File size: 15,803 Bytes
7d07e42
 
 
 
149c85a
 
 
 
 
 
 
7d07e42
 
 
 
149c85a
7d07e42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ea25e34
 
7d07e42
 
 
 
 
 
 
 
 
 
149c85a
 
 
02f4591
7d07e42
 
 
 
 
 
 
 
 
 
ea25e34
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7d07e42
 
 
ea25e34
7d07e42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ea25e34
 
 
 
 
 
 
 
 
 
 
7d07e42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
149c85a
 
 
7d07e42
 
149c85a
 
 
 
7d07e42
 
 
149c85a
 
 
 
7d07e42
 
149c85a
 
c3356eb
7d07e42
 
 
 
 
 
 
 
 
 
 
c3356eb
 
4d62ba6
c3356eb
 
7d07e42
 
 
 
 
 
c3356eb
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7d07e42
 
 
 
 
 
149c85a
7d07e42
 
 
63fae6b
 
7d07e42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
63fae6b
 
149c85a
 
 
7d07e42
 
 
 
 
149c85a
4d62ba6
149c85a
63fae6b
 
 
 
7d07e42
63fae6b
 
 
 
 
7d07e42
 
 
63fae6b
7d07e42
 
 
 
 
 
149c85a
7d07e42
 
 
 
 
 
 
 
 
 
ea25e34
 
7d07e42
 
 
149c85a
7d07e42
 
 
149c85a
 
 
7d07e42
149c85a
7d07e42
 
 
 
 
 
 
 
 
ea25e34
149c85a
7d07e42
 
 
149c85a
 
7d07e42
 
149c85a
 
 
7d07e42
 
 
 
 
 
 
 
 
ea25e34
 
7d07e42
 
 
149c85a
 
7d07e42
 
 
149c85a
7d07e42
 
 
149c85a
 
7d07e42
 
 
 
 
 
 
ea25e34
 
7d07e42
 
149c85a
 
 
 
7d07e42
 
149c85a
 
 
7d07e42
 
 
149c85a
 
7d07e42
 
 
149c85a
 
7d07e42
 
 
149c85a
 
 
 
 
 
 
 
7d07e42
149c85a
7d07e42
 
 
 
 
 
149c85a
 
 
 
 
 
 
 
 
 
 
 
02f4591
 
 
 
 
 
149c85a
 
 
7d07e42
 
 
149c85a
 
 
 
 
7d07e42
149c85a
7d07e42
149c85a
7d07e42
149c85a
 
7d07e42
 
 
 
 
 
 
149c85a
7d07e42
149c85a
7d07e42
 
149c85a
 
7d07e42
 
149c85a
7d07e42
149c85a
7d07e42
149c85a
 
 
 
7d07e42
 
 
 
 
 
 
 
 
149c85a
 
 
 
7d07e42
 
 
 
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
"""
CV API routes.

Pola yang dipakai:
- Semua heavy endpoint pakai `def` (bukan `async def`) supaya FastAPI
  jalankan di threadpool, bukan event loop. Ini prevent blocking event loop
  yang menyebabkan health checks, readiness polling, dan semua request lain
  ikut hang.
- Endpoint /ready dipakai UI buat polling.
- _MODEL_WAIT_TIMEOUT dikurangi ke 30s (HF edge proxy timeout ~60s,
  jadi ada buffer 30s untuk compute aktual).
"""

from __future__ import annotations

import asyncio
import threading
from fastapi import APIRouter, HTTPException, UploadFile, File, status
from fastapi.responses import Response
from pydantic import BaseModel
from loguru import logger

from .schemas import (
    AnalyzeURLRequest, FullAnalysisResponse,
    ClassifyRequest, ClassificationResponse,
    SimilarityRequest, SimilarityResponse,
    VisualQARequest, VisualQAResponse,
    CaptionResponse, DetectionResponse, OCRResponse,
)
from .readiness import get_readiness
from ..cv_pipeline import CVPipeline

router = APIRouter()

_pipeline: CVPipeline = None
_pipeline_lock = threading.Lock()

_trigger_lock = threading.Lock()


def get_pipeline() -> CVPipeline:
    global _pipeline
    if _pipeline is None:
        with _pipeline_lock:
            if _pipeline is None:
                _pipeline = CVPipeline()
    return _pipeline


