File size: 18,984 Bytes
353a28d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2c94ade
353a28d
 
 
 
 
 
 
 
 
 
4f12e6d
353a28d
 
 
 
 
 
 
 
 
 
 
 
 
 
4f12e6d
 
 
 
 
 
 
 
 
 
 
 
 
 
353a28d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4f12e6d
353a28d
 
 
 
 
 
 
 
 
4f12e6d
353a28d
 
 
 
 
 
 
 
 
 
 
 
4f12e6d
353a28d
 
4f12e6d
353a28d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4f12e6d
353a28d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
import argparse
import csv
import html
import json
import os
import re
import sqlite3
import urllib.parse
from datetime import datetime, timezone
from pathlib import Path

import chromadb
import httpx
from dotenv import load_dotenv
from openai import OpenAI


load_dotenv()

BASE_DIR = Path(__file__).resolve().parent
DATA_DIR = BASE_DIR / "data"

SPECIES_CSV = BASE_DIR / "unique_species_labels.csv"
RAG_DB_PATH = Path(os.getenv("RAG_DB_PATH", str(DATA_DIR / "plant_rag")))
SQLITE_DB_PATH = Path(os.getenv("PLANTS_SQLITE_PATH", str(DATA_DIR / "plants.db")))

DEFAULT_MODEL = os.getenv("OPENAI_MODEL", "gpt-4o-mini")
PROFILE_KEYS = (
    "annaffiatura_gg",
    "annaffiatura_time",
    "luce",
    "temperatura",
    "umidita",
    "altezza_media",
    "pulizia",
    "terriccio",
    "concimazione",
    "prevenzione",
)

RHS_SEARCH_URL = "https://www.rhs.org.uk/plants/search-results?query={query}"
MISSOURI_SEARCH_URL = (
    "https://www.missouribotanicalgarden.org/PlantFinder/PlantFinderSearch.aspx?basic={query}"
)
EPPO_SEARCH_URL = "https://gd.eppo.int/search?query={query}"

HTTP_TIMEOUT = 12.0
HTTP_USER_AGENT = os.getenv(
    "EXTERNAL_SOURCES_USER_AGENT",
    "clorofilla/1.0 (contact: local-dev)",
)


def init_db(conn: sqlite3.Connection) -> None:
    conn.execute(
        """
        CREATE TABLE IF NOT EXISTS plants (
            id INTEGER PRIMARY KEY AUTOINCREMENT,
            species_name TEXT NOT NULL UNIQUE,
            indexed INTEGER NOT NULL DEFAULT 0,
            image_paths TEXT,
            annaffiatura_gg INTEGER,
            annaffiatura_time TEXT,
            luce TEXT,
            temperatura TEXT,
            umidita TEXT,
            altezza_media TEXT,
            pulizia TEXT,
            terriccio TEXT,
            concimazione TEXT,
            prevenzione TEXT,
            updated_at TEXT NOT NULL
        )
        """
    )
    # Migration for existing DBs created before image_paths support.
    try:
        conn.execute("ALTER TABLE plants ADD COLUMN image_paths TEXT")
        conn.commit()
    except Exception:
        pass
    conn.execute(
        """
        CREATE TABLE IF NOT EXISTS leafsnap_aliases (
            leafsnap_label TEXT PRIMARY KEY,
            db_species_name TEXT NOT NULL
        )
        """
    )
    conn.commit()


def load_species() -> list[str]:
    species: list[str] = []
    with open(SPECIES_CSV, "r", encoding="utf-8") as f:
        for row in csv.DictReader(f):
            name = (row.get("species_name") or "").strip()
            if name:
                species.append(name)
    return species


def get_rag_collection():
    client = chromadb.PersistentClient(path=str(RAG_DB_PATH))
    return client.get_collection(name="plants")


