

  • Add powerio-prob for complete numerical problem instances (#238). Its
    default build is matrix free; the matrix feature adds sparse projections,
    DC OPF bundle output.
  • Move DC OPF instance types and bundle output out of powerio-matrix.
  • Keep solver formulations and KKT operators outside powerio-prob.
  • Demote powerio-json from the public case format surface (#229). It leaves
    the CLI format help and the generated format tables; pio_parse_str(..., "powerio-json", ...) and pio_to_format(..., "powerio-json", ...) keep
    working as ABI v4 compatibility aliases, and pio_to_json / pio_from_json
    are the documented balanced model JSON API.
  • NetworkPackage::from_balanced is format neutral. Source adapters, such as
    GOC3 operating point extraction, run only for parsed reader input through
    from_parsed_balanced.
  • Building a SCOPF instance from text requires an explicit source format.
  • Add AC OPF problem instances (#248): powerio-prob gains AcOpfInstance
    and build_ac_opf_instance, carrying pi model branch data with separate tap
    and shift, per terminal charging, bus shunts, active and reactive demand,
    voltage bands, generator PQ bounds, and quadratic cost including the
    constant term (GenCost::quadratic_with_constant). Relaxations such as SOC
    forms consume the same instance. Matrix free; C ABI and Python exposure is
    #249.
  • powerio-prob first publish review fixes: reserve membership sets now
    assign zone indices from the same document order as the reserve rows
    (sorted order previously crossed n_p/n_q between the two tables and
    diverged from src/goc3.jl past nine zones); a GOC3 branch with
    r = x = 0 is rejected by name instead of writing NaN into the wire
    rows; a missing device_type defaults to producer, the balanced
    reader's rule; the AC instance folds self-loop branch admittance into
    the bus shunt vectors, matching build_ybus; both instance builders
    reject a non-positive base MVA before scaling; the DC OPF bundle
    directory name stays confined to the output directory, and the bundle
    manifest reports the powerio core version (powerio::VERSION, new).
    SCOPF row structs and DcOpfOutputs are #[non_exhaustive].
  • DSS reader: linecode= now sets a line's conductor count the way the
    engine's FetchLineCode does, the later of phases=/linecode= winning.
    A 4-wire line without an explicit phases= keeps its neutral instead of
    truncating the terminal map to 3 against a 4x4 linecode
    (frederikgeth/BMOPFTools.jl#332); thanks @frederikgeth

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.6.3...v0.7.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 1 month ago

PowerIO - v0.6.3

  • Arrow matrix axes (#234): the C ABI Arrow export gains a table catalog and
    dense matrix axis maps. pio_to_arrow exposes matrix_bus and
    matrix_branch axis tables alongside the ybus, incidence, bprime, and
    bdoubleprime COO tables, which carry powerio.row_axis / powerio.col_axis
    schema metadata so a consumer maps dense matrix rows and incidence columns
    back to source bus and branch rows. Matrix rows are labeled with the
    matrix_bus axis, which stays correct when 3-winding transformer star-point
    lowering expands the bus set past the handle bus order.
  • FDPF matrices (#234): bprime / bdoubleprime follow MATPOWER makeB
    semantics, with self-loop handling and asymmetric Matrix Market writes pinned
    by Rust, C ABI, and Python coverage.
  • Summary JSON (#234): the C ABI exposes balanced (pio_summary_json) and
    distribution summary JSON, so a binding can render network summaries without
    materializing the full model payload.
  • Matrix Arrow export (#234): the numeric matrix export path fills Julia-owned
    buffers directly for the copy-free fast path.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.6.2...v0.6.3

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 1 month ago

PowerIO - v0.6.2

  • Normalization (#210): angle bound clamping now keeps every repaired branch
    interval ordered. One sided intervals wholly outside the supported window are
    widened to the configured pad instead of producing angmin > angmax; Rust,
    C ABI, and Python normalize option coverage pin the behavior.
  • Binding coverage (#185): the already shipped study block and distribution
    graph projection now have C ABI and Python accessors. The remaining geo
    binding symbols stay in the v0.6.3 follow through.
  • BMOPF diagnostics (#219): distribution conversions now carry structured
    diagnostics alongside warning strings, and transformer export losses expose
    stable EMIT.BMOPF.* codes for downstream tests and capability checks.
  • BMOPF transformer fidelity (#214, #215, #216, #217): OpenDSS fixed
    transformer taps, center tap convention fields, delta_wye leakage referral,
    and n_winding delta_roll now export directly in BMOPF form with regression
    coverage against schema valid output and unaffected fixture byte identity.
  • BMOPF source fidelity (#218): per phase OpenDSS voltage sources on the same
    bus merge into one BMOPF voltage_source when their phase angles are
    coherent; ambiguous, bounded, priced, or conflicting source banks stay split
    with warnings.
  • Distribution capabilities (#213): the C ABI dist feature exposes
    pio_dist_capabilities_json, reporting the six BMOPF fidelity flags that
    PowerIO.jl and downstream tools can probe at runtime.
  • Geographic fields (#180): balanced and distribution models now share typed
    GeoMeta / Location JSON shapes. Network.geo, Bus.location,
    DistNetwork.geo, and DistBus.location are optional and omitted when
    absent. OpenDSS Buscoords and BMOPF longitude/latitude sideloads promote into
    typed bus locations; OpenDSS writes a Buscoords sidecar when locations are
    present, while BMOPF longitude/latitude output remains opt in and only emits
    declared geographic coordinates.
  • .pio.json model JSON: the balanced and multiconductor payload schema
    versions move from 1.0.0 to 1.1.0 for the additive geographic fields.
    The package metadata schema, C ABI version, and Python package surface stay
    in the 0.6 compatibility band.
  • JSON strategy: .pio.json docs now state that it is PowerIO's compiled
    artifact, not a case format; payload schemas are for validating model JSON
    inside .pio.json documents; powerio-json remains supported, is deprecated
    for CLI file handoffs, and is no longer shown in the PR conversion matrix.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.6.1...v0.6.2

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 1 month ago

PowerIO - v0.6.1

  • CI: added wasm32 coverage for the core Rust crates (#186), external BMOPF
    JSON Schema validation for emitted distribution documents (#192), and
    generated .pio.json / model JSON schema drift checks (#178).
  • Distribution fidelity (#197): OpenDSS and BMOPF writers preserve transformer
    winding voltage bases, no load admittance, tap settings, neutral impedances,
    and multi winding transformer structure. Roundtripped OpenDSS decks now run
    through a solve oracle that checks voltage agreement, load voltage model
    behavior, and neutral return handling.
  • Distribution DER mapping (#197): typed IBR and control profile data now round
    trips through OpenDSS PVSystem / Generator / InvControl and BMOPF
    ibr / control_profile records, with warnings for unsupported control
    details.
  • .pio.json documents (#181, #193): added the study block for replayable
    balanced model edits, materialization helpers, deterministic uid stamping,
    and balanced reader warnings as structured .pio.json diagnostics.
  • Normalization (#188): added an opt-in angle bound clamp pass with Rust, C ABI,
    and Python entry points; existing normalization behavior is unchanged.
  • Distribution graph projection (#182): added a bus and terminal graph view for
    DistNetwork, including transformers, open switches, and terminal metadata.
  • Matrix bindings (#190): added Arrow C ABI matrix exports as COO triplet tables
    for Ybus, incidence, B', and B'', with C and Python golden coverage.
  • Sensitivities (#8): added sparse and iterative PTDF/LODF export paths with
    drop tolerance metadata, while retaining the dense path as the small case
    oracle.
  • Documentation (#191): standardized released docs, READMEs, and crate metadata
    around .pio.json document, model JSON, and metadata terminology.
    powerio-py continues to inherit the workspace version; no separate Python
    version bump is needed.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.6.0...v0.6.1

