Spaces:
Running
Running
File size: 5,639 Bytes
27d04ef | 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 | """
API routes for topic extraction v2.
"""
from fastapi import APIRouter, HTTPException
from pydantic import BaseModel
from typing import List, Optional, Dict
from pathlib import Path
import json
from services.micro_topic_service import (
extract_micro_topics_v2,
process_events_batch,
get_aggregated_topics,
extract_hashtags
)
topic_router = APIRouter(prefix="/api/topics", tags=["Topics"])
class ExtractRequest(BaseModel):
text: str
language_type: str = "english"
class ExtractResponse(BaseModel):
text: str
language_type: str
hashtags: List[str]
ner: List[str]
nouns: List[str]
text_v1: Optional[str] = None
micro_topics: List[str]
@topic_router.post("/extract", response_model=ExtractResponse)
async def extract_topics_single(request: ExtractRequest):
"""
Extract micro topics from a single text.
Useful for testing the extraction pipeline.
"""
# Create a mock event to use the extraction function
mock_event = {
"type": "watch",
"engagement": "active",
"text_clean": request.text,
"language_type": request.language_type
}
# Process it
result = extract_micro_topics_v2(mock_event)
return ExtractResponse(
text=request.text,
language_type=request.language_type,
hashtags=result.get("hashtags", []),
ner=result.get("ner", []),
nouns=result.get("nouns", []),
text_v1=result.get("text_v1"),
micro_topics=result.get("micro_topics", [])
)
@topic_router.post("/{token}/enrich")
async def enrich_session_topics(token: str):
"""
Enrich all events in a session with micro topics.
Only processes events with type=watch and engagement=active.
Adds hashtags, ner, nouns, text_v1, and micro_topics fields.
"""
storage_dir = Path("storage")
file_path = storage_dir / f"preprocessed_{token}.json"
if not file_path.exists():
raise HTTPException(status_code=404, detail="Session not found")
# Load events
with open(file_path, "r", encoding="utf-8") as f:
data = json.load(f)
events = data.get("events", [])
# Count qualifying events before processing
active_watch_count = sum(
1 for e in events
if e.get("type") == "watch" and e.get("engagement") == "active"
)
# Process events
processed_events = process_events_batch(events)
# Count results
events_with_topics = sum(1 for e in processed_events if e.get("micro_topics"))
total_topics = sum(len(e.get("micro_topics", [])) for e in processed_events)
total_hashtags = sum(len(e.get("hashtags", [])) for e in processed_events)
total_ner = sum(len(e.get("ner", [])) for e in processed_events)
total_nouns = sum(len(e.get("nouns", [])) for e in processed_events)
# Save updated data
data["events"] = processed_events
data["micro_topics_extracted"] = True
data["extraction_version"] = "v2"
with open(file_path, "w", encoding="utf-8") as f:
json.dump(data, f, ensure_ascii=False, indent=2)
return {
"token": token,
"total_events": len(events),
"active_watch_events": active_watch_count,
"events_with_topics": events_with_topics,
"extraction_stats": {
"total_hashtags": total_hashtags,
"total_ner": total_ner,
"total_nouns": total_nouns,
"total_micro_topics": total_topics
},
"status": "enriched"
}
@topic_router.get("/{token}/aggregate")
async def get_session_topics(token: str, top_n: int = 50):
"""
Get aggregated micro topics for a session.
Returns:
- top_hashtags: Most common hashtags
- top_ner: Most common named entities
- top_nouns: Most common nouns
- top_micro_topics: Most common overall
"""
storage_dir = Path("storage")
file_path = storage_dir / f"preprocessed_{token}.json"
if not file_path.exists():
raise HTTPException(status_code=404, detail="Session not found")
# Load events
with open(file_path, "r", encoding="utf-8") as f:
data = json.load(f)
events = data.get("events", [])
# Check if topics are extracted
if not data.get("micro_topics_extracted"):
raise HTTPException(
status_code=400,
detail="Topics not extracted yet. Call POST /{token}/enrich first."
)
# Aggregate topics
aggregated = get_aggregated_topics(events, top_n)
# Add language breakdown
from collections import Counter
language_topics = {"english": [], "hindi": [], "hinglish": [], "unknown": []}
for event in events:
if event.get("type") == "watch" and event.get("engagement") == "active":
lang = event.get("language_type", "unknown")
topics = event.get("micro_topics", [])
if lang in language_topics:
language_topics[lang].extend(topics)
language_breakdown = {
lang: [{"topic": t, "count": c} for t, c in Counter(topics).most_common(20)]
for lang, topics in language_topics.items()
if topics # Only include languages with topics
}
return {
"token": token,
"version": data.get("extraction_version", "v1"),
"stats": aggregated["stats"],
"top_hashtags": aggregated["top_hashtags"],
"top_ner": aggregated["top_ner"],
"top_nouns": aggregated["top_nouns"],
"top_micro_topics": aggregated["top_micro_topics"],
"by_language": language_breakdown
}
|