improved templating for extracted text

emcf · Apr 20, 2024 · 2edefca · 2edefca
1 parent 46274df
commit 2edefca
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 2 deletions.
diff --git a/setup.py b/setup.py
@@ -2,7 +2,7 @@
 
 setup(
     name='thepipe_api',
-    version='0.2.7',
+    version='0.2.8',
     author='Emmett McFarlane',
     author_email='[email protected]',
     description='Automate information extraction for multimodal LLMs.',

diff --git a/thepipe_api/thepipe.py b/thepipe_api/thepipe.py
@@ -14,8 +14,10 @@ def save_outputs(chunks: List[Chunk], verbose: bool = False, text_only: bool = F
     for i, chunk in enumerate(chunks):
         if chunk is None:
             continue
+        if chunk.path is not None:
+            text += f'{chunk.path}:\n'
         if chunk.text is not None:
-            text += chunk.text
+            text += f'```\n{chunk.text}```\n\n'
         if (chunk.image is not None) and (not text_only):
             if chunk.path is None:
                 clean_path = f"image"