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 1 month ago

PowerIO - v0.6.0

  • Breaking (#175): ElementRef.row is Option<usize>, the honest form of the
    0.5.1 wire semantics. None addresses by identity alone (refs built with
    by_source_uid); the private wire-presence shim (wire_row()) is gone, and
    row itself says whether the wire carried one. The .pio.json wire format
    is unchanged. The other break collected in #175, keyed-object addressing for
    multiconductor operating point updates, needs design and moves to the 1.0
    window (#196).
  • C ABI: the package payload extraction inverses land as additive symbols (no
    ABI version change; probe the symbols like the other feature surfaces):
    pio_package_to_balanced_network and pio_package_to_multiconductor_network
    materialize an owned network handle from a parsed .pio.json package handle,
    the inverses of the pio_package_from_* constructors. A handle built from a
    payload retains no source text, so a same-format write is a fresh
    serialization rather than a byte-exact echo; the multiconductor payload's
    parse warnings ride along.
  • C ABI: pio_to_json / pio_from_json are the function form of the balanced
    model JSON (byte identical to the powerio-json writer); the format token
    remains as a compatibility alias for file based workflows. This is the
    additive half of #194; retiring the token waits for 1.0.
  • C ABI: pio_dist_to_json / pio_dist_from_json serialize a distribution
    handle to its model JSON and back in one call each: the same object a
    .pio.json document carries under model.multiconductor_network, without
    the surrounding document. Bindings materialize element tables with this
    instead of building a throwaway package; it is not a case format (the
    converter, CLI, and inference do not know it), so BMOPF JSON remains the
    distribution JSON exchanged with other tools.
  • C ABI: pio_classify_str classifies in-memory JSON by the same top level
    markers the transmission parser's .json sniffing uses, and recognizes
    .pio.json envelopes: transmission:<format>, distribution:<format>,
    package, ambiguous, or unknown, size-then-fill. Bindings can route a
    bare .json before choosing a parser instead of matching error text.
  • The JSON classifier reports a .pio.json envelope as its own outcome
    (routing::JsonClass), so every consumer handles it: the CLI, the Python
    readers, and the Python classify_json_text now name the package surface
    for an envelope instead of a generic cannot-infer error (or, for the Python
    string reader, a MATPOWER syntax error). Envelope detection requires
    model_kind to be balanced or multiconductor, so a case document
    carrying those key names with other values still classifies as a case, and
    classification parses the document once.
  • Directed errors at the transmission boundary: a .dss path, a distribution
    from token (dss/pmd/bmopf), and a .pio.json envelope handed to the
    balanced parser now name the surface that reads them instead of a generic
    unknown-format message.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.5.1...v0.6.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 1 month ago

PowerIO - v0.5.1

The .pio.json payload schema is declared and versioned independently of the envelope (payload_schema, payload_schema_version; envelope schema_version 0.1.1). Payload rows carry stable uid identities. Operating point updates resolve by identity, a present row is a consistency check, and unknown identities are rejected at materialization and by validation. Packages from 0.5.0 read unchanged.

Python table dicts expose uid. No C ABI signature changes. Docs are restructured and served at https://powerio.dev.

Details in CHANGELOG.md.

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by samtalki about 1 month ago