# Dikurangi dari 180s ke 30s.
# Semua model prewarmed dalam <2s. Kalau masih belum ready dalam 30s,
# ada masalah serius — lebih baik fail fast daripada block sampai HF proxy timeout.
_MODEL_WAIT_TIMEOUT = 90.0


def _trigger_and_wait(model_name: str):
    """
    Trigger lazy load model (akses pipeline property),
    lalu tunggu ReadinessTracker konfirmasi ready.
    Thread-safe: hanya satu thread yang load, sisanya tunggu.
    """
    readiness = get_readiness()

    with _trigger_lock:
        status_info = readiness.get_status(model_name)

        if status_info.state.value == "error":
            raise HTTPException(
                status_code=503,
                detail={
                    "error": "model_failed_to_load",
                    "model": model_name,
                    "message": status_info.error_message or "Model gagal dimuat.",
                    "hint": "Cek logs container untuk detail error.",
                },
            )

        need_spawn = status_info.state.value in ("not_loaded",)
        if need_spawn:
            readiness.mark_loading(model_name)

    if status_info.state.value == "ready":
        return

    if need_spawn:
        def _do_load():
            try:
                p = get_pipeline()
                if model_name == "captioner":
                    _ = p.captioner
                elif model_name == "yolo":
                    _ = p.yolo
                elif model_name == "clip":
                    _ = p.clip
                elif model_name == "ocr":
                    _ = p.ocr
                readiness.mark_ready(model_name)
                logger.info(f"Model '{model_name}' lazy-loaded dan ready.")
            except Exception as e:
                readiness.mark_error(model_name, str(e))
                logger.error(f"Lazy-load '{model_name}' failed: {e}")

        t = threading.Thread(target=_do_load, daemon=True, name=f"lazy-load-{model_name}")
        t.start()

    ok = readiness.wait_for(model_name, timeout=_MODEL_WAIT_TIMEOUT)
    if not ok:
        current = readiness.get_status(model_name).state.value
        if current == "error":
            err_msg = readiness.get_status(model_name).error_message
            raise HTTPException(
                status_code=503,
                detail={
                    "error": "model_failed_to_load",
                    "model": model_name,
                    "message": err_msg or f"Model '{model_name}' gagal dimuat.",
                    "hint": "Cek logs container untuk traceback lengkap.",
                },
            )
        raise HTTPException(
            status_code=503,
            detail={
                "error": "model_not_ready",
                "model": model_name,
                "current_state": current,
                "message": f"Model '{model_name}' belum siap setelah {_MODEL_WAIT_TIMEOUT}s.",
                "hint": "Cek GET /api/v1/ready. Coba request lagi dalam beberapa saat.",
            },
        )


def _ensure_models_ready(*model_names: str):
    """Pastikan semua model yang dibutuhkan endpoint sudah ready."""
    for name in model_names:
        _trigger_and_wait(name)


# === HEALTH & READINESS ===

@router.get("/health", tags=["meta"])
def health():
    return {"status": "ok"}


@router.get("/ready", tags=["meta"])
def ready():
    readiness = get_readiness()
    snap = readiness.snapshot()
    return snap


# === ANALYSIS ENDPOINTS ===
# PENTING: semua heavy endpoint pakai `def` bukan `async def`.
# FastAPI otomatis jalankan sync def di threadpool (anyio worker thread),
# sehingga blocking code (httpx, ONNX, Tesseract) tidak freeze event loop.

@router.post("/analyze/url", response_model=FullAnalysisResponse, tags=["analysis"])
def analyze_from_url(req: AnalyzeURLRequest):
    """Analisis gambar dari URL (caption + opsional detection/OCR/CLIP)."""
    import concurrent.futures as _cf
    needed = []
    if req.run_caption:
        needed.append("captioner")
    if req.run_detection:
        needed.append("yolo")
    if req.classification_labels:
        needed.append("clip")
    if req.run_ocr:
        needed.append("ocr")
    _ensure_models_ready(*needed)