def get_rag_context(collection, species_name: str, max_chars: int = 9000) -> str:
    results = collection.get(
        where={"species_name": {"$eq": species_name}},
        limit=20,
    )
    docs = (results or {}).get("documents", [])
    if not docs:
        return ""

    context = "\n\n".join(docs)
    if len(context) > max_chars:
        context = context[:max_chars] + "\n..."
    return context


def _clean_json_payload(raw_text: str) -> str:
    txt = (raw_text or "").strip()
    if txt.startswith("```"):
        txt = txt.strip("`")
        if txt.startswith("json"):
            txt = txt[4:]
    return txt.strip()


def normalize_profile_data(data: dict) -> dict:
    allowed_keys = set(PROFILE_KEYS)
    normalized = {k: data.get(k) for k in allowed_keys}

    raw_days = normalized.get("annaffiatura_gg")
    if raw_days is None:
        normalized["annaffiatura_gg"] = None
    else:
        try:
            normalized["annaffiatura_gg"] = int(raw_days)
        except (TypeError, ValueError):
            normalized["annaffiatura_gg"] = None

    valid_time = {"mattino", "sera", "entrambi"}
    t = normalized.get("annaffiatura_time")
    if isinstance(t, str):
        t = t.strip().lower()
        normalized["annaffiatura_time"] = t if t in valid_time else None
    else:
        normalized["annaffiatura_time"] = None

    for key in allowed_keys - {"annaffiatura_gg", "annaffiatura_time"}:
        value = normalized.get(key)
        if value is None:
            continue
        normalized[key] = str(value).strip() or None

    return normalized


def _html_to_text(value: str) -> str:
    txt = re.sub(r"<script[\\s\\S]*?</script>", " ", value, flags=re.IGNORECASE)
    txt = re.sub(r"<style[\\s\\S]*?</style>", " ", txt, flags=re.IGNORECASE)
    txt = re.sub(r"<[^>]+>", " ", txt)
    txt = html.unescape(txt)
    txt = re.sub(r"\\s+", " ", txt)
    return txt.strip()


def _fetch_page_text(url: str, species_name: str, max_chars: int = 5000) -> str:
    headers = {
        "User-Agent": HTTP_USER_AGENT,
        "Accept": "text/html,application/xhtml+xml",
    }
    try:
        with httpx.Client(timeout=HTTP_TIMEOUT, follow_redirects=True, headers=headers) as client:
            resp = client.get(url)
            if resp.status_code != 200:
                return ""
            page_text = _html_to_text(resp.text)
            if not page_text:
                return ""
            species_low = species_name.lower()
            if species_low not in page_text.lower():
                return ""
            return page_text[:max_chars]
    except Exception:
        return ""


def fetch_external_sources(species_name: str) -> dict[str, str]:
    query = urllib.parse.quote_plus(species_name)
    rhs_text = _fetch_page_text(RHS_SEARCH_URL.format(query=query), species_name)
    missouri_text = _fetch_page_text(MISSOURI_SEARCH_URL.format(query=query), species_name)
    eppo_text = _fetch_page_text(EPPO_SEARCH_URL.format(query=query), species_name)

    return {
        "rhs": rhs_text,
        "missouri": missouri_text,
        "eppo": eppo_text,
    }


def normalize_profile_with_evidence(
    client: OpenAI,
    model: str,
    species_name: str,
    rag_context: str,
    partial_profile: dict | None,
    external_sources: dict[str, str] | None,
) -> dict:
    external_sources = external_sources or {}
    partial_profile = partial_profile or {}
    rhs = external_sources.get("rhs", "")
    missouri = external_sources.get("missouri", "")
    eppo = external_sources.get("eppo", "")

    system_msg = (
        "Sei un botanico professionista. Compila i campi solo usando le evidenze fornite. "
        "Priorita: RAG locale, poi RHS/Missouri per cura pratica, poi EPPO per prevenzione. "
        "Non inventare dati: se non ci sono evidenze affidabili usa null. "
        "Rispondi SOLO con JSON valido e senza testo extra."
    )

