StepParser

StepParser is a C# CLI for ISO 10303-21 clear-text STEP files. It tokenizes, lexes, and parses STEP exchange files, with an AP242 (ISO 10303-242) Managed Model-Based 3D Engineering entity catalog used to classify PMI/MBD content.

Features

• Tokenization with block-comment (/* … */) stripping and precise source coordinates
• Lexer covering STEP structural keywords, string/enum/binary literals, entity references, and typed parameters including Unicode escape sequences (\X2\, \X4\)
• Recursive-descent parser for HEADER and DATA sections that reports diagnostics instead of throwing
• Recovery from non-conformant complex entity instances (missing outer parentheses per ISO 10303-21 §11.3.3) — emits a WARNING diagnostic and continues rather than failing
• AP242 semantic model: ~600 entity types across geometry, topology, representations, product structure, assemblies, units/measurement, PMI dimensions, GD&T, datums, shape aspects, presentations, materials, and kinematics — surfaced as the semantics section of the output
• JSON and text output modes
• AOP verbose debug logging via Fody/Serilog (--log / --log-file)
• Corpus sweep mode (--sweep) for bulk processing of .stp/.step files

Build

$env:DOTNET_CLI_HOME="$PWD\.dotnet_home"
$env:DOTNET_SKIP_FIRST_TIME_EXPERIENCE="1"
dotnet build StepParser.slnx --ignore-failed-sources

Test

$env:DOTNET_CLI_HOME="$PWD\.dotnet_home"
$env:DOTNET_SKIP_FIRST_TIME_EXPERIENCE="1"
dotnet test StepParser.slnx

Run a single test with dotnet test --filter "FullyQualifiedName~ParsesMinimalSample".

Coverage and mutation testing:

dotnet test StepParser.slnx --collect:"XPlat Code Coverage"
dotnet stryker                        # requires: dotnet tool restore

Run

$env:DOTNET_CLI_HOME="$PWD\.dotnet_home"
$env:DOTNET_SKIP_FIRST_TIME_EXPERIENCE="1"
dotnet run --project src/StepParser/StepParser.csproj -c Debug --no-build -- sample.stp

CLI Options

Usage: StepParser [OPTIONS] <input-file>

Options:
  -o, --output <file|auto>   Write output to a file. Use 'auto' for <input>.json beside the input.
  -f, --format <fmt>         Output format: json (default), text
  --phase <phase>            Stop after phase: tokens, lex, parse (default: parse)
  --strict                   Treat warnings as errors (exit 2; prints annotation to stderr)
  --no-color                 Disable ANSI colour in text output
  -v, --verbose              Print progress to stderr
  --log                      Enable AOP verbose debug logging
  --log-file <path>          Log file path (default: <input>.log beside the input file)
  -h, --help                 Show help
  --version                  Show version

Corpus sweep:
  --sweep <csv>              Parse every path read from stdin; write per-file results to <csv>

Exit codes: 0 clean, 1 input file not found, 2 errors (or warnings under --strict), 3 warnings only, 4 bad arguments.

Text Output

Text format (--format text) includes:

• Source path, detected schema, STEP edition
• Errors: N | Warnings: N counts (or Diagnostics: 0 when clean)
• All diagnostic messages with severity, line:column, and message
• An AP242: … | MBD: … | PMI: N | GD&T: N summary line, plus any detected dimensions and datums
• Entity-type frequency breakdown (top 20 by count)

Sweep Local STEP Files

$files = (@(es.exe *.stp) + @(es.exe *.step)) | Sort-Object -Unique
$files | dotnet run --project src/StepParser/StepParser.csproj -c Debug --no-build -- --sweep .\step_sweep_results.csv

License

GPL-3.0 — see LICENSE.