Spaces:
Running
Running
File size: 10,823 Bytes
93c9664 29ed661 93c9664 |
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 |
"""
V4 API endpoint for structured summarization with streaming.
"""
import json
import time
from fastapi import APIRouter, HTTPException, Request
from fastapi.responses import StreamingResponse
from app.api.v4.schemas import StructuredSummaryRequest
from app.core.logging import get_logger
from app.services.article_scraper import article_scraper_service
from app.services.structured_summarizer import structured_summarizer_service
router = APIRouter()
logger = get_logger(__name__)
@router.post("/scrape-and-summarize/stream")
async def scrape_and_summarize_stream(
request: Request, payload: StructuredSummaryRequest
):
"""
V4: Structured summarization with streaming support.
Supports two modes:
1. URL mode: Scrape article from URL then generate structured summary
2. Text mode: Generate structured summary from provided text
Returns structured JSON summary with:
- title: Click-worthy title
- main_summary: 2-4 sentence summary
- key_points: 3-5 bullet points
- category: Topic category
- sentiment: positive/negative/neutral
- read_time_min: Estimated reading time
Response format:
Server-Sent Events stream with:
- Metadata event (if include_metadata=true)
- Content chunks (streaming JSON tokens)
- Done event (final latency)
"""
request_id = getattr(request.state, "request_id", "unknown")
# Determine input mode and prepare data
if payload.url:
# URL Mode: Scrape + Summarize
logger.info(f"[{request_id}] V4 URL mode: {payload.url[:80]}...")
scrape_start = time.time()
try:
article_data = await article_scraper_service.scrape_article(
url=payload.url, use_cache=payload.use_cache
)
except Exception as e:
logger.error(f"[{request_id}] Scraping failed: {e}")
raise HTTPException(
status_code=502, detail=f"Failed to scrape article: {str(e)}"
)
scrape_latency_ms = (time.time() - scrape_start) * 1000
logger.info(
f"[{request_id}] Scraped in {scrape_latency_ms:.2f}ms, "
f"extracted {len(article_data['text'])} chars"
)
# Validate scraped content
if len(article_data["text"]) < 100:
raise HTTPException(
status_code=422,
detail="Insufficient content extracted from URL. "
"Article may be behind paywall or site may block scrapers.",
)
text_to_summarize = article_data["text"]
metadata = {
"input_type": "url",
"url": payload.url,
"title": article_data.get("title"),
"author": article_data.get("author"),
"date": article_data.get("date"),
"site_name": article_data.get("site_name"),
"scrape_method": article_data.get("method", "static"),
"scrape_latency_ms": scrape_latency_ms,
"extracted_text_length": len(article_data["text"]),
"style": payload.style.value,
}
else:
# Text Mode: Direct Summarization
logger.info(f"[{request_id}] V4 text mode: {len(payload.text)} chars")
text_to_summarize = payload.text
metadata = {
"input_type": "text",
"text_length": len(payload.text),
"style": payload.style.value,
}
# Stream structured summarization
return StreamingResponse(
_stream_generator(text_to_summarize, payload, metadata, request_id),
media_type="text/event-stream",
headers={
"Cache-Control": "no-cache",
"Connection": "keep-alive",
"X-Accel-Buffering": "no",
"X-Request-ID": request_id,
},
)
async def _stream_generator(text: str, payload, metadata: dict, request_id: str):
"""Generate SSE stream for structured summarization."""
# Send metadata event first
if payload.include_metadata:
metadata_event = {"type": "metadata", "data": metadata}
yield f"data: {json.dumps(metadata_event)}\n\n"
# Stream structured summarization chunks
summarization_start = time.time()
tokens_used = 0
try:
async for chunk in structured_summarizer_service.summarize_structured_stream(
text=text,
style=payload.style.value,
max_tokens=payload.max_tokens,
):
# Track tokens
if not chunk.get("done", False):
tokens_used = chunk.get("tokens_used", tokens_used)
# Forward chunks in SSE format
yield f"data: {json.dumps(chunk)}\n\n"
except Exception as e:
logger.error(f"[{request_id}] V4 summarization failed: {e}")
error_event = {"type": "error", "error": str(e), "done": True}
yield f"data: {json.dumps(error_event)}\n\n"
return
summarization_latency_ms = (time.time() - summarization_start) * 1000
# Calculate total latency (include scrape time for URL mode)
total_latency_ms = summarization_latency_ms
if metadata.get("input_type") == "url":
total_latency_ms += metadata.get("scrape_latency_ms", 0)
logger.info(
f"[{request_id}] V4 request completed in {total_latency_ms:.2f}ms "
f"(scrape: {metadata.get('scrape_latency_ms', 0):.2f}ms, "
f"summary: {summarization_latency_ms:.2f}ms)"
)
else:
logger.info(
f"[{request_id}] V4 text mode completed in {total_latency_ms:.2f}ms"
)
@router.post("/scrape-and-summarize/stream-ndjson")
async def scrape_and_summarize_stream_ndjson(
request: Request, payload: StructuredSummaryRequest
):
"""
V4: NDJSON patch-based structured summarization with streaming.
