Clean up

server/.gitignore

@@ -1,5 +1,7 @@
 .env
 
+/api/data/*
+
 # Terraform stuff
 *.tfvars
 *.tfstate

server/api/code/lacity_data_api/routers/reports.py

@@ -5,7 +5,7 @@
 from fastapi import APIRouter, Query, BackgroundTasks
 from fastapi.responses import FileResponse
 
-from lacity_data_api.services.reports import run_report, make_csv_cache
+from lacity_data_api.services.reports import run_report, make_year_csv, make_csv
 from lacity_data_api.config import DATA_DIR
 
 router = APIRouter()
@@ -38,8 +38,30 @@ async def get_report(
     return results
 
 
-@router.get("/export/{year}/{file}.csv")
-async def get_csv(
+# TODO: Need to rethink. These don't work in production with our tiny CPUs.
+# Gunicorn kills the worker when it sees a CPU/Memory spike.
+@router.get("/export/service_requests.csv", description="EXPERIMENTAL")
+async def get_csv(background_tasks: BackgroundTasks):
+    csv_file = f"{DATA_DIR}/service_requests.csv"
+    if os.path.exists(csv_file):
+        return FileResponse(csv_file)
+    else:
+        background_tasks.add_task(make_csv)
+        return "Started file generation"
+
+
+@router.get("/export/service_requests.csv.gz", description="EXPERIMENTAL")
+async def get_gzip(background_tasks: BackgroundTasks):
+    gzip_file = f"{DATA_DIR}/service_requests.csv.gz"
+    if os.path.exists(gzip_file):
+        return FileResponse(gzip_file)
+    else:
+        background_tasks.add_task(make_csv)
+        return "Started file generation"
+
+
+@router.get("/export/{year}/{file}.csv", description="EXPERIMENTAL")
+async def get_year_csv(
     background_tasks: BackgroundTasks,
     file: str = "service_requests",
     year: int = 2020,
@@ -48,12 +70,12 @@ async def get_csv(
     if os.path.exists(csv_file):
         return FileResponse(csv_file)
     else:
-        background_tasks.add_task(make_csv_cache, file, year)
+        background_tasks.add_task(make_year_csv, file, year)
         return "Started file generation"
 
 
-@router.get("/export/{year}/{file}.csv.gz")
-async def get_gzip(
+@router.get("/export/{year}/{file}.csv.gz", description="EXPERIMENTAL")
+async def get_year_gzip(
     background_tasks: BackgroundTasks,
     file: str = "service_requests",
     year: int = 2020,
@@ -62,5 +84,5 @@ async def get_gzip(
     if os.path.exists(gzip_file):
         return FileResponse(gzip_file)
     else:
-        background_tasks.add_task(make_csv_cache, file, year)
+        background_tasks.add_task(make_year_csv, file, year)
         return "Started file generation"

server/api/code/lacity_data_api/services/reports.py

@@ -81,7 +81,7 @@ async def run_report(field, filter):
     return result
 
 
-async def make_csv_cache(table: str, year: int = 2020) -> None:
+async def make_year_csv(table: str, year: int = 2020) -> None:
 
     columns = [
         "request_id",
@@ -113,3 +113,38 @@ async def make_csv_cache(table: str, year: int = 2020) -> None:
     with open(f"{DATA_DIR}/{table}-{year}.csv", 'rb') as f_in:
         with gzip.open(f"{DATA_DIR}/{table}-{year}.csv.gz", 'wb') as f_out:
             shutil.copyfileobj(f_in, f_out)
+
+
+async def make_csv():
+    FILE = "service_requests"
+
+    columns = [
+        "request_id",
+        "srnumber",
+        "created_date",
+        "closed_date",
+        "type_id",
+        "agency_id",
+        "source_id",
+        "council_id",
+        "region_id",
+        "address",
+        "latitude",
+        "longitude",
+        "city_id"
+    ]
+
+    query = db.text(f"""
+        select * from {FILE};
+    """)
+
+    result = await db.all(query)
+
+    with open(f"{DATA_DIR}/{FILE}.csv", 'w', newline='') as f:
+        writer = csv.writer(f)
+        writer.writerow(columns)
+        writer.writerows(result)
+
+    with open(f"{DATA_DIR}/{FILE}.csv", 'rb') as f_in:
+        with gzip.open(f"{DATA_DIR}/{FILE}.csv.gz", 'wb') as f_out:
+            shutil.copyfileobj(f_in, f_out)