    user_msg = (
        f"Specie: {species_name}\n\n"
        "Profilo parziale gia estratto:\n"
        f"{json.dumps(partial_profile, ensure_ascii=False)}\n\n"
        "Compila/normalizza i campi JSON con queste chiavi esatte:\n"
        "annaffiatura_gg (numero intero o null),\n"
        "annaffiatura_time (mattino|sera|entrambi|null),\n"
        "luce, temperatura, umidita, altezza_media, pulizia, terriccio, concimazione, prevenzione.\n\n"
        "Evidenze RAG:\n"
        f"{rag_context or 'N/A'}\n\n"
        "Evidenze RHS (cura):\n"
        f"{rhs or 'N/A'}\n\n"
        "Evidenze Missouri Botanical Garden (cura):\n"
        f"{missouri or 'N/A'}\n\n"
        "Evidenze EPPO (prevenzione):\n"
        f"{eppo or 'N/A'}"
    )

    completion = client.chat.completions.create(
        model=model,
        temperature=0,
        response_format={"type": "json_object"},
        messages=[
            {"role": "system", "content": system_msg},
            {"role": "user", "content": user_msg},
        ],
    )
    payload = completion.choices[0].message.content or "{}"
    payload = _clean_json_payload(payload)
    data = json.loads(payload)
    return normalize_profile_data(data)


def extract_plant_profile(client: OpenAI, model: str, species_name: str, context: str) -> dict:
    system_msg = (
        "Sei un botanico professionista. Estrai solo dati supportati dal contesto fornito. "
        "Rispondi SOLO con JSON valido e senza testo extra. "
        "Se un dato manca, usa null."
    )
    user_msg = (
        f"Specie: {species_name}\n\n"
        "Estrai i seguenti campi in JSON con queste chiavi esatte:\n"
        "annaffiatura_gg (numero intero o null),\n"
        "annaffiatura_time (mattino|sera|entrambi|null),\n"
        "luce, temperatura, umidita, altezza_media, pulizia, terriccio, concimazione, prevenzione.\n"
        "\nContesto:\n"
        f"{context}"
    )

    completion = client.chat.completions.create(
        model=model,
        temperature=0,
        response_format={"type": "json_object"},
        messages=[
            {"role": "system", "content": system_msg},
            {"role": "user", "content": user_msg},
        ],
    )
    payload = completion.choices[0].message.content or "{}"
    payload = _clean_json_payload(payload)
    data = json.loads(payload)
    return normalize_profile_data(data)


def profile_has_missing_fields(profile: dict | None) -> bool:
    if not profile:
        return True
    return any(profile.get(key) is None for key in PROFILE_KEYS)


def merge_missing_fields(base_profile: dict | None, fallback_profile: dict | None) -> dict:
    merged = dict(base_profile or {})
    fallback_profile = fallback_profile or {}
    for key in PROFILE_KEYS:
        if merged.get(key) is None and fallback_profile.get(key) is not None:
            merged[key] = fallback_profile[key]
    return merged


def extract_plant_profile_generic(client: OpenAI, model: str, species_name: str, partial_profile: dict | None) -> dict:
    partial = partial_profile or {}
    system_msg = (
        "Sei un botanico professionista. Usa conoscenza generale botanica per stimare i campi mancanti. "
        "Rispondi SOLO con JSON valido e senza testo extra. "
        "Se non sei ragionevolmente sicuro, lascia null."
    )
    user_msg = (
        f"Specie: {species_name}\n\n"
        "Hai gia questi valori (da mantenere):\n"
        f"{json.dumps(partial, ensure_ascii=False)}\n\n"
        "Compila SOLO i campi mancanti in JSON con queste chiavi esatte:\n"
        "annaffiatura_gg (numero intero o null),\n"
        "annaffiatura_time (mattino|sera|entrambi|null),\n"
        "luce, temperatura, umidita, altezza_media, pulizia, terriccio, concimazione, prevenzione."
    )

