mitai-jinkendo/backend/routers/activity.py

"""
Activity Tracking Endpoints for Mitai Jinkendo

Handles workout/activity logging, statistics, and Apple Health CSV import.
"""
import csv
import io
import uuid
import logging
from typing import Optional

from fastapi import APIRouter, HTTPException, UploadFile, File, Header, Depends, Query

from db import get_db, get_cursor, r2d
from auth import require_auth, check_feature_access, increment_feature_usage
from models import ActivityEntry
from routers.profiles import get_pid
from feature_logger import log_feature_usage
from quality_filter import get_quality_filter_sql

# Evaluation import with error handling (Phase 1.2)
try:
    from evaluation_helper import evaluate_and_save_activity
    EVALUATION_AVAILABLE = True
except Exception as e:
    logger.warning(f"[AUTO-EVAL] Evaluation system not available: {e}")
    EVALUATION_AVAILABLE = False
    evaluate_and_save_activity = None

router = APIRouter(prefix="/api/activity", tags=["activity"])
logger = logging.getLogger(__name__)


@router.get("")
def list_activity(
    limit: int = Query(200, ge=1, le=50_000),
    days: Optional[int] = Query(None, ge=1, le=4000, description="Nur Einträge mit date >= HEUTE − days (Kalendertage)"),
    x_profile_id: Optional[str] = Header(default=None),
    session: dict = Depends(require_auth),
):
    """Get activity entries for current profile. Optional *days* filter by calendar window (not the same as *limit*)."""
    pid = get_pid(x_profile_id)
    with get_db() as conn:
        cur = get_cursor(conn)

        # Issue #31: Apply global quality filter (profile from DB = saved level)
        cur.execute("SELECT * FROM profiles WHERE id=%s", (pid,))
        profile = r2d(cur.fetchone())
        quality_filter = get_quality_filter_sql(profile)

        if days is not None:
            cur.execute(
                f"""
                SELECT * FROM activity_log
                WHERE profile_id=%s
                  {quality_filter}
                  AND date >= (CURRENT_DATE - %s::integer)
                ORDER BY date DESC, start_time DESC
                LIMIT %s
                """,
                (pid, days, limit),
            )
        else:
            cur.execute(
                f"""
                SELECT * FROM activity_log
                WHERE profile_id=%s
                  {quality_filter}
                ORDER BY date DESC, start_time DESC
                LIMIT %s
                """,
                (pid, limit),
            )
        return [r2d(r) for r in cur.fetchall()]


@router.post("")
def create_activity(e: ActivityEntry, x_profile_id: Optional[str]=Header(default=None), session: dict=Depends(require_auth)):
    """Create new activity entry."""
    pid = get_pid(x_profile_id)

    # Phase 4: Check feature access and ENFORCE
    access = check_feature_access(pid, 'activity_entries')
    log_feature_usage(pid, 'activity_entries', access, 'create')

    if not access['allowed']:
        logger.warning(
            f"[FEATURE-LIMIT] User {pid} blocked: "
            f"activity_entries {access['reason']} (used: {access['used']}, limit: {access['limit']})"
        )
        raise HTTPException(
            status_code=403,
            detail=f"Limit erreicht: Du hast das Kontingent für Aktivitätseinträge überschritten ({access['used']}/{access['limit']}). "
                   f"Bitte kontaktiere den Admin oder warte bis zum nächsten Reset."
        )

    eid = str(uuid.uuid4())
    d = e.model_dump()
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute("""INSERT INTO activity_log
            (id,profile_id,date,start_time,end_time,activity_type,duration_min,kcal_active,kcal_resting,
             hr_avg,hr_max,distance_km,rpe,source,notes,created)
            VALUES (%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,CURRENT_TIMESTAMP)""",
            (eid,pid,d['date'],d['start_time'],d['end_time'],d['activity_type'],d['duration_min'],
             d['kcal_active'],d['kcal_resting'],d['hr_avg'],d['hr_max'],d['distance_km'],
             d['rpe'],d['source'],d['notes']))