PowerIO - v0.5.0

  • powerio-pkg: NetworkPackage is the one package type name (CompilerPackage
    is gone); the Julia binding already leads with NetworkPackage. The .pio.json
    format is unchanged.
  • Python API: the seven module level package_* functions are replaced by the
    powerio.Package handle class, which parses the envelope once and exposes
    model_kind, operating_points(), materialize_operating_point(),
    as_balanced()/as_multiconductor(), validate(), validation(),
    diagnostics(), and the multiconductor to balanced preflight and lowering.
  • .pio.json operating points: the per point label and duration_hours fields
    are gone; time_axis.labels and time_axis.duration_hours (indexed by
    points[].index) are the one source of truth. Readers ignore the old fields.
  • Transmission formats: added GOC3 JSON input and Surge JSON read and write paths.
    GOC3 packages lift source time series into .pio.json operating_points,
    and package APIs can materialize one point into a static package.
  • GOC3 reader fixes: branches with additional_shunt keep the line charging
    (b/2 per terminal added to the extra shunts, per the GO Challenge 3
    formulation); ta_lb/ta_ub map to an ActiveFlow transformer control
    range instead of fabricating angmin/angmax bus angle limits; producers
    and consumers honor initial_status.on_status like every other record type;
    object form section keys sort under a total order (mixed numeric and non
    numeric keys no longer risk a sort panic).
  • powerio-pkg: GOC3 operating point extraction now consumes the parser's own
    document walking (device_rows, section, cost_at shared through a
    bridge), so update row indices match the payload by construction, including
    devices without a uid. A failed extraction attaches a
    READ.GOC3.OPERATING_POINTS_DROPPED diagnostic instead of silently
    producing a static only package. Materialized packages clear package_id
    (the parent id lives in origin.parent_package_id).
  • PSS/E .raw: revision aware record layouts for v34/v35 transformer winding
    lines (twelve ratings, NODE), v35 generator records (NREG, BASLOD),
    and v35 switched shunts (NREG, per block status triples), on both read and
    write; the 2W/3W transformer split accepts float form K fields. The
    case14_v34.raw/case14_v35.raw fixtures are regenerated in the genuine
    layouts.
  • PSLF .epc writer: parallel loads and shunts on one bus get distinct ids
    (extras["id"] preferred, positional fallback); the reader captures load,
    shunt, and SVD ids into extras["id"] so they survive cross format writes.
  • PowerWorld .pwb: the table location search runs under a work budget, so a
    crafted file fails with a read error instead of pinning a core for hours.
  • Surge JSON writer warns when named branch rating sets are dropped, like every
    other lossy writer.
  • Writing a read only format (goc3-json) returns the new
    Error::WriteUnsupported instead of a misleading UnknownFormat.
  • C ABI: the panic guard now covers index construction in the parse entry
    points; pio_package_validate documents its exclusive access requirement
    (the one non const entry point) and the header preamble scopes the
    concurrent read guarantee accordingly; PioDistNetwork gains the same
    compile time Send + Sync assertion as the other handles.
  • SourceFormat::name() is the one source format name mapping; the package,
    CLI, and Python copies are gone.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.4.0...v0.5.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 1 month ago

PowerIO - v0.4.0

  • powerio-pkg: .pio.json reads now enforce the envelope compatibility rule:
    same major schema_version values load, while incompatible major versions
    fail before payload use. The mdBook schema page documents the rule.
  • powerio-pkg: balanced package output now emits source maps for stable bus,
    load, shunt, branch, and generator fields; validation diagnostics attach the
    matching source reference where a map exists. Format folded fields use
    mapping kinds such as split, and defaulted fields are not marked as exact
    source fields.
  • Converter tests now compare stable per element values across writable legacy
    formats, not only counts and totals. PSLF export now warns when transformer
    charging admittance is dropped.
  • powerio-dist BMOPF: OpenDSS fixed P/Q generators now emit as BMOPF
    generator.* entries with pinned P/Q bounds instead of negative load.*
    entries. The old negative load warning is gone; generators without source
    costs keep the existing cost 0 warning.
  • Python API: removed the one release powerio.Case and
    powerio.dist.DistCase compatibility aliases. Use powerio.Network /
    powerio.BalancedNetwork and powerio.dist.MulticonductorNetwork /
    powerio.dist.DistNetwork.
  • No C ABI rename in this migration slice: PIO_ABI_VERSION stays 4 and
    PIO_DIST_ABI_VERSION stays 1.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.3.3...v0.4.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 2 months ago