This is the NEW streaming protocol that outputs NDJSON patches.
Each event contains:
- delta: The patch object (e.g., {"op": "set", "field": "title", "value": "..."})
- state: The current accumulated state
- done: Boolean indicating completion
- tokens_used: Number of tokens generated
- latency_ms: Total latency (final event only)
Supports two modes:
1. URL mode: Scrape article from URL then generate structured summary
2. Text mode: Generate structured summary from provided text
Response format:
Server-Sent Events stream with:
- Metadata event (if include_metadata=true)
- NDJSON patch events (streaming state updates)
- Final event (with latency)
"""
request_id = getattr(request.state, "request_id", "unknown")
# Determine input mode and prepare data
if payload.url:
# URL Mode: Scrape + Summarize
logger.info(f"[{request_id}] V4 NDJSON URL mode: {payload.url[:80]}...")
scrape_start = time.time()
try:
article_data = await article_scraper_service.scrape_article(
url=payload.url, use_cache=payload.use_cache
)
except Exception as e:
logger.error(f"[{request_id}] Scraping failed: {e}")
raise HTTPException(
status_code=502, detail=f"Failed to scrape article: {str(e)}"
)
scrape_latency_ms = (time.time() - scrape_start) * 1000
logger.info(
f"[{request_id}] Scraped in {scrape_latency_ms:.2f}ms, "
f"extracted {len(article_data['text'])} chars"
)
# Validate scraped content
if len(article_data["text"]) < 100:
raise HTTPException(
status_code=422,
detail="Insufficient content extracted from URL. "
"Article may be behind paywall or site may block scrapers.",
)
text_to_summarize = article_data["text"]
metadata = {
"input_type": "url",
"url": payload.url,
"title": article_data.get("title"),
"author": article_data.get("author"),
"date": article_data.get("date"),
"site_name": article_data.get("site_name"),
"scrape_method": article_data.get("method", "static"),
"scrape_latency_ms": scrape_latency_ms,
"extracted_text_length": len(article_data["text"]),
"style": payload.style.value,
}
else:
# Text Mode: Direct Summarization
logger.info(f"[{request_id}] V4 NDJSON text mode: {len(payload.text)} chars")
text_to_summarize = payload.text
metadata = {
"input_type": "text",
"text_length": len(payload.text),
"style": payload.style.value,
}
# Stream NDJSON structured summarization
return StreamingResponse(
_stream_generator_ndjson(text_to_summarize, payload, metadata, request_id),
media_type="text/event-stream",
headers={
"Cache-Control": "no-cache",
"Connection": "keep-alive",
"X-Accel-Buffering": "no",
"X-Request-ID": request_id,
},
)
async def _stream_generator_ndjson(text: str, payload, metadata: dict, request_id: str):
"""Generate SSE stream for NDJSON patch-based structured summarization."""
# Send metadata event first
if payload.include_metadata:
metadata_event = {"type": "metadata", "data": metadata}
yield f"data: {json.dumps(metadata_event)}\n\n"
# Stream NDJSON structured summarization
summarization_start = time.time()
try:
async for (
event
) in structured_summarizer_service.summarize_structured_stream_ndjson(
text=text,
style=payload.style.value,
max_tokens=payload.max_tokens,
):
# Forward events in SSE format
yield f"data: {json.dumps(event)}\n\n"
except Exception as e:
logger.error(f"[{request_id}] V4 NDJSON summarization failed: {e}")
error_event = {
"delta": None,
"state": None,
"done": True,
"error": str(e),
}
yield f"data: {json.dumps(error_event)}\n\n"
return
summarization_latency_ms = (time.time() - summarization_start) * 1000
# Calculate total latency (include scrape time for URL mode)
total_latency_ms = summarization_latency_ms
if metadata.get("input_type") == "url":
total_latency_ms += metadata.get("scrape_latency_ms", 0)
logger.info(
f"[{request_id}] V4 NDJSON request completed in {total_latency_ms:.2f}ms "
f"(scrape: {metadata.get('scrape_latency_ms', 0):.2f}ms, "
f"summary: {summarization_latency_ms:.2f}ms)"
)
else:
logger.info(
f"[{request_id}] V4 NDJSON text mode completed in {total_latency_ms:.2f}ms"
)
|