ps-parser

A dynamic, schema-aware library for Parasolid XT binary part files (.x_b).

The library reads the XT binary node stream against a base schema and resolves per-node schemas on the fly — handling files that embed full or delta schema definitions for node types that differ from (or extend) the base schema. It also supports writing edited documents back through the library API.

Installation

No third-party dependencies — Python 3.10+ and the standard library only.

git clone https://github.com/khoanguyen-3fc/ps-parser.git
cd ps-parser

Usage

# Default: emit the node list as JSON Lines (one object per line) to stdout
python cli.py samples/model.x_b

# Dump everything: full decoded fields per node on stdout,
# parser diagnostics on stderr
python cli.py --debug samples/model.x_b

# Use a different base schema
python cli.py --schema assets/sch_13006.s_t samples/model.x_b

# Display the topology tree
python cli.py --tree samples/model.x_b

Output

• Default — one JSON Lines object per node with all decoded fields.
• --debug — same JSON Lines output, plus detailed parser diagnostics (header, schema resolution, per-field values) written to stderr.
• --tree — ASCII topology tree to stdout. Nodes are placed using type-specific parent pointers (SHELL→body, FACE→shell, LOOP→face, etc.). A count of nodes with unrecognized types is written to stderr.

Because the default output is JSON Lines, it composes with standard tools:

python cli.py samples/model.x_b | head -1 | python -m json.tool

Default schema

The default base schema is assets/sch_13006.s_t (schema 13006). Override it with --schema. The file's schema name must be compatible with the base schema, or parsing aborts.

Project layout

cli.py                 # command-line entry point
psparser/
  __init__.py          # public API
  reader.py            # binary readers + field-type codes
  schema.py            # schema data models + base/embedded schema parsing
  parser.py            # file header + dynamic node-stream decoder
  writer.py            # XT serializer for decoded documents
  tree.py              # topology tree builder + ASCII renderer
assets/                # default base schema

Using as a library

from psparser import load_schema, parse_ps

schema = load_schema("assets/sch_13006.s_t")
with open("samples/model.x_b", "rb") as f:
    nodes = parse_ps(f, schema)

parse_ps returns a list[dict] of decoded nodes and emits diagnostics through the logging module (no printing), so callers control verbosity.

For edit/write workflows, use read_document and write_document.

Reference

The primary reference for the Parasolid XT format is the official documentation:

• Parasolid XT V35 documentation — http://www.q-solid.com/Parasolid_Docs_V35/xt_index.html

License

Released under the MIT License. Feel free to use, modify, and distribute.