ignore all output types correctly and write to a folder instead of the git root

Signed-off-by: John Seekins <[email protected]>

Author: johnseekins

.gitignore

@@ -1,5 +1,11 @@
 __pycache__/
 *.csv
+**/.csv
 *.json
+**/*.json
 *.xlsx
+**/.xlsx
 *.xlsx#
+**/.xlsx#
+*.parquet
+**/.parquet

file_utils.py

@@ -1,9 +1,11 @@
 import copy
 import json
+import os
 from schemas import enrichment_print_schema
 from utils import (
     convert_to_dataframe,
     logger,
+    output_folder,
 )
 import xlsxwriter  # type: ignore [import-untyped]
 
@@ -16,8 +18,9 @@ def export_to_file(
     if not facilities_data or not facilities_data.get("facilities", []):
         logger.warning("No data to export!")
         return ""
-
-    full_name = f"{filename}.{file_type}"
+    # make sure the folder we're dropping files into exists
+    os.makedirs(output_folder, exist_ok=True)
+    full_name = f"{output_folder}/{filename}.{file_type}"
     if file_type in ["csv", "xlsx", "parquet"]:
         writer = convert_to_dataframe(facilities_data["facilities"])
         match file_type:
@@ -36,10 +39,9 @@ def export_to_file(
             json.dump(facilities_data, f_out, indent=2, sort_keys=True, default=str)
 
     logger.info(
-        "%s file '%s.%s' created successfully with %s facilities.",
-        file_type,
-        filename,
+        "%s file '%s' created successfully with %s facilities.",
         file_type,
+        full_name,
         len(facilities_data["facilities"]),
     )
     return filename

utils.py

@@ -1,10 +1,12 @@
 # For general helpers, regexes, or shared logic (e.g. phone/address parsing functions).
 import logging
+import os
 import polars
 import requests
 from requests.adapters import HTTPAdapter
 import urllib3
 
+SCRIPTDIR = os.path.dirname(os.path.realpath(__file__))
 logger = logging.getLogger(__name__)
 logger.setLevel(logging.INFO)
 logger.addHandler(logging.StreamHandler())
@@ -20,6 +22,7 @@
 session.mount("http://", _adapter)
 session.headers.update(default_headers)
 
+output_folder = f"{SCRIPTDIR}/output/"
 default_timestamp = "1970-01-01T00:00:00-+0000"
 timestamp_format = "%Y-%m-%dT%H:%M:%S-%z"