64 lines
2.7 KiB
Python
64 lines
2.7 KiB
Python
|
"""Additional integrity checks"""
|
||
|
|
import asyncio
|
||
|
|
from sqlalchemy import text
|
||
|
|
from sqlalchemy.ext.asyncio import create_async_engine
|
||
|
|
|
||
|
|
DATABASE_URL = 'postgresql+asyncpg://scilit:scilit_prod_2026@postgres:5432/scilit'
|
||
|
|
|
||
|
|
async def main():
|
||
|
|
engine = create_async_engine(DATABASE_URL)
|
||
|
|
async with engine.connect() as conn:
|
||
|
|
# 1. Records where article_date has real month/day but pub_date is Jan-01
|
||
|
|
r = await conn.execute(text("""
|
||
|
|
SELECT id, pmid, pub_date, article_date FROM global_literature
|
||
|
|
WHERE EXTRACT(MONTH FROM pub_date)=1 AND EXTRACT(DAY FROM pub_date)=1
|
||
|
|
AND article_date IS NOT NULL
|
||
|
|
AND NOT (EXTRACT(MONTH FROM article_date)=1 AND EXTRACT(DAY FROM article_date)=1)
|
||
|
|
LIMIT 20
|
||
|
|
"""))
|
||
|
|
print('=== Records where article_date can fix pub_date ===')
|
||
|
|
for row in r:
|
||
|
|
print(f' pmid={row.pmid} pub_date={row.pub_date} article_date={row.article_date}')
|
||
|
|
|
||
|
|
# 2. Check for records with NULL pub_year but valid pub_date
|
||
|
|
r = await conn.execute(text("""
|
||
|
|
SELECT COUNT(*) FROM global_literature
|
||
|
|
WHERE pub_date IS NOT NULL AND pub_year IS NULL
|
||
|
|
"""))
|
||
|
|
print(f'\npub_date set but pub_year NULL: {r.scalar()}')
|
||
|
|
|
||
|
|
# 3. Check for records with NULL pub_date but valid pub_year
|
||
|
|
r = await conn.execute(text("""
|
||
|
|
SELECT COUNT(*) FROM global_literature
|
||
|
|
WHERE pub_date IS NULL AND pub_year IS NOT NULL
|
||
|
|
"""))
|
||
|
|
print(f'pub_year set but pub_date NULL: {r.scalar()}')
|
||
|
|
|
||
|
|
# 4. Check if there are records WITHOUT any date info at all
|
||
|
|
r = await conn.execute(text("""
|
||
|
|
SELECT COUNT(*) FROM global_literature
|
||
|
|
WHERE pub_date IS NULL AND pub_year IS NULL AND article_date IS NULL
|
||
|
|
"""))
|
||
|
|
print(f'No date info at all: {r.scalar()}')
|
||
|
|
|
||
|
|
# 5. Latest literature records - sample top 10 for sanity check
|
||
|
|
r = await conn.execute(text("""
|
||
|
|
SELECT pmid, pub_date, pub_year, title FROM global_literature
|
||
|
|
ORDER BY pub_year DESC NULLS LAST, pmid DESC LIMIT 5
|
||
|
|
"""))
|
||
|
|
print('\n=== Latest 5 records ===')
|
||
|
|
for row in r:
|
||
|
|
title_short = (row.title or '')[:60]
|
||
|
|
print(f' PMID={row.pmid} date={row.pub_date} year={row.pub_year} title={title_short}...')
|
||
|
|
|
||
|
|
# 6. Check for records that are in user_feed but patient hasn't been generated
|
||
|
|
r = await conn.execute(text("""
|
||
|
|
SELECT COUNT(*) FROM user_feed uf
|
||
|
|
WHERE NOT EXISTS (SELECT 1 FROM global_literature gl WHERE gl.id = uf.literature_id)
|
||
|
|
"""))
|
||
|
|
print(f'\nUserFeed pointing to non-existent literature: {r.scalar()}')
|
||
|
|
|
||
|
|
await engine.dispose()
|
||
|
|
|
||
|
|
asyncio.run(main())
|