    # Hard outer deadline — TOTAL_TIMEOUT (40s) di cv_pipeline sudah handle ini,
    # tapi kita tambah satu lapis lagi di route untuk jaga-jaga.
    ROUTE_TIMEOUT = 78.0  # sedikit lebih dari CVPipeline.TOTAL_TIMEOUT
    def _run_analyze():
        return get_pipeline().analyze(
            source=req.url,
            run_caption=req.run_caption,
            run_detection=req.run_detection,
            run_ocr=req.run_ocr,
            classification_labels=req.classification_labels,
        )

    with _cf.ThreadPoolExecutor(max_workers=1) as exc:
        fut = exc.submit(_run_analyze)
        try:
            result = fut.result(timeout=ROUTE_TIMEOUT)
        except _cf.TimeoutError:
            raise HTTPException(
                status_code=504,
                detail=(
                    "Analyze timeout setelah 42s. "
                    "Kemungkinan server gambar lambat atau memblok HF. "
                    "Coba URL gambar lain (imgur, ibb.co, raw GitHub, dll)."
                ),
            )
        except HTTPException:
            raise
        except Exception as e:
            logger.error(f"Analyze error: {e}")
            raise HTTPException(status_code=500, detail=str(e))

    return _to_response(result)


@router.post("/analyze/upload", response_model=FullAnalysisResponse, tags=["analysis"])
async def analyze_upload(
    file: UploadFile = File(...),
    run_caption: bool = True,
    run_detection: bool = False,
    run_ocr: bool = False,
):
    """Upload dan analisis gambar langsung (multipart)."""
    import concurrent.futures as _cf

    allowed = {"image/jpeg", "image/png", "image/webp", "image/gif"}
    if file.content_type not in allowed:
        raise HTTPException(400, detail=f"Tipe file tidak didukung: {file.content_type}")

    data = await file.read()
    if len(data) > 10 * 1024 * 1024:
        raise HTTPException(400, detail="Ukuran file maksimum 10MB")

    needed = []
    if run_caption:
        needed.append("captioner")
    if run_detection:
        needed.append("yolo")
    if run_ocr:
        needed.append("ocr")

    # Run blocking work in threadpool so we don't block event loop.
    # Pakai get_running_loop() (bukan get_event_loop() yang deprecated di Py3.10+).
    def _run():
        _ensure_models_ready(*needed)
        return get_pipeline().analyze(
            source=data,
            run_caption=run_caption,
            run_detection=run_detection,
            run_ocr=run_ocr,
        )

    UPLOAD_TIMEOUT = 78.0  # sama dengan analyze/url — hard deadline
    try:
        loop = asyncio.get_running_loop()
        with _cf.ThreadPoolExecutor(max_workers=1) as exc:
            fut = loop.run_in_executor(exc, _run)
            result = await asyncio.wait_for(fut, timeout=UPLOAD_TIMEOUT)
        return _to_response(result)
    except asyncio.TimeoutError:
        raise HTTPException(
            status_code=504,
            detail="Upload analyze timeout setelah 42s. Coba lagi atau kurangi ukuran gambar.",
        )
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Upload analyze error: {e}")
        raise HTTPException(status_code=500, detail=str(e))


# === INDIVIDUAL TASKS ===

@router.post("/caption", response_model=CaptionResponse, tags=["tasks"])
def caption(url: str, prompt: str = None):
    """Generate deskripsi teks dari gambar."""
    _ensure_models_ready("captioner")
    try:
        from ..processors.image_preprocessor import ImagePreprocessor
        image = ImagePreprocessor.load(url)
        result = get_pipeline().captioner.caption(image, prompt=prompt)
        return CaptionResponse(caption=result.caption, model=result.model)
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Caption error: {e}")
        raise HTTPException(500, detail=f"Caption gagal: {e}")


@router.post("/detect", response_model=DetectionResponse, tags=["tasks"])
def detect(url: str, conf: float = None):
    """Deteksi objek dalam gambar dengan YOLOv8."""
    _ensure_models_ready("yolo")
    try:
        from ..processors.image_preprocessor import ImagePreprocessor
        image = ImagePreprocessor.load(url)
        result = get_pipeline().yolo.detect(image, conf_threshold=conf)
        return DetectionResponse(
            detections=[_det_to_schema(d) for d in result.detections],
            count=result.count,
            labels_summary=result.labels_summary,
            image_width=result.image_width,
            image_height=result.image_height,
            inference_time_ms=result.inference_time_ms,
        )
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Detect error: {e}")
        raise HTTPException(500, detail=f"Detection gagal: {e}")


@router.post("/classify", response_model=ClassificationResponse, tags=["tasks"])
def classify(req: ClassifyRequest):
    """Zero-shot classification dengan CLIP."""
    _ensure_models_ready("clip")
    try:
        from ..processors.image_preprocessor import ImagePreprocessor
        image = ImagePreprocessor.load(req.url)
        result = get_pipeline().clip.classify(image, req.labels)
        return ClassificationResponse(
            top_label=result.top_label,
            top_score=result.top_score,
            labels=result.labels,
            probabilities=result.probabilities,
        )
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Classify error: {e}")
        raise HTTPException(500, detail=f"Classify gagal: {e}")