        # Phase 1.2: Auto-evaluation after INSERT
        if EVALUATION_AVAILABLE:
            # Load the activity data to evaluate
            cur.execute("""
                SELECT id, profile_id, date, training_type_id, duration_min,
                       hr_avg, hr_max, distance_km, kcal_active, kcal_resting,
                       rpe, pace_min_per_km, cadence, elevation_gain
                FROM activity_log
                WHERE id = %s
            """, (eid,))
            activity_row = cur.fetchone()
            if activity_row:
                activity_dict = dict(activity_row)
                training_type_id = activity_dict.get("training_type_id")
                if training_type_id:
                    try:
                        evaluate_and_save_activity(cur, eid, activity_dict, training_type_id, pid)
                        logger.info(f"[AUTO-EVAL] Evaluated activity {eid} on INSERT")
                    except Exception as eval_error:
                        logger.error(f"[AUTO-EVAL] Failed to evaluate activity {eid}: {eval_error}")

    # Phase 2: Increment usage counter (always for new entries)
    increment_feature_usage(pid, 'activity_entries')

    return {"id":eid,"date":e.date}


@router.put("/{eid}")
def update_activity(eid: str, e: ActivityEntry, x_profile_id: Optional[str]=Header(default=None), session: dict=Depends(require_auth)):
    """Update existing activity entry."""
    pid = get_pid(x_profile_id)
    with get_db() as conn:
        d = e.model_dump()
        cur = get_cursor(conn)
        cur.execute(f"UPDATE activity_log SET {', '.join(f'{k}=%s' for k in d)} WHERE id=%s AND profile_id=%s",
                     list(d.values())+[eid,pid])

        # Phase 1.2: Auto-evaluation after UPDATE
        if EVALUATION_AVAILABLE:
            # Load the updated activity data to evaluate
            cur.execute("""
                SELECT id, profile_id, date, training_type_id, duration_min,
                       hr_avg, hr_max, distance_km, kcal_active, kcal_resting,
                       rpe, pace_min_per_km, cadence, elevation_gain
                FROM activity_log
                WHERE id = %s
            """, (eid,))
            activity_row = cur.fetchone()
            if activity_row:
                activity_dict = dict(activity_row)
                training_type_id = activity_dict.get("training_type_id")
                if training_type_id:
                    try:
                        evaluate_and_save_activity(cur, eid, activity_dict, training_type_id, pid)
                        logger.info(f"[AUTO-EVAL] Re-evaluated activity {eid} on UPDATE")
                    except Exception as eval_error:
                        logger.error(f"[AUTO-EVAL] Failed to re-evaluate activity {eid}: {eval_error}")

    return {"id":eid}


@router.delete("/{eid}")
def delete_activity(eid: str, x_profile_id: Optional[str]=Header(default=None), session: dict=Depends(require_auth)):
    """Delete activity entry."""
    pid = get_pid(x_profile_id)
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute("DELETE FROM activity_log WHERE id=%s AND profile_id=%s", (eid,pid))
    return {"ok":True}


@router.get("/stats")
def activity_stats(x_profile_id: Optional[str]=Header(default=None), session: dict=Depends(require_auth)):
    """Get activity statistics (last 30 entries)."""
    pid = get_pid(x_profile_id)
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute(
            "SELECT * FROM activity_log WHERE profile_id=%s ORDER BY date DESC LIMIT 30", (pid,))
        rows = [r2d(r) for r in cur.fetchall()]
    if not rows: return {"count":0,"total_kcal":0,"total_min":0,"by_type":{}}
    total_kcal=sum(float(r.get('kcal_active') or 0) for r in rows)
    total_min=sum(float(r.get('duration_min') or 0) for r in rows)
    by_type={}
    for r in rows:
        t=r['activity_type']; by_type.setdefault(t,{'count':0,'kcal':0,'min':0})
        by_type[t]['count']+=1
        by_type[t]['kcal']+=float(r.get('kcal_active') or 0)
        by_type[t]['min']+=float(r.get('duration_min') or 0)
    return {"count":len(rows),"total_kcal":round(total_kcal),"total_min":round(total_min),"by_type":by_type}


def get_training_type_for_activity(activity_type: str, profile_id: str = None):
    """
    Map activity_type to training_type_id using database mappings.

