Admin step A: privacy-respecting first-party event logging

- events table (kind, article_id, visitor_hash, day) with a UNIQUE key that dedups
  to one row per visitor-day — caps volume and makes counts mean distinct
  visitor-days. NO ip/ua/referrer/url. Groupings derived from article_id at query
  time, never stored.
- POST /api/events (public): whitelisted kinds (visit/open/share_ub/copy_source/
  native_share/source_click); visitor token hashed server-side (never raw).
- Frontend analytics.js: random localStorage visitor token; track() via sendBeacon;
  visit once/day; open on article click; share_ub/copy_source/native_share from the
  share menu; /a landing pages fire source_click. 127 tests pass.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
jay
2026-06-03 18:21:49 +00:00
parent ab5caada0b
commit 1a778e1334
7 changed files with 180 additions and 4 deletions
+31
View File
@@ -336,6 +336,23 @@ class PrefsBody(BaseModel):
prefs: dict = {}
class EventBody(BaseModel):
kind: str
article_id: int | None = None
visitor: str | None = None
# The only event kinds we record (per the agreed vocabulary).
_EVENT_KINDS = {"visit", "open", "share_ub", "copy_source", "native_share", "source_click"}
def _visitor_hash(token: str | None) -> str:
token = (token or "").strip()[:200]
if not token:
return ""
return hashlib.sha256(f"{SESSION_SECRET}:{token}".encode()).hexdigest()
# --- App --------------------------------------------------------------------
@@ -666,6 +683,20 @@ def create_app() -> FastAPI:
_kick_summary(aid, background_tasks) # generate for next time; page polls
return HTMLResponse(share.render_share_page(dict(row), PUBLIC_BASE_URL, summary=summary))
# --- Privacy-respecting first-party analytics -------------------------
@app.post("/api/events")
def record_event(body: EventBody) -> dict:
if body.kind in _EVENT_KINDS:
with get_conn() as conn:
conn.execute(
"INSERT OR IGNORE INTO events (kind, article_id, visitor_hash, day) "
"VALUES (?, ?, ?, date('now'))",
(body.kind, body.article_id or 0, _visitor_hash(body.visitor)),
)
conn.commit()
return {"ok": True} # always identical; dedup'd by the unique key
@app.get("/api/summary/{article_id}")
def article_summary(article_id: int, background_tasks: BackgroundTasks) -> dict:
with get_conn() as conn:
+18
View File
@@ -205,6 +205,24 @@ CREATE TABLE IF NOT EXISTS article_summaries (
model TEXT,
created_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP
);
-- Privacy-respecting, first-party analytics. NO IP / user-agent / referrer / raw
-- URL. visitor_hash is a hash of a random localStorage token (never email/IP).
-- The UNIQUE key dedups to one row per (kind, article, visitor, day) — that both
-- caps volume and makes counts mean "distinct visitor-days". Groupings are derived
-- from article_id at query time, never stored here.
CREATE TABLE IF NOT EXISTS events (
id INTEGER PRIMARY KEY AUTOINCREMENT,
kind TEXT NOT NULL,
article_id INTEGER NOT NULL DEFAULT 0,
visitor_hash TEXT NOT NULL DEFAULT '',
day TEXT NOT NULL,
created_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP,
UNIQUE (kind, article_id, visitor_hash, day)
);
CREATE INDEX IF NOT EXISTS idx_events_day ON events(day);
CREATE INDEX IF NOT EXISTS idx_events_kind ON events(kind);
CREATE INDEX IF NOT EXISTS idx_events_article ON events(article_id);
"""
+16
View File
@@ -72,6 +72,21 @@ def render_share_page(article: dict, base_url: str, summary: str | None = None)
}}
go();
}})();
</script>"""
# Fire a privacy-respecting 'source_click' when the reader heads to the source
# (reuses the SPA's anonymous visitor token from same-origin localStorage).
src_click = f"""<script>
(function(){{
var b=document.querySelector('[data-src-click]'); if(!b) return;
b.addEventListener('click',function(){{
try{{
var v=localStorage.getItem('goodnews:visitor')||'';
var body=JSON.stringify({{kind:'source_click',article_id:{aid},visitor:v}});
if(navigator.sendBeacon) navigator.sendBeacon('/api/events',new Blob([body],{{type:'application/json'}}));
}}catch(e){{}}
}});
}})();
</script>"""
return f"""<!doctype html>
@@ -134,6 +149,7 @@ def render_share_page(article: dict, base_url: str, summary: str | None = None)
</div>
</article>
</main>
{src_click}
{poll}
</body>
</html>"""