    completion = client.chat.completions.create(
        model=model,
        temperature=0,
        response_format={"type": "json_object"},
        messages=[
            {"role": "system", "content": system_msg},
            {"role": "user", "content": user_msg},
        ],
    )
    payload = completion.choices[0].message.content or "{}"
    payload = _clean_json_payload(payload)
    data = json.loads(payload)
    return normalize_profile_data(data)


def upsert_plant(
    conn: sqlite3.Connection,
    species_name: str,
    indexed: bool,
    profile: dict | None,
    image_paths: str | None = None,
) -> None:
    now_iso = datetime.now(timezone.utc).isoformat()
    profile = profile or {}

    conn.execute(
        """
        INSERT INTO plants (
            species_name,
            indexed,
            image_paths,
            annaffiatura_gg,
            annaffiatura_time,
            luce,
            temperatura,
            umidita,
            altezza_media,
            pulizia,
            terriccio,
            concimazione,
            prevenzione,
            updated_at
        )
        VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
        ON CONFLICT(species_name) DO UPDATE SET
            indexed=excluded.indexed,
            image_paths=COALESCE(excluded.image_paths, plants.image_paths),
            annaffiatura_gg=excluded.annaffiatura_gg,
            annaffiatura_time=excluded.annaffiatura_time,
            luce=excluded.luce,
            temperatura=excluded.temperatura,
            umidita=excluded.umidita,
            altezza_media=excluded.altezza_media,
            pulizia=excluded.pulizia,
            terriccio=excluded.terriccio,
            concimazione=excluded.concimazione,
            prevenzione=excluded.prevenzione,
            updated_at=excluded.updated_at
        """,
        (
            species_name,
            1 if indexed else 0,
            image_paths,
            profile.get("annaffiatura_gg"),
            profile.get("annaffiatura_time"),
            profile.get("luce"),
            profile.get("temperatura"),
            profile.get("umidita"),
            profile.get("altezza_media"),
            profile.get("pulizia"),
            profile.get("terriccio"),
            profile.get("concimazione"),
            profile.get("prevenzione"),
            now_iso,
        ),
    )


def already_enriched(conn: sqlite3.Connection, species_name: str) -> bool:
    row = conn.execute(
        """
        SELECT indexed, annaffiatura_gg, annaffiatura_time, luce, temperatura,
               umidita, altezza_media, pulizia, terriccio, concimazione, prevenzione
        FROM plants
        WHERE species_name = ?
        """,
        (species_name,),
    ).fetchone()
    if not row:
        return False
    indexed = bool(row[0])
    any_data = any(value is not None and str(value).strip() != "" for value in row[1:])
    return indexed and any_data


def main() -> None:
    parser = argparse.ArgumentParser(
        description="Crea/aggiorna data/plants.db da CSV + RAG, con arricchimento OpenAI.",
    )
    parser.add_argument("--limit", type=int, default=0, help="Processa solo le prime N specie")
    parser.add_argument(
        "--force-refresh",
        action="store_true",
        help="Ricalcola anche le specie gia arricchite nel DB",
    )
    parser.add_argument(
        "--generic-fallback",
        action=argparse.BooleanOptionalAction,
        default=True,
        help="Se mancano campi, tenta una stima OpenAI senza contesto RAG (default: true)",
    )
    parser.add_argument(
        "--external-sources",
        action=argparse.BooleanOptionalAction,
        default=True,
        help="Integra fonti esterne (RHS, Missouri, EPPO) prima della normalizzazione finale",
    )
    parser.add_argument("--model", default=DEFAULT_MODEL, help="Modello OpenAI da usare")
    args = parser.parse_args()

    species = load_species()
    if args.limit and args.limit > 0:
        species = species[: args.limit]

    if not species:
        raise RuntimeError("Nessuna specie trovata nel CSV.")