    Priority:
    1. User-specific mapping (profile_id)
    2. Global mapping (profile_id = NULL)
    3. No mapping found → returns (None, None, None)

    Returns: (training_type_id, category, subcategory) or (None, None, None)
    """
    with get_db() as conn:
        cur = get_cursor(conn)

        # Try user-specific mapping first
        if profile_id:
            cur.execute("""
                SELECT m.training_type_id, t.category, t.subcategory
                FROM activity_type_mappings m
                JOIN training_types t ON m.training_type_id = t.id
                WHERE m.activity_type = %s AND m.profile_id = %s
                LIMIT 1
            """, (activity_type, profile_id))
            row = cur.fetchone()
            if row:
                return (row['training_type_id'], row['category'], row['subcategory'])

        # Try global mapping
        cur.execute("""
            SELECT m.training_type_id, t.category, t.subcategory
            FROM activity_type_mappings m
            JOIN training_types t ON m.training_type_id = t.id
            WHERE m.activity_type = %s AND m.profile_id IS NULL
            LIMIT 1
        """, (activity_type,))
        row = cur.fetchone()
        if row:
            return (row['training_type_id'], row['category'], row['subcategory'])

    return (None, None, None)


@router.get("/uncategorized")
def list_uncategorized_activities(x_profile_id: Optional[str]=Header(default=None), session: dict=Depends(require_auth)):
    """Get activities without assigned training type, grouped by activity_type."""
    pid = get_pid(x_profile_id)
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute("""
            SELECT activity_type, COUNT(*) as count,
                   MIN(date) as first_date, MAX(date) as last_date
            FROM activity_log
            WHERE profile_id=%s AND training_type_id IS NULL
            GROUP BY activity_type
            ORDER BY count DESC
        """, (pid,))
        return [r2d(r) for r in cur.fetchall()]


@router.post("/bulk-categorize")
def bulk_categorize_activities(
    data: dict,
    x_profile_id: Optional[str]=Header(default=None),
    session: dict=Depends(require_auth)
):
    """
    Bulk update training type for activities.

    Also saves the mapping to activity_type_mappings for future imports.

    Body: {
        "activity_type": "Running",
        "training_type_id": 1,
        "training_category": "cardio",
        "training_subcategory": "running"
    }
    """
    pid = get_pid(x_profile_id)
    activity_type = data.get('activity_type')
    training_type_id = data.get('training_type_id')
    training_category = data.get('training_category')
    training_subcategory = data.get('training_subcategory')

    if not activity_type or not training_type_id:
        raise HTTPException(400, "activity_type and training_type_id required")

    with get_db() as conn:
        cur = get_cursor(conn)

        # Update existing activities
        cur.execute("""
            UPDATE activity_log
            SET training_type_id = %s,
                training_category = %s,
                training_subcategory = %s
            WHERE profile_id = %s
              AND activity_type = %s
              AND training_type_id IS NULL
        """, (training_type_id, training_category, training_subcategory, pid, activity_type))
        updated_count = cur.rowcount