PowerIO - v0.3.3

  • MCP server: unified the advertised tool surface to semantic verbs:
    convert, save, summary, parse, normalize, matrix, and display.
    The tools route transmission cases, distribution cases, PyPSA CSV folders, and
    gridfm datasets by format. Distribution parse returns canonical bmopf-json
    as its serial transport; transmission parse returns powerio-json.
    summary now has one canonical JSON schema across MCP and the CLI's new
    powerio summary command. Gridfm is a format behind parse/save, not its
    own MCP tool. PowerWorld .pwd display files parse through display, leaving
    room for a future open display format without renaming the tool. Older case,
    matrix, and PyPSA helper names stay as direct Python compatibility callables
    for one release, but they are no longer advertised as MCP tools.
  • Python API: restored the undocumented powerio.Case = Network alias for one
    release, but left it out of __all__ and docs; remove it in 0.4.0. The
    experimental distribution surface now uses powerio.dist.DistNetwork as
    the primary name to match the native DistNetwork hub type, while the
    exported powerio.dist.DistCase = DistNetwork alias stays for one release.
    powerio.dist is gated on the draft BMOPF schema (PIO_DIST_ABI_VERSION = 1)
    and not yet under the stability guarantee.
  • No C ABI change: PIO_ABI_VERSION stays 4 and PIO_DIST_ABI_VERSION stays 1,
    and the matrix builders are unchanged. The native extension's internal pyclass
    names changed (PyCase → PyNetwork, _DistCase → _DistNetwork) so repr()
    now renders the public Network(...) / DistNetwork(...) form directly; a
    rebuilt wheel is required.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.3.2...v0.3.3

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 2 months ago

PowerIO - v0.3.2

  • powerio-dist OpenDSS: grounding reactors written from a bus terminal to the
    same bus's node 0 now type as shunts in BMOPF instead of staying untyped.
    Impedance form reactors use the equivalent admittance matrix, so neutral
    grounding resistors survive DSS to BMOPF conversion.
  • powerio-dist OpenDSS: delta capacitor and reactor banks now type as shunt
    admittance matrices, including off diagonal branch terms, instead of being
    dropped as untyped objects.
  • DSS writing now regenerates conductance bearing shunts as grounding reactors
    and preserves delta shunts as conn=delta where the typed model identifies
    them.
  • No core or distribution C ABI break; PIO_ABI_VERSION stays 4 and
    PIO_DIST_ABI_VERSION stays 1.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.3.1...v0.3.2

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 2 months ago

PowerIO - v0.3.1

  • Parser warnings: PSS/E and PowerWorld .aux parse warnings now surface
    through Parsed::warnings and the C ABI's pio_warnings path instead of
    living only in docs or writer warnings.
  • PSS/E: hardened record tokenization and continuation handling. Slash
    characters inside quoted fields are no longer treated as comments; incomplete
    transformer and two-terminal DC continuation records now error clearly instead
    of consuming section terminators; transformer records with non-unit CW/CZ
    now warn that impedance and turns values were read without conversion.
  • PSS/E: load ZIP components and v34/v35 load tail fields are retained in extras
    and replayed on write. If typed load p/q no longer match retained
    PL/QL/IP/IQ/YP/YQ, the writer emits typed constant power and reports the
    stale extras instead of replaying wrong source components.
  • PSS/E: quoted IDs, names, and HVDC names are sanitized before duplicate ID
    allocation, so collisions created by replacing quotes or / are handled
    deterministically and reported in conversion warnings.
  • Normalization: generator cost per-unit scaling now dispatches through explicit
    cost models, and slack bus selection ignores NaN generator pmax values
    when choosing among candidate reference buses.
  • PSLF and PowerWorld AUX tokenization: quoted / and // text is kept as data
    rather than treated as continuation or comments. PowerWorld .aux now reports
    unmodeled DATA blocks as parse warnings while retaining source text for
    same-format writeback.
  • powerio-dist OpenDSS: quoted comment markers are preserved in lexer values,
    indented block comments are skipped, capacitor and reactor kvar shunts share
    validation, reactors with kvar/kv map to typed shunts with negative
    susceptance, and invalid shunt forms stay untyped with explicit warnings.
  • powerio-dist BMOPF: fixed OpenDSS generators with fixed P/Q setpoints now
    encode as negative BMOPF loads with warnings. The vendored draft schema was
    refreshed for multi-digit matrix keys, corrected $id, and nonnegative
    switch i_max, so 10-conductor linecode output validates without the old
    schema warning.
  • C distribution ABI v1 (PIO_DIST_ABI_VERSION 1): direct pio_dist_* callers
    get a separate version check; the supported one-shot conversion order is
    (input, from, to, ...).
  • C ABI tests now reject the old target-before-source conversion order for both
    pio_convert_* and pio_dist_convert_*, including the compiled C smoke test
    against powerio.h.
  • C ABI hardening: unit tests pin every public PIO_* macro, opaque typedef,
    and pio_* prototype in powerio.h; Cargo now checks Rust source/header
    symbol parity; CI builds the no-default core ABI plus the release
    arrow,gridfm,dist feature smoke test and C++ header/link sanity checks.
  • No core C ABI break; PIO_ABI_VERSION stays 4. No existing Rust or Python
    API was removed or reordered.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.3.0...v0.3.1

