generate_all_processes

What it is

A standalone script that generates a set of Turtle (.ttl) ontology files describing “processes” (BFO 7 Buckets compliant) from a hardcoded catalog (ALL_PROCESSES). Each process is written as a separate .ttl file under a category folder.

Public API

• ttl_string(s: str) -> str
  • Escapes Python strings for safe inclusion in Turtle literals (backslashes, quotes, newlines).
• write_process(folder, filename, label, definition, impl, who, where, when, how_know, how_is, why_roles, why_disps, steps, data_props=None, precedes=None, preceded_by=None) -> pathlib.Path
  • Writes one process ontology file to <OUTPUT_DIR>/<folder>/<filename>.ttl.
  • Returns the written file path.
• p(**kwargs) -> None
  • Appends a process definition dict into the global ALL_PROCESSES list.
• main() -> None
  • Generates all .ttl files for the processes accumulated in ALL_PROCESSES and prints a summary.

Configuration/Dependencies

• Python standard library only
  • pathlib.Path
• Output location
  • OUTPUT_DIR = Path(__file__).parent.parent / "processes"
• Hardcoded Turtle prefixes
  • Stored in PREFIXES string and prepended to every output file.
• Input data
  • All process definitions are hardcoded in ALL_PROCESSES via repeated calls to p(...).

Usage

Run the script directly to generate all process .ttl files:

# from a shell
python libs/naas-abi/naas_abi/apps/nexus/ontology/scripts/generate_all_processes.py

Or invoke programmatically:

from naas_abi.apps.nexus.ontology.scripts.generate_all_processes import main
 
main()

Caveats

• precedes parameter exists in write_process(...) but is not used to emit any Turtle content.
• Only the first item in each of who, where, when, and why_roles is used to generate object-property relationships for the process; additional entries are still emitted as classes but not linked via generated relationships.
• Output uses a fixed dc:created date ("2026-02-09"^^xsd:date) for all generated ontologies.