        # Phase 1.2: Auto-evaluation after bulk categorization
        if EVALUATION_AVAILABLE:
            # Load all activities that were just updated and evaluate them
            cur.execute("""
                SELECT id, profile_id, date, training_type_id, duration_min,
                       hr_avg, hr_max, distance_km, kcal_active, kcal_resting,
                       rpe, pace_min_per_km, cadence, elevation_gain
                FROM activity_log
                WHERE profile_id = %s
                  AND activity_type = %s
                  AND training_type_id = %s
            """, (pid, activity_type, training_type_id))

            activities_to_evaluate = cur.fetchall()
            evaluated_count = 0
            for activity_row in activities_to_evaluate:
                activity_dict = dict(activity_row)
                try:
                    evaluate_and_save_activity(cur, activity_dict["id"], activity_dict, training_type_id, pid)
                    evaluated_count += 1
                except Exception as eval_error:
                    logger.warning(f"[AUTO-EVAL] Failed to evaluate bulk-categorized activity {activity_dict['id']}: {eval_error}")

            logger.info(f"[AUTO-EVAL] Evaluated {evaluated_count}/{updated_count} bulk-categorized activities")

        # Save mapping for future imports (upsert)
        cur.execute("""
            INSERT INTO activity_type_mappings (activity_type, training_type_id, profile_id, source, updated_at)
            VALUES (%s, %s, %s, 'bulk', CURRENT_TIMESTAMP)
            ON CONFLICT (activity_type, profile_id)
            DO UPDATE SET
                training_type_id = EXCLUDED.training_type_id,
                source = 'bulk',
                updated_at = CURRENT_TIMESTAMP
        """, (activity_type, training_type_id, pid))

        logger.info(f"[MAPPING] Saved bulk mapping: {activity_type} → training_type_id {training_type_id} (profile {pid})")

    return {"updated": updated_count, "activity_type": activity_type, "mapping_saved": True}


@router.post("/import-csv")
async def import_activity_csv(file: UploadFile=File(...), x_profile_id: Optional[str]=Header(default=None), session: dict=Depends(require_auth)):
    """Import Apple Health workout CSV with automatic training type mapping."""
    pid = get_pid(x_profile_id)
    raw = await file.read()
    try:    text = raw.decode('utf-8')
    except: text = raw.decode('latin-1')
    if text.startswith('\ufeff'): text = text[1:]
    if not text.strip(): raise HTTPException(400,"Leere Datei")
    reader = csv.DictReader(io.StringIO(text))
    inserted = skipped = 0
    with get_db() as conn:
        cur = get_cursor(conn)
        for row in reader:
            wtype = row.get('Workout Type','').strip()
            start = row.get('Start','').strip()
            if not wtype or not start: continue
            try: date = start[:10]
            except: continue
            dur = row.get('Duration','').strip()
            duration_min = None
            if dur:
                try:
                    p = dur.split(':')
                    duration_min = round(int(p[0])*60+int(p[1])+int(p[2])/60,1)
                except: pass
            def kj(v):
                try: return round(float(v)/4.184) if v else None
                except: return None
            def tf(v):
                try: return round(float(v),1) if v else None
                except: return None
            # Map activity_type to training_type_id using database mappings
            training_type_id, training_category, training_subcategory = get_training_type_for_activity(wtype, pid)

            try:
                # Check if entry already exists (duplicate detection by date + start_time)
                cur.execute("""
                    SELECT id FROM activity_log
                    WHERE profile_id = %s AND date = %s AND start_time = %s
                """, (pid, date, start))
                existing = cur.fetchone()

                if existing:
                    # Update existing entry (e.g., to add training type mapping)
                    existing_id = existing['id']
                    cur.execute("""
                        UPDATE activity_log
                        SET end_time = %s,
                            activity_type = %s,
                            duration_min = %s,
                            kcal_active = %s,
                            kcal_resting = %s,
                            hr_avg = %s,
                            hr_max = %s,
                            distance_km = %s,
                            training_type_id = %s,
                            training_category = %s,
                            training_subcategory = %s
                        WHERE id = %s
                    """, (
                        row.get('End',''), wtype, duration_min,
                        kj(row.get('Aktive Energie (kJ)','')),
                        kj(row.get('Ruheeinträge (kJ)','')),
                        tf(row.get('Durchschn. Herzfrequenz (count/min)','')),
                        tf(row.get('Max. Herzfrequenz (count/min)','')),
                        tf(row.get('Distanz (km)','')),
                        training_type_id, training_category, training_subcategory,
                        existing_id
                    ))
                    skipped += 1  # Count as skipped (not newly inserted)