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 2 months ago

PowerIO - v0.3.0

  • Distribution systems: new powerio-dist crate for multi conductor unbalanced
    networks. Reads OpenDSS and the PowerModelsDistribution engineering JSON, and
    reads/writes the IEEE BMOPF Taskforce JSON (schema v0.0.1). First crates.io
    release of powerio-dist. thanks @frederikgeth @deakinmt @matheusduartedm
  • PSS/E: read and write support for v34 and v35 alongside v33.
  • GE PSLF: an .epc writer, with better interoperability between PSLF and PSS/E.
  • Transformers with three or more windings.
  • C ABI v4 (PIO_ABI_VERSION 4): a smaller canonical surface designed so future
    changes stay additive. Breaking ABI change in this release.
  • Memory safety hardening across the readers.

What's Changed

New Contributors

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.2.4...v0.3.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 2 months ago

PowerIO - v0.2.4

  • PSLF .epc: read support for GE PSLF power flow cases, including .epc
    extension inference and pslf / epc input aliases. The reader is read only
    and keeps source text plus warnings for sections outside Network.
  • PowerWorld .pwb: expanded binary reader coverage across older and newer
    header constants, with stricter record probes, companion format parity checks,
    and clearer rejection of unsupported vintages.
  • PowerWorld .pwd: display parsing keeps the separate display API path and
    retains the malformed input invariant: corrupt or truncated display files
    return a structured error or a parsed display, not a panic.
  • No C ABI break; PIO_ABI_VERSION stays 3.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.2.3...v0.2.4

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] about 2 months ago

PowerIO - v0.2.3

  • Normalization: Network::to_normalized preserves source bus ids instead of
    renumbering surviving buses to dense 1-based ids. Dense row mapping remains
    available through IndexedNetwork and the C ABI table order.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.2.2...v0.2.3

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 2 months ago

PowerIO - v0.2.2

  • Display API: parse_display_file / parse_display_bytes read display
    artifacts separately from network cases. PowerWorld .pwd returns
    DisplayData::PowerWorld(PwdDisplay) in Rust and
    DisplayData("powerworld", PwdDisplay(...)) in Python. parse_file
    remains Network only and points .pwd callers at the display API.
  • PowerWorld AUX: name keyed complete case exports can resolve
    BusName_NomVolt labels for loads, shunts, generators, and branches.
  • PSS/E: the reader accepts comment headers, system wide records before
    BEGIN BUS DATA, and v34 named branch records without misclassifying
    long v33 branch rows.
  • MCP: add dedicated tools for PyPSA CSV folders and gridfm Parquet datasets.
  • DC sensitivities: PTDF/LODF fall back to dense Gaussian elimination for
    invertible indefinite grounded Laplacians.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.2.1...v0.2.2

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 2 months ago

PowerIO - v0.2.1

Hardening fixes only; no API or ABI change (PIO_ABI_VERSION stays 3).

  • MATPOWER: a crafted gencost NCOST (e.g. 1e20) overflowed the row
    width arithmetic and panicked on every build profile, a denial of
    service on untrusted input through the Rust API and the CLI. The width
    now saturates and the row is rejected as a ShortRow parse error.
    Found by malformed input fuzzing.
  • C ABI: error and warning messages were clipped at a raw byte count,
    which could split a multibyte UTF-8 character and hand the caller an
    invalid string. Truncation now lands on a character boundary.
  • PowerWorld .pwd: the reader's byte accessors return Option instead
    of indexing, so an out of range offset from a corrupt file rejects the
    record instead of panicking. A corruption sweep test pins the
    invariant; the differential oracle tests pass unchanged.
  • powerio.h: a doc comment contained a literal */ that terminated
    the generated block comment, so compiling with -DPIO_GRIDFM against
    the shipped 0.2.0 header failed with unknown type name 'raw'.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.2.0...v0.2.1