    SQLITE_DB_PATH.parent.mkdir(parents=True, exist_ok=True)
    conn = sqlite3.connect(SQLITE_DB_PATH)
    init_db(conn)

    collection = get_rag_collection()

    api_key = os.getenv("OPENAI_API_KEY", "").strip()
    client = OpenAI(api_key=api_key) if api_key else None

    if client is None:
        print("OPENAI_API_KEY non impostata: verra compilato solo indexed true/false.")

    indexed_count = 0
    enriched_count = 0
    generic_fallback_count = 0
    external_sources_count = 0
    not_indexed_count = 0

    total = len(species)
    for i, species_name in enumerate(species, start=1):
        context = get_rag_context(collection, species_name)
        is_indexed = bool(context.strip())

        profile = None
        external_sources: dict[str, str] = {}
        if args.external_sources:
            external_sources = fetch_external_sources(species_name)
            if any(external_sources.values()):
                external_sources_count += 1

        if not is_indexed:
            if client is not None and args.generic_fallback:
                try:
                    profile = extract_plant_profile_generic(
                        client,
                        args.model,
                        species_name,
                        partial_profile=None,
                    )
                    if external_sources:
                        profile = normalize_profile_with_evidence(
                            client=client,
                            model=args.model,
                            species_name=species_name,
                            rag_context="",
                            partial_profile=profile,
                            external_sources=external_sources,
                        )
                    generic_fallback_count += 1
                    enriched_count += 1
                    print(f"[{i}/{total}] {species_name}: indexed=0, arricchita (fallback)")
                except Exception as exc:
                    print(f"[{i}/{total}] {species_name}: indexed=0, errore fallback OpenAI ({exc})")
            upsert_plant(conn, species_name, indexed=False, profile=profile)
            not_indexed_count += 1
            if profile is None:
                print(f"[{i}/{total}] {species_name}: indexed=0")
            continue

        indexed_count += 1
        if not args.force_refresh and already_enriched(conn, species_name):
            upsert_plant(conn, species_name, indexed=True, profile=None)
            print(f"[{i}/{total}] {species_name}: indexed=1 (gia arricchita)")
            continue

        if client is not None:
            try:
                profile = extract_plant_profile(client, args.model, species_name, context)
                if external_sources:
                    profile = normalize_profile_with_evidence(
                        client=client,
                        model=args.model,
                        species_name=species_name,
                        rag_context=context,
                        partial_profile=profile,
                        external_sources=external_sources,
                    )
                enriched_count += 1
                if args.generic_fallback and profile_has_missing_fields(profile):
                    fallback = extract_plant_profile_generic(client, args.model, species_name, partial_profile=profile)
                    profile = merge_missing_fields(profile, fallback)
                    generic_fallback_count += 1
                    print(f"[{i}/{total}] {species_name}: indexed=1, arricchita + fallback")
                else:
                    print(f"[{i}/{total}] {species_name}: indexed=1, arricchita")
            except Exception as exc:
                print(f"[{i}/{total}] {species_name}: indexed=1, errore OpenAI ({exc})")
        else:
            print(f"[{i}/{total}] {species_name}: indexed=1")

        upsert_plant(conn, species_name, indexed=True, profile=profile)

        if i % 50 == 0:
            conn.commit()

    conn.commit()
    conn.close()

    print("\n=== Completato ===")
    print(f"DB SQLite: {SQLITE_DB_PATH}")
    print(f"Specie processate: {total}")
    print(f"Presenti in RAG (indexed=1): {indexed_count}")
    print(f"Non presenti in RAG (indexed=0): {not_indexed_count}")
    print(f"Arricchite con OpenAI: {enriched_count}")
    print(f"Fallback generico usato: {generic_fallback_count}")
    print(f"Fonti esterne usate: {external_sources_count}")


if __name__ == "__main__":
    main()