election-game/backend/data/pipeline/src/election_data/people/output.py

117 lines
3.7 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

"""PersonDraft → people.json 的 Person(packages/protocol/src/people.ts)。"""
from __future__ import annotations
from collections import defaultdict
from collections.abc import Sequence
from typing import Any
from .model import PersonDraft, Sourced
from .names import name_key, slug_name
STAT_KEYS: tuple[str, ...] = (
"awareness",
"network",
"affinity",
"eloquence",
"strategy",
"online",
"image",
"wealth",
"stamina",
)
def slug_id(p: PersonDraft) -> str:
year = p.birth_date.value[:4] if p.birth_date and p.birth_date.value else "x"
return f"tw-{p.category}-{slug_name(p.name)}-{year}"
def _sort_key(p: PersonDraft) -> tuple[str, ...]:
"""同 id 的人物排序(決定 -2、-3 的歸屬),只用資料內容,不受輸入順序影響。"""
return (
name_key(p.name),
p.birth_date.value or "" if p.birth_date else "",
p.name,
",".join(sorted(p.roles)),
",".join(sorted(p.counties)),
",".join(sorted(p.districts)),
",".join(f"{x.org}/{x.title}" for x in p.positions),
)
def assign_ids(persons: Sequence[PersonDraft]) -> list[str]:
"""有 Wikidata QID 就用 QID;否則用 tw-<類別>-<姓名>-<出生年或 x>。同 id 依固定排序加上 -2、-3。"""
base = [p.wikidata_id.value if p.wikidata_id and p.wikidata_id.value else slug_id(p) for p in persons]
groups: dict[str, list[int]] = defaultdict(list)
for i, b in enumerate(base):
groups[b].append(i)
ids = list(base)
for b, idx in groups.items():
if len(idx) > 1:
for n, i in enumerate(sorted(idx, key=lambda k: _sort_key(persons[k]))):
ids[i] = b if n == 0 else f"{b}-{n + 1}"
if len(set(ids)) != len(ids):
raise ValueError("人物 id 不唯一")
return ids
def _field_sources(p: PersonDraft) -> dict[str, list[str]]:
out: dict[str, list[str]] = {"name": list(p.name_sources)}
scalars: dict[str, Sourced | None] = {
"gender": p.gender,
"birthDate": p.birth_date,
"party": p.party,
"wikidataId": p.wikidata_id,
}
for field, value in scalars.items():
if value is not None and (value.value is not None or field == "party"):
out[field] = list(value.source_ids)
return out
def _value(v: Sourced | None) -> str | None:
return v.value if v else None
def person_json(p: PersonDraft, person_id: str, as_of: str) -> dict[str, Any]:
return {
"id": person_id,
"kind": "real",
"category": p.category,
"name": p.name,
"gender": _value(p.gender),
"birthDate": _value(p.birth_date),
"party": _value(p.party),
"wikidataId": _value(p.wikidata_id),
"positions": [x.to_json() for x in p.positions],
"education": [x.to_json() for x in p.education],
"electionHistory": [x.to_json() for x in p.elections],
"stats": dict.fromkeys(STAT_KEYS),
"scoring": [],
"fieldSources": _field_sources(p),
"asOf": as_of,
}
def people_file(
persons: Sequence[PersonDraft], ids: Sequence[str], protocol_version: str, as_of: str
) -> dict[str, Any]:
rows = sorted(zip(ids, persons, strict=True), key=lambda t: (t[1].category, name_key(t[1].name), t[0]))
return {
"protocolVersion": protocol_version,
"asOf": as_of,
"people": [person_json(p, pid, as_of) for pid, p in rows],
}
def referenced_source_ids(doc: dict[str, Any]) -> set[str]:
out: set[str] = set()
for person in doc["people"]:
for ids in person["fieldSources"].values():
out.update(ids)
for key in ("positions", "education", "electionHistory"):
for item in person[key]:
out.update(item["sourceIds"])
return out