                    # Phase 1.2: Auto-evaluation after CSV import UPDATE
                    if EVALUATION_AVAILABLE and training_type_id:
                        try:
                            # Build activity dict for evaluation
                            activity_dict = {
                                "id": existing_id,
                                "profile_id": pid,
                                "date": date,
                                "training_type_id": training_type_id,
                                "duration_min": duration_min,
                                "hr_avg": tf(row.get('Durchschn. Herzfrequenz (count/min)','')),
                                "hr_max": tf(row.get('Max. Herzfrequenz (count/min)','')),
                                "distance_km": tf(row.get('Distanz (km)','')),
                                "kcal_active": kj(row.get('Aktive Energie (kJ)','')),
                                "kcal_resting": kj(row.get('Ruheeinträge (kJ)','')),
                                "rpe": None,
                                "pace_min_per_km": None,
                                "cadence": None,
                                "elevation_gain": None
                            }
                            evaluate_and_save_activity(cur, existing_id, activity_dict, training_type_id, pid)
                            logger.debug(f"[AUTO-EVAL] Re-evaluated updated activity {existing_id}")
                        except Exception as eval_error:
                            logger.warning(f"[AUTO-EVAL] Failed to re-evaluate updated activity {existing_id}: {eval_error}")
                else:
                    # Insert new entry
                    new_id = str(uuid.uuid4())
                    cur.execute("""INSERT INTO activity_log
                        (id,profile_id,date,start_time,end_time,activity_type,duration_min,kcal_active,kcal_resting,
                         hr_avg,hr_max,distance_km,source,training_type_id,training_category,training_subcategory,created)
                        VALUES (%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,'apple_health',%s,%s,%s,CURRENT_TIMESTAMP)""",
                        (new_id,pid,date,start,row.get('End',''),wtype,duration_min,
                         kj(row.get('Aktive Energie (kJ)','')),kj(row.get('Ruheeinträge (kJ)','')),
                         tf(row.get('Durchschn. Herzfrequenz (count/min)','')),
                         tf(row.get('Max. Herzfrequenz (count/min)','')),
                         tf(row.get('Distanz (km)','')),
                         training_type_id,training_category,training_subcategory))
                    inserted+=1

                    # Phase 1.2: Auto-evaluation after CSV import INSERT
                    if EVALUATION_AVAILABLE and training_type_id:
                        try:
                            # Build activity dict for evaluation
                            activity_dict = {
                                "id": new_id,
                                "profile_id": pid,
                                "date": date,
                                "training_type_id": training_type_id,
                                "duration_min": duration_min,
                                "hr_avg": tf(row.get('Durchschn. Herzfrequenz (count/min)','')),
                                "hr_max": tf(row.get('Max. Herzfrequenz (count/min)','')),
                                "distance_km": tf(row.get('Distanz (km)','')),
                                "kcal_active": kj(row.get('Aktive Energie (kJ)','')),
                                "kcal_resting": kj(row.get('Ruheeinträge (kJ)','')),
                                "rpe": None,
                                "pace_min_per_km": None,
                                "cadence": None,
                                "elevation_gain": None
                            }
                            evaluate_and_save_activity(cur, new_id, activity_dict, training_type_id, pid)
                            logger.debug(f"[AUTO-EVAL] Evaluated imported activity {new_id}")
                        except Exception as eval_error:
                            logger.warning(f"[AUTO-EVAL] Failed to evaluate imported activity {new_id}: {eval_error}")
            except Exception as e:
                logger.warning(f"Import row failed: {e}")
                skipped+=1
    return {"inserted":inserted,"skipped":skipped,"message":f"{inserted} Trainings importiert"}