| """Build small authored behavior examples; these fictional packets are not factual training data.""" |
| import json |
| from pathlib import Path |
|
|
|
|
| def build(): |
| |
| |
| specifications = [ |
| ("alder", "train", "Alder Valley Observatory", "an astronomical observatory", "2008", |
| "opened to visitors", "2014", "added a public telescope-viewing programme"), |
| ("birch", "train", "Birch Harbour Library", "a public library", "1996", |
| "opened its reading room", "2011", "introduced a local-history collection"), |
| ("cedar", "train", "Cedar Ridge Railway Museum", "a transport museum", "1982", |
| "opened a collection of railway equipment", "2005", "added a carriage-restoration workshop"), |
| ("dune", "train", "Dune Field Wetland Reserve", "a wetland reserve", "2001", |
| "opened a marked visitor trail", "2017", "introduced seasonal bird counts"), |
| ("elm", "train", "Elm River Chamber Ensemble", "a chamber music ensemble", "1993", |
| "gave its first documented public concert", "2010", "began an annual series of school concerts"), |
| ("fern", "train", "Fern Coast Maritime Archive", "a maritime archive", "1978", |
| "opened a reading room", "2009", "began digitising shipping registers"), |
| ("glen", "validation", "Glen Brook Textile Museum", "a textile museum", "1987", |
| "opened a loom display", "2013", "introduced demonstrations of weaving techniques"), |
| ("hazel", "validation", "Hazel Plain Seed Library", "a community seed library", "2006", |
| "started lending seed packets", "2018", "added workshops on seed storage"), |
| ("iris", "test", "Iris Bay Community Theatre", "a community theatre", "1999", |
| "staged its first documented production", "2016", "introduced a youth drama workshop"), |
| ("juniper", "test", "Juniper Hill Geological Centre", "a geology education centre", "2003", |
| "opened a rock collection", "2019", "added guided geology walks"), |
| ] |
| packets = [] |
| for key, split, name, category, first_year, event, second_year, activity in specifications: |
| topic = f"{name} (fictional practice topic)" |
| rows = [ |
| [f"Practice Gazette report about {name}", f"https://example.org/practice/{key}/report", |
| f"Fictional practice material: {name} is {category}. It {event} in {first_year}."], |
| [f"Practice Review account of {name}", f"https://example.org/practice/{key}/review", |
| f"Fictional practice material: {name} {activity} in {second_year}. No founder or visitor total is stated."], |
| ] |
| article = { |
| "lead": [{"text": f"{name} is {category} in a fictional practice setting.", "sources": ["S1"]}], |
| "sections": [{"heading": "Development", "paragraphs": [ |
| {"text": f"The organisation {event} in {first_year}.", "sources": ["S1"]}, |
| {"text": f"It {activity} in {second_year}.", "sources": ["S2"]}]}], |
| "gaps": ["The supplied excerpts do not identify a founder or provide visitor figures.", |
| "These are fictional practice sources, not evidence of a real topic's notability."], |
| } |
| outline = {"sections": [{"heading": "History and activities", "questions": [ |
| "Which dated milestones do the supplied excerpts support?", |
| "What further independent coverage would a human editor need?"], "sources": ["S1", "S2"]}], |
| "gaps": ["Do not draft article prose for English Wikipedia using an LLM.", |
| "These fictional sources cannot establish a real subject's notability."]} |
| packets.append({"id": key, "split": split, "topic": topic, "sources": rows, |
| "article_target": article, "outline_target": outline}) |
| payload = {"license": "GPL-3.0-only", "authorship": "Project-authored synthetic behavior examples", |
| "limitations": "Fictional topics and reserved URLs. Not real references, model outputs, or accuracy evidence.", |
| "packets": packets} |
| return payload |
|
|
|
|
| if __name__ == "__main__": |
| path = Path(__file__).with_name("article_examples.json") |
| path.write_text(json.dumps(build(), ensure_ascii=False, indent=2), encoding="utf-8") |
| print(f"Wrote {path}") |
|
|