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 2 months ago

PowerIO - v0.2.0

  • PowerWorld .pwb binary reader (#95, #102, #105): read only, covering
    June 2016 through 2022 era exports under header constants 425, 483, 508,
    537, 550, and 551, parity tested against same vintage .aux/.RAW/.m
    siblings up to the 6717 bus Texas7k. Unsupported writer vintages are
    rejected with the format constant named.
  • pandapower JSON converter (#106): read and write pandapowerNet JSON.
    Written trafo parameters reproduce the source Y_bus exactly through
    pandapower 3.x's transformer model, ZIP load columns go out in both the
    <= 3.1 and >= 3.2 namings, and CI validates the converter against
    pandapower itself over the vendored fixtures.
  • PyPSA CSV folder converter (#106): read and write the static network
    CSV folder, CI validated against PyPSA over the vendored fixtures.
    Folders parse through parse_file(..., "pypsa-csv"), auto-detected for
    a directory holding network.csv; the CLI takes --from pypsa-csv and
    --to pypsa-csv -o <dir>.
  • PowerWorld .pwd display reader (#102): substation diagram coordinates,
    matched 1-1 against the aux substations on every probed save with a same
    vintage aux (the v19 resave matches 1248/1250 against the published
    case, a vintage skew).
  • Read fidelity channel (#106): parse_file/parse_str return
    Parsed { network, warnings }, so what a reader cannot carry is
    itemized instead of dropped silently. Python exposes
    Network.read_warnings and the MCP tools report it; the C ABI gains
    pio_parse_warnings and pio_write_pypsa_csv_folder (additive, ABI
    version stays 3).
  • Full .aux fidelity (#95): all three field naming generations through
    Simulator 21+, validated against the vendored ACTIVSg200 set.
  • docs/powerworld.md records the decode evidence, mapping notes, and the
    coverage matrix the corpus tests assert.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.1.1...v0.2.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 2 months ago

PowerIO - v0.1.1

  • File extension detection is case-insensitive (#97, #101): parse_file
    accepts .RAW/.M/.JSON/.AUX and any mixed case alongside the
    lowercase forms, and the CLI batch discovery and TUI file browser find
    such files too. Reported by @jd-foster.
  • MCP server error hardening (#93): an unreadable input file surfaces as
    the documented ValueError shape instead of a raw PermissionError, with
    defensive guards on the JSON load and matrix dispatch paths.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.1.0...v0.1.1

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 2 months ago

PowerIO - v0.1.0

  • gridfm read path (#70): read_gridfm_dataset / read_gridfm_scenarios /
    gridfm_base_case in powerio-matrix, pio_read_gridfm /
    pio_gridfm_scenario_ids in the C ABI behind --features gridfm, and
    powerio.read_gridfm / read_gridfm_scenarios in Python. Release tarballs
    now build the C ABI with the gridfm feature, so the symbols ship to the
    Julia bindings.
  • convert_str (#88): in-memory conversion through the hub in Rust and
    Python; the MCP server's inline conversion no longer stages temp files.
    Closes #66.
  • The MCP server grows from two tools to eight (#90): parse_case,
    normalize_case, and case_to_json emit the JSON transport,
    compute_matrix returns nine sparse kinds in COO form, dense_view
    returns the dense table view, and save_case writes converted cases to
    disk; convert_case and case_summary are unchanged.
  • Docs (#92): Pages landing page with the released/development split, guide
    links, and the logo; the crate homepage points at the docs site; release
    drafts carry the CHANGELOG section instead of a bare title.

What's Changed

Full Changelog: https://github.com/eigenergy/powerio/compare/v0.0.1...v0.1.0

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 2 months ago

PowerIO - v0.0.1

Energy Systems - Energy Data Accessibility and Integration - Rust
Published by github-actions[bot] 2 months ago