Add fastapi-cache2 for pydantic v2 compatibility

requirements.txt

@@ -8,3 +8,4 @@ s3fs
 uvicorn
 xarray>=2022.06
 zarr
+fastapi-cache2 @ git+https://github.com/andersy005/fastapi-cache@pydantic-v2-compat # for pydantic v2 compatilibity

src/app.py

@@ -1,13 +1,57 @@
 from __future__ import annotations
 
+import os
+import traceback
+from contextlib import asynccontextmanager
+
 import pydantic
 from fastapi import FastAPI, Query, status
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import JSONResponse
+from fastapi_cache import FastAPICache
+from fastapi_cache.backends.inmemory import InMemoryBackend
+from fastapi_cache.decorator import cache
+
+from .log import get_logger
 
 origins = ['*']
 
-app = FastAPI()
+logger = get_logger()
+
+
+@asynccontextmanager
+async def lifespan_event(app: FastAPI):
+    """
+    Context manager that yields the application's startup and shutdown events.
+    """
+    logger.info('⏱️ Application startup...')
+
+    worker_num = int(os.environ.get('APP_WORKER_ID', 9999))
+
+    logger.info(f'👷 Worker num: {worker_num}')
+
+    # set up cache
+    logger.info('🔥 Setting up cache...')
+    expiration = int(60 * 60 * 1)  # 24 hours
+    cache_status_header = 'x-html-reprs-cache'
+    FastAPICache.init(
+        InMemoryBackend(),
+        expire=expiration,
+        cache_status_header=cache_status_header,
+    )
+    logger.info(
+        f'🔥 Cache set up with expiration={expiration:,} seconds | {cache_status_header} cache status header.'
+    )
+
+    yield
+
+    logger.info('Application shutdown...')
+    logger.info('Clearing cache...')
+    FastAPICache.reset()
+    logger.info('👋 Goodbye!')
+
+
+app = FastAPI(lifespan=lifespan_event)
 app.add_middleware(
     CORSMiddleware,
     allow_origins=origins,
@@ -23,34 +67,60 @@ def index():
 
 
 @app.get('/xarray/')
+@cache(namespace='html-reprs')
 def xarray(
     url: pydantic.AnyUrl = Query(
         ...,
         description='URL to a zarr store',
         example='https://ncsa.osn.xsede.org/Pangeo/pangeo-forge/HadISST-feedstock/hadisst.zarr',
     ),
 ):
+    logger.info(f'🚀 Starting to process request for URL: {url}')
+
+    import time
+
     import xarray as xr
     import zarr
 
-    error_message = f'An error occurred while fetching the data from URL: {url}'
+    start_time = time.time()
+
+    error_message = f'❌ An error occurred while fetching the data from URL: {url}'
 
     try:
+        logger.info(f'📂 Attempting to open dataset from URL: {url}')
         with xr.open_dataset(url.unicode_string(), engine='zarr', chunks={}) as ds:
+            logger.info('✅ Successfully opened dataset. Generating HTML representation.')
             html = ds._repr_html_().strip()
 
         del ds
+        logger.info('🧹 Cleaned up dataset object')
+
+        end_time = time.time()
+        processing_time = end_time - start_time
+        logger.info(f'🏁 Request processed successfully in {processing_time:.2f} seconds')
 
         return {'html': html, 'dataset': url}
 
     except (zarr.errors.GroupNotFoundError, FileNotFoundError):
+        message = traceback.format_exc()
+        logger.error(f'🔍 {error_message}: Dataset not found\n{message}')
         return JSONResponse(
             status_code=status.HTTP_404_NOT_FOUND,
-            content={'detail': f'{error_message}. Dataset not found.'},
+            content={'detail': f'{error_message} Dataset not found. 😕'},
         )
 
     except PermissionError:
+        message = traceback.format_exc()
+        logger.error(f'🔒 {error_message}: Permission denied\n{message}')
         return JSONResponse(
             status_code=status.HTTP_403_FORBIDDEN,
-            content={'detail': f'{error_message}. Permission denied.'},
+            content={'detail': f'{error_message} Permission denied. 🚫'},
+        )
+
+    except Exception as e:
+        message = traceback.format_exc()
+        logger.error(f'💥 {error_message}: Unexpected error\n{message}')
+        return JSONResponse(
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+            content={'detail': f'{error_message} {str(e)} 😱'},
         )

src/log.py

@@ -0,0 +1,21 @@
+import logging
+import os
+import sys
+
+
+def get_logger() -> logging.Logger:
+    logger = logging.getLogger('html-reprs')
+    worker_id = os.environ.get('APP_WORKER_ID', '')
+
+    if not logger.handlers:
+        handler = logging.StreamHandler(stream=sys.stdout)
+        if worker_id != '':
+            handler.setFormatter(
+                logging.Formatter(f'[%(name)s] [worker {worker_id}] [%(levelname)s] %(message)s')
+            )
+        else:
+            handler.setFormatter(logging.Formatter('[%(name)s] [%(levelname)s] %(message)s'))
+        logger.addHandler(handler)
+
+    logger.setLevel(logging.DEBUG)
+    return logger