oliver/StaySense
1
0
Fork 0
Code Issues 11 Pull requests Projects Releases Packages Wiki Activity Actions

Add roadmap and implement top UX improvements for score transparency and map selection

Browse source
This commit is contained in:
Oliver 2026-02-16 08:47:41 +01:00
parent a118c3ca33
commit c0ea660e48
No known key found for this signature in database
6 changed files with 324 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

126
backend/server.py
View file

@ -476,6 +476,98 @@ def source_health(sources: list[dict], now: dt.datetime) -> dict:
}
def classify_quality(
health: dict, used_fallback_pois: bool, degraded: bool, factor_count: int, community_factor_count: int
) -> dict:
score = 100
reasons = []
if degraded:
score -= 45
reasons.append("Backend im degradierten Modus")
if used_fallback_pois:
score -= 20
reasons.append("Fallback-POIs statt Live-POIs")
if not health.get("has_data"):
score -= 20
reasons.append("Keine Quellenmetadaten")
else:
stalest_age = health.get("stalest_age_hours")
if isinstance(stalest_age, (int, float)):
if stalest_age > 48:
score -= 20
reasons.append("Mindestens eine Quelle >48h alt")
elif stalest_age > 24:
score -= 10
reasons.append("Mindestens eine Quelle >24h alt")
stale_count = len(health.get("stale_sources") or [])
if stale_count >= 2:
score -= 10
reasons.append("Mehrere stale Quellen")
elif stale_count == 1:
score -= 5
reasons.append("Eine stale Quelle")
if factor_count <= 2:
score -= 8
reasons.append("Wenig Score-Faktoren verfuegbar")
if community_factor_count == 0:
score -= 5
reasons.append("Keine aktuellen Community-Signale")
elif community_factor_count >= 2:
score += 4
reasons.append("Mehrere aktuelle Community-Signale")
score = max(0, min(100, int(round(score))))
if score >= 75:
level = "high"
label = "hoch"
elif score >= 45:
level = "medium"
label = "mittel"
else:
level = "low"
label = "niedrig"
return {
"score": score,
"level": level,
"label": label,
"reasons": reasons[:4],
}
def build_explanation(factors: list[dict], spot: dict) -> dict:
details = []
for item in factors:
points = float(item["points"])
impact = "neutral"
if points > 0:
impact = "positive"
elif points < 0:
impact = "negative"
details.append(
{
"key": item["key"],
"label": item["label"],
"points": round(points, 2),
"source": item["source"],
"impact": impact,
}
)
return {
"factors": details,
"spot_context": {
"area_type": spot.get("osm_area_type", "unknown"),
"road_type": spot.get("road_type", "unknown"),
"distance_police_m": int(spot.get("distance_police_m", 5000)),
"distance_fire_m": int(spot.get("distance_fire_m", 5000)),
"distance_hospital_m": int(spot.get("distance_hospital_m", 5000)),
},
}
def compute_score_payload(lat: float, lon: float, at_time: dt.datetime) -> dict:
now_iso = to_iso(utc_now())
spot = ensure_spot(lat, lon, now_iso)
@ -504,8 +596,16 @@ def compute_score_payload(lat: float, lon: float, at_time: dt.datetime) -> dict:
raw_score = 100.0 + sum(item["points"] for item in factors)
score = clamp_score(raw_score)
top_reasons = sorted(factors, key=lambda item: abs(item["points"]), reverse=True)[:4]
sorted_factors = sorted(factors, key=lambda item: abs(item["points"]), reverse=True)
top_reasons = sorted_factors[:4]
reasons = [f"{item['label']} ({item['points']:+.0f})" for item in top_reasons]
quality = classify_quality(
health=health,
used_fallback_pois=bool(spot.get("used_fallback_pois", False)),
degraded=False,
factor_count=len(sorted_factors),
community_factor_count=sum(1 for item in factors if item.get("source") == "community"),
)
return {
"spot_id": spot["id"],
@ -513,6 +613,7 @@ def compute_score_payload(lat: float, lon: float, at_time: dt.datetime) -> dict:
"ampel": ampel(score),
"reasons": reasons,
"factors": top_reasons,
"explanation": build_explanation(sorted_factors, spot),
"night_window": {
"start": to_iso(night_start),
"end": to_iso(night_end),
@ -524,6 +625,7 @@ def compute_score_payload(lat: float, lon: float, at_time: dt.datetime) -> dict:
"sources": sources,
"health": health,
"used_fallback_pois": bool(spot.get("used_fallback_pois", False)),
"quality": quality,
},
}
@ -550,8 +652,24 @@ def compute_score_payload_fallback(lat: float, lon: float, at_time: dt.datetime,
raw_score = 100.0 + sum(item["points"] for item in factors)
score = clamp_score(raw_score)
top_reasons = sorted(factors, key=lambda item: abs(item["points"]), reverse=True)[:4]
sorted_factors = sorted(factors, key=lambda item: abs(item["points"]), reverse=True)
top_reasons = sorted_factors[:4]
reasons = [f"{item['label']} ({item['points']:+.0f})" for item in top_reasons] or ["Fallback-Berechnung aktiv (0)"]
fallback_spot = {
"osm_area_type": "residential",
"road_type": "unknown",
"distance_police_m": police_d,
"distance_fire_m": fire_d,
"distance_hospital_m": hosp_d,
}
health = {"freshest_age_hours": None, "stalest_age_hours": None, "stale_sources": [], "has_data": False}
quality = classify_quality(
health=health,
used_fallback_pois=True,
degraded=True,
factor_count=len(sorted_factors),
community_factor_count=0,
)
return {
"spot_id": spot_id_for(lat, lon),
@ -559,6 +677,7 @@ def compute_score_payload_fallback(lat: float, lon: float, at_time: dt.datetime,
"ampel": ampel(score),
"reasons": reasons,
"factors": top_reasons,
"explanation": build_explanation(sorted_factors, fallback_spot),
"night_window": {
"start": to_iso(night_start),
"end": to_iso(night_end),
@ -568,10 +687,11 @@ def compute_score_payload_fallback(lat: float, lon: float, at_time: dt.datetime,
"region": "DE-NW (Pilot: Kreis Mettmann)",
"attribution": "Kartendaten: OpenStreetMap-Mitwirkende (ODbL)",
"sources": [],
"health": {"freshest_age_hours": None, "stalest_age_hours": None, "stale_sources": [], "has_data": False},
"health": health,
"used_fallback_pois": True,
"degraded": True,
"degraded_reason": error_code,
"quality": quality,
},
}