@router.post("/ocr", response_model=OCRResponse, tags=["tasks"])
def ocr(url: str):
    """Ekstrak teks dari gambar dengan Tesseract OCR."""
    _ensure_models_ready("ocr")
    try:
        from ..processors.image_preprocessor import ImagePreprocessor
        image = ImagePreprocessor.load(url)
        result = get_pipeline().ocr.extract_text(image)
        return OCRResponse(
            full_text=result.full_text,
            boxes=[{"text": b.text, "confidence": b.confidence, "bbox": b.bbox}
                   for b in (result.boxes or [])],
            word_count=result.word_count,
            language=result.language,
            engine=result.engine,
        )
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"OCR error: {e}")
        raise HTTPException(500, detail=f"OCR gagal: {e}")


@router.post("/vqa", response_model=VisualQAResponse, tags=["tasks"])
def visual_qa(req: VisualQARequest):
    """Visual Question Answering."""
    _ensure_models_ready("captioner")
    try:
        from ..processors.image_preprocessor import ImagePreprocessor
        image = ImagePreprocessor.load(req.url)
        result = get_pipeline().captioner.visual_qa(image, req.question)
        return VisualQAResponse(question=req.question, answer=result.caption)
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"VQA error: {e}")
        raise HTTPException(500, detail=f"VQA gagal: {e}")


@router.post("/similarity", response_model=SimilarityResponse, tags=["tasks"])
def similarity(req: SimilarityRequest):
    """Hitung similarity antara gambar dan teks dengan CLIP."""
    _ensure_models_ready("clip")
    try:
        score = get_pipeline().image_text_similarity(req.url, req.text)
        if score > 0.3:
            interpretation = "Sangat relevan"
        elif score > 0.2:
            interpretation = "Cukup relevan"
        elif score > 0.1:
            interpretation = "Sedikit relevan"
        else:
            interpretation = "Tidak relevan"
        return SimilarityResponse(
            similarity_score=round(score, 4),
            text=req.text,
            interpretation=interpretation,
        )
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Similarity error: {e}")
        raise HTTPException(500, detail=f"Similarity gagal: {e}")


# === HELPERS ===

def _det_to_schema(d):
    from .schemas import DetectionSchema, BBoxSchema
    return DetectionSchema(
        label=d.label,
        confidence=d.confidence,
        bbox=BBoxSchema(
            x1=d.bbox.x1,
            y1=d.bbox.y1,
            x2=d.bbox.x2,
            y2=d.bbox.y2,
            width=d.bbox.width,
            height=d.bbox.height,
        ),
        class_id=d.class_id,
    )


def _to_response(result) -> FullAnalysisResponse:
    from .schemas import (
        FullAnalysisResponse, CaptionResponse, DetectionResponse,
        ClassificationResponse, OCRResponse, BBoxSchema, DetectionSchema, OCRBoxSchema,
    )
    cap = None
    if result.caption:
        cap = CaptionResponse(caption=result.caption.caption, model=result.caption.model)

    det = None
    if result.detections:
        det = DetectionResponse(
            detections=[_det_to_schema(d) for d in result.detections.detections],
            count=result.detections.count,
            labels_summary=result.detections.labels_summary,
            image_width=result.detections.image_width,
            image_height=result.detections.image_height,
            inference_time_ms=result.detections.inference_time_ms,
        )

    cls = None
    if result.classification:
        cls = ClassificationResponse(
            top_label=result.classification.top_label,
            top_score=result.classification.top_score,
            labels=result.classification.labels or [],
            probabilities=result.classification.probabilities or [],
        )

    ocr = None
    if result.ocr:
        ocr = OCRResponse(
            full_text=result.ocr.full_text,
            boxes=[
                OCRBoxSchema(text=b.text, confidence=b.confidence, bbox=b.bbox)
                for b in (result.ocr.boxes or [])
            ],
            word_count=result.ocr.word_count,
            language=result.ocr.language,
            engine=result.ocr.engine,
        )

    return FullAnalysisResponse(
        image_width=result.image_width,
        image_height=result.image_height,
        source=result.source,
        caption=cap,
        detections=det,
        classification=cls,
        ocr=ocr,
        summary_text=result.to_summary(),
        models_used=result.models_used,
        total_latency_ms=result.total_latency_ms,
    )