feat: Refactor outlets with multilingual support and dynamic queries

- Replace static articles array with dynamic source_keyword queries
- Use MongoDB _id as unique identifier for outlets
- Add multilingual translations (9 languages: ko, en, zh_cn, zh_tw, ja, fr, de, es, it)
- Add OutletService for database operations
- Add outlet migration script with Korean source_keyword matching
- Remove JSON file-based outlet loading
- Add /outlets/{outlet_id}/articles endpoint for dynamic article retrieval

This resolves the design issues with:
1. Static articles array requiring constant updates
2. Lack of multilingual support for outlet names/descriptions
3. Broken image URLs
4. Korean entity matching for article queries

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
jungwoo choi
2025-10-13 16:52:34 +09:00
parent deb52e51f2
commit e467e76d02
6 changed files with 515 additions and 28 deletions

View File

@ -2,25 +2,14 @@ from fastapi import APIRouter, HTTPException, Query
from typing import Optional
from app.services.article_service import ArticleService
from app.services.comment_service import CommentService
from app.services.outlet_service import OutletService
from app.models.article import ArticleList, Article, ArticleSummary
from app.models.comment import Comment, CommentCreate, CommentList
from app.models.outlet import Outlet
from typing import List
import json
import os
router = APIRouter()
# Load outlets data
OUTLETS_FILE = os.path.join(os.path.dirname(__file__), '../../outlets-extracted.json')
outlets_data = None
def load_outlets():
global outlets_data
if outlets_data is None:
with open(OUTLETS_FILE, 'r', encoding='utf-8') as f:
outlets_data = json.load(f)
return outlets_data
@router.get("/{language}/articles", response_model=ArticleList)
async def get_articles(
language: str,
@ -84,28 +73,48 @@ async def get_categories(language: str):
@router.get("/outlets")
async def get_outlets(category: Optional[str] = Query(None, description="Filter by category: people, topics, companies")):
"""Get outlets list - people, topics, companies"""
data = load_outlets()
if category:
if category in ['people', 'topics', 'companies']:
return {category: data[category]}
else:
raise HTTPException(status_code=400, detail=f"Invalid category: {category}. Must be one of: people, topics, companies")
# Get outlets for specific category
outlets = await OutletService.get_all_outlets(category=category)
return {category: outlets}
return data
# Get all outlets grouped by category
result = {}
for cat in ['people', 'topics', 'companies']:
outlets = await OutletService.get_all_outlets(category=cat)
result[cat] = outlets
return result
@router.get("/outlets/{outlet_id}")
async def get_outlet_by_id(outlet_id: str):
"""Get specific outlet by ID"""
data = load_outlets()
"""Get specific outlet by ID (_id)"""
outlet = await OutletService.get_outlet_by_id(outlet_id)
return outlet
# Search in all categories
for category in ['people', 'topics', 'companies']:
for outlet in data[category]:
if outlet['id'] == outlet_id:
return outlet
@router.get("/{language}/outlets/{outlet_id}/articles")
async def get_outlet_articles(
language: str,
outlet_id: str,
page: int = Query(1, ge=1, description="Page number"),
page_size: int = Query(20, ge=1, le=100, description="Items per page")
):
"""Get articles for a specific outlet using source_keyword"""
if not ArticleService.validate_language(language):
raise HTTPException(status_code=400, detail=f"Unsupported language: {language}")
raise HTTPException(status_code=404, detail=f"Outlet not found: {outlet_id}")
# Get outlet to retrieve source_keyword
outlet = await OutletService.get_outlet_by_id(outlet_id)
# Query articles by source_keyword dynamically
articles_result = await ArticleService.get_articles_by_source_keyword(
language,
outlet['source_keyword'],
page,
page_size
)
return articles_result
# Comment endpoints
@router.get("/comments", response_model=CommentList)