refactor: share official records parser

apps/api/Dockerfile

@@ -1,8 +1,9 @@
 FROM python:3.12-slim
 WORKDIR /app
 ENV PYTHONDONTWRITEBYTECODE=1 PYTHONUNBUFFERED=1
-COPY requirements.txt .
+COPY apps/api/requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
-COPY . .
+COPY apps/api .
+COPY rf4_research ./rf4_research
 EXPOSE 8000
 CMD ["sh", "-c", "alembic upgrade head && python -m app.seed && uvicorn app.main:app --host 0.0.0.0 --port 8000"]

apps/api/app/importer.py

@@ -7,9 +7,9 @@ from dataclasses import asdict, dataclass
 from datetime import date, datetime, time, timezone
 
 import httpx
-from bs4 import BeautifulSoup, Tag
 from sqlalchemy import select
 from sqlalchemy.orm import Session
+from rf4_research.official_parser import RecordsContractError, parse_official_records
 
 from .models import (
     Bait, BaitKind, CatchReport, Fish, ImportStatus, ModerationStatus,
@@ -60,54 +60,16 @@ def external_id(record: RawRecord) -> str:
     return hashlib.sha256("|".join(normalize(part) for part in parts).encode()).hexdigest()
 
 
-def parse_weight(raw: str) -> int:
-    value = " ".join(raw.replace("\xa0", " ").split()).lower()
-    match = re.fullmatch(r"([\d .,'’]+)\s*(kg|g)", value)
-    if not match:
-        raise ImportSourceError(f"unsupported weight {raw!r}")
-    number, unit = match.groups()
-    number = number.replace(" ", "").replace("'", "").replace("’", "")
-    if unit == "g":
-        return int(number.replace(".", "").replace(",", ""))
-    if "," in number and "." not in number:
-        number = number.replace(",", ".")
-    return round(float(number) * 1000)
-
-
-def _text(node: Tag | None) -> str:
-    return node.get_text(" ", strip=True) if node else ""
-
-
 def parse_html(html: str, *, region: str, category: str) -> list[RawRecord]:
-    soup = BeautifulSoup(html, "html.parser")
-    table = soup.select_one("div.records.flex_table")
-    if table is None:
-        raise ImportSourceError("records table not found")
-    header = table.select_one(":scope > .row.header")
-    expected = ["fish", "weight", "location", "bait", "gamername", "data"]
-    cells = header.find_all("div", recursive=False) if header else []
-    actual = [next((key for key in expected if key in cell.get("class", [])), "") for cell in cells]
-    if actual != expected:
-        raise ImportSourceError(f"record columns changed: {actual}")
-
-    records: list[RawRecord] = []
-    for group in table.select(":scope > .rows > .row > .records_subtable"):
-        group_header = group.select_one(":scope > .row.header")
-        if group_header is None:
-            continue
-        fish = _text(group_header.select_one(".fish .text"))
-        rows = [group_header, *group.select(":scope > .rows > .row")]
-        for row in rows:
-            bait_node = row.select_one(".bait_icon")
-            bait = bait_node.get("title", "").strip() if bait_node else ""
-            try:
-                record_date = datetime.strptime(_text(row.select_one(".data")), "%d.%m.%y").date()
-            except ValueError as exc:
-                raise ImportSourceError("record date format changed") from exc
-            records.append(RawRecord(region.upper(), category, _text(row.select_one(".gamername")) or None, fish, parse_weight(_text(row.select_one(".weight"))), _text(row.select_one(".location")), bait or None, record_date))
-    if not records:
-        raise ImportSourceError("records table is empty")
-    return records
+    try:
+        rows = parse_official_records(html)
+    except RecordsContractError as exc:
+        raise ImportSourceError(str(exc)) from exc
+    return [RawRecord(
+        region=region.upper(), category=category, player=row.player, fish=row.fish,
+        weight_g=row.weight_g, waterbody=row.waterbody, bait=row.bait,
+        record_date=row.record_date,
+    ) for row in rows]
 
 
 def fetch_records(