From f00d7cbb81c2be6de9df7b46619f18001d77f297 Mon Sep 17 00:00:00 2001
From: Emmett McFaralne <staminacode@gmail.com>
Date: Fri, 26 Apr 2024 22:15:33 -0400
Subject: [PATCH] fixed json datetime encoding

---
 setup.py                 |  2 +-
 thepipe_api/extractor.py | 10 +++++++++-
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/setup.py b/setup.py
index 87ef909..9bec88e 100644
--- a/setup.py
+++ b/setup.py
@@ -2,7 +2,7 @@
 
 setup(
     name='thepipe_api',
-    version='0.3.1',
+    version='0.3.2',
     author='Emmett McFarlane',
     author_email='emmett@thepi.pe',
     description='Automate information extraction for multimodal LLMs.',
diff --git a/thepipe_api/extractor.py b/thepipe_api/extractor.py
index 2dae60e..d88128d 100644
--- a/thepipe_api/extractor.py
+++ b/thepipe_api/extractor.py
@@ -15,8 +15,16 @@
 import tempfile
 import mimetypes
 import dotenv
+from datetime import datetime
+from json import JSONEncoder
 dotenv.load_dotenv()
 
+class JSONDateEncoder(JSONEncoder):
+    def default(self, obj):
+        if isinstance(obj, datetime):
+            return obj.isoformat()
+        return JSONEncoder.default(self, obj)
+
 FILES_TO_IGNORE = {'package-lock.json', '.gitignore', '.bin', '.pyc', '.pyo', '.exe', '.bat', '.dll', '.obj', '.o', '.a', '.lib', '.so', '.dylib', '.ncb', '.sdf', '.suo', '.pdb', '.idb', '.pyd', '.ipynb_checkpoints', '.npy', '.pth'} # Files to ignore, please feel free to customize!
 CODE_EXTENSIONS = {'.h', '.json', '.js', '.jsx', '.ts', '.tsx',  '.cs', '.java', '.html', '.css', '.ini', '.xml', '.yaml', '.xaml', '.sh'} # Plaintext files that should not be compressed with LLMLingua
 CTAGS_CODE_EXTENSIONS = {'.c', '.cpp', '.py'} # code files that work with ctags
@@ -244,7 +252,7 @@ def extract_spreadsheet(file_path: str) -> Chunk:
     elif file_path.endswith(".xls") or file_path.endswith(".xlsx"):
         df = pd.read_excel(file_path)
     dict = df.to_dict(orient='records')
-    json_dict = json.dumps(dict, indent=4)
+    json_dict = json.dumps(dict, indent=4, cls=JSONDateEncoder)
     return Chunk(path=file_path, text=json_dict, image=None, source_type=SourceTypes.SPREADSHEET)
     
 def extract_url(url: str, text_only: bool = False, local: bool = True, limit: int = None) -> List[Chunk]: