068073423f
Local-first RSS/Atom ingestion pipeline with metadata-only storage, heuristic + local-LLM scoring, and daily brief builder. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
168 lines
5.2 KiB
Python
168 lines
5.2 KiB
Python
from __future__ import annotations
|
|
|
|
import sqlite3
|
|
from datetime import date
|
|
|
|
|
|
def build_daily_brief(
|
|
conn: sqlite3.Connection,
|
|
brief_date: str | None = None,
|
|
limit: int = 5,
|
|
replace: bool = False,
|
|
) -> int:
|
|
target_date = brief_date or date.today().isoformat()
|
|
existing = conn.execute("SELECT id FROM daily_briefs WHERE brief_date = ?", (target_date,)).fetchone()
|
|
if existing and not replace:
|
|
return int(existing["id"])
|
|
if existing and replace:
|
|
conn.execute("DELETE FROM daily_briefs WHERE id = ?", (existing["id"],))
|
|
|
|
brief_id = conn.execute(
|
|
"INSERT INTO daily_briefs (brief_date, title) VALUES (?, ?)",
|
|
(target_date, f"Five Good Things Today - {target_date}"),
|
|
).lastrowid
|
|
|
|
rows = _candidate_articles(conn, target_date)
|
|
selected = _select_diverse(rows, limit)
|
|
for index, row in enumerate(selected, start=1):
|
|
conn.execute(
|
|
"""
|
|
INSERT INTO daily_brief_items (brief_id, article_id, rank, selection_reason)
|
|
VALUES (?, ?, ?, ?)
|
|
""",
|
|
(
|
|
brief_id,
|
|
row["id"],
|
|
index,
|
|
_selection_reason(row),
|
|
),
|
|
)
|
|
conn.commit()
|
|
return int(brief_id)
|
|
|
|
|
|
def show_brief(conn: sqlite3.Connection, brief_date: str | None = None, limit: int = 10) -> list[sqlite3.Row]:
|
|
target_date = brief_date or _latest_brief_date(conn)
|
|
if not target_date:
|
|
return []
|
|
return conn.execute(
|
|
"""
|
|
SELECT
|
|
b.brief_date,
|
|
bi.rank,
|
|
bi.selection_reason,
|
|
a.title,
|
|
a.description,
|
|
a.canonical_url,
|
|
a.published_at,
|
|
src.name AS source_name,
|
|
src.default_category,
|
|
s.constructive_score,
|
|
s.cortisol_score,
|
|
s.ragebait_score,
|
|
s.agency_score,
|
|
s.human_benefit_score,
|
|
s.reason_code,
|
|
s.reason_text,
|
|
s.model_name
|
|
FROM daily_briefs b
|
|
JOIN daily_brief_items bi ON bi.brief_id = b.id
|
|
JOIN articles a ON a.id = bi.article_id
|
|
JOIN sources src ON src.id = a.source_id
|
|
LEFT JOIN article_scores s ON s.article_id = a.id
|
|
WHERE b.brief_date = ?
|
|
ORDER BY bi.rank
|
|
LIMIT ?
|
|
""",
|
|
(target_date, limit),
|
|
).fetchall()
|
|
|
|
|
|
def _candidate_articles(conn: sqlite3.Connection, target_date: str) -> list[sqlite3.Row]:
|
|
return conn.execute(
|
|
"""
|
|
SELECT
|
|
a.id,
|
|
a.title,
|
|
a.description,
|
|
a.canonical_url,
|
|
a.published_at,
|
|
a.discovered_at,
|
|
src.name AS source_name,
|
|
src.default_category,
|
|
src.trust_score,
|
|
s.constructive_score,
|
|
s.cortisol_score,
|
|
s.ragebait_score,
|
|
s.agency_score,
|
|
s.human_benefit_score,
|
|
s.novelty_score,
|
|
s.pr_risk_score,
|
|
s.reason_code,
|
|
s.reason_text,
|
|
s.model_name
|
|
FROM articles a
|
|
JOIN sources src ON src.id = a.source_id
|
|
JOIN article_scores s ON s.article_id = a.id
|
|
WHERE s.accepted = 1
|
|
AND date(COALESCE(a.published_at, a.discovered_at)) = date(?)
|
|
ORDER BY
|
|
(s.constructive_score + s.agency_score + s.human_benefit_score + src.trust_score
|
|
- s.cortisol_score - s.ragebait_score - s.pr_risk_score) DESC,
|
|
COALESCE(a.published_at, a.discovered_at) DESC
|
|
LIMIT 50
|
|
""",
|
|
(target_date,),
|
|
).fetchall()
|
|
|
|
|
|
def _select_diverse(rows: list[sqlite3.Row], limit: int) -> list[sqlite3.Row]:
|
|
selected = []
|
|
seen_sources = set()
|
|
seen_categories = set()
|
|
|
|
for row in rows:
|
|
if len(selected) >= limit:
|
|
break
|
|
source = row["source_name"]
|
|
category = row["default_category"]
|
|
if source in seen_sources and len(rows) > limit:
|
|
continue
|
|
selected.append(row)
|
|
seen_sources.add(source)
|
|
seen_categories.add(category)
|
|
|
|
if len(selected) < limit:
|
|
selected_ids = {row["id"] for row in selected}
|
|
for row in rows:
|
|
if len(selected) >= limit:
|
|
break
|
|
if row["id"] in selected_ids:
|
|
continue
|
|
selected.append(row)
|
|
selected_ids.add(row["id"])
|
|
|
|
if len(seen_categories) < 2 and len(rows) > limit:
|
|
selected_ids = {row["id"] for row in selected}
|
|
for row in rows:
|
|
if row["id"] in selected_ids:
|
|
continue
|
|
if row["default_category"] not in seen_categories:
|
|
selected[-1] = row
|
|
break
|
|
|
|
return selected
|
|
|
|
|
|
def _selection_reason(row: sqlite3.Row) -> str:
|
|
return (
|
|
f"{row['reason_code']}; constructive={row['constructive_score']}, "
|
|
f"agency={row['agency_score']}, human_benefit={row['human_benefit_score']}, "
|
|
f"cortisol={row['cortisol_score']}, source={row['source_name']}"
|
|
)
|
|
|
|
|
|
def _latest_brief_date(conn: sqlite3.Connection) -> str | None:
|
|
row = conn.execute("SELECT brief_date FROM daily_briefs ORDER BY brief_date DESC LIMIT 1").fetchone()
|
|
return row["brief_date"] if row else None
|