Spaces:
Sleeping
Sleeping
Update src/documentProcessing.py
Browse files- src/documentProcessing.py +26 -12
src/documentProcessing.py
CHANGED
|
@@ -36,11 +36,27 @@ def extract_text_from_txt(txt_path):
|
|
| 36 |
|
| 37 |
def process_document(document_path, gemini_api_key, language, content_type):
|
| 38 |
try:
|
| 39 |
-
|
| 40 |
-
with open(temp_file, 'wb') as f:
|
| 41 |
-
f.write(document_path.read())
|
| 42 |
-
|
| 43 |
file_extension = os.path.splitext(document_path.name)[-1].lower()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
if file_extension == '.pdf':
|
| 45 |
text = extract_text_from_pdf(temp_file)
|
| 46 |
elif file_extension == '.docx':
|
|
@@ -49,19 +65,17 @@ def process_document(document_path, gemini_api_key, language, content_type):
|
|
| 49 |
text = extract_text_from_txt(temp_file)
|
| 50 |
else:
|
| 51 |
raise Exception(f"Unsupported file type: {file_extension}")
|
| 52 |
-
|
| 53 |
text_file_path = tempfile.mktemp(suffix='.txt')
|
| 54 |
with open(text_file_path, 'w', encoding='utf-8') as f:
|
| 55 |
f.write(text)
|
| 56 |
-
|
|
|
|
| 57 |
formatted_output, json_path, txt_path = analyze_document(
|
| 58 |
-
text,
|
| 59 |
-
gemini_api_key,
|
| 60 |
-
language,
|
| 61 |
-
content_type
|
| 62 |
)
|
| 63 |
-
|
| 64 |
return f"Document processed successfully", text_file_path, formatted_output, txt_path, json_path
|
| 65 |
except Exception as e:
|
| 66 |
error_message = f"Error processing document: {str(e)}"
|
| 67 |
-
return error_message, None, error_message, None, None
|
|
|
|
| 36 |
|
| 37 |
def process_document(document_path, gemini_api_key, language, content_type):
|
| 38 |
try:
|
| 39 |
+
# Create a temporary file
|
|
|
|
|
|
|
|
|
|
| 40 |
file_extension = os.path.splitext(document_path.name)[-1].lower()
|
| 41 |
+
temp_file = tempfile.mktemp(suffix=file_extension)
|
| 42 |
+
|
| 43 |
+
# Handle different file-like objects
|
| 44 |
+
if hasattr(document_path, 'read'):
|
| 45 |
+
# If it's a file-like object with read method
|
| 46 |
+
with open(temp_file, 'wb') as f:
|
| 47 |
+
f.write(document_path.read())
|
| 48 |
+
elif hasattr(document_path, 'file'):
|
| 49 |
+
# If it's a Django or similar web framework file upload
|
| 50 |
+
with open(temp_file, 'wb') as f:
|
| 51 |
+
for chunk in document_path.file.chunks():
|
| 52 |
+
f.write(chunk)
|
| 53 |
+
elif isinstance(document_path, str):
|
| 54 |
+
# If it's a file path string
|
| 55 |
+
temp_file = document_path
|
| 56 |
+
else:
|
| 57 |
+
raise Exception("Unsupported document_path type")
|
| 58 |
+
|
| 59 |
+
# Process based on file type
|
| 60 |
if file_extension == '.pdf':
|
| 61 |
text = extract_text_from_pdf(temp_file)
|
| 62 |
elif file_extension == '.docx':
|
|
|
|
| 65 |
text = extract_text_from_txt(temp_file)
|
| 66 |
else:
|
| 67 |
raise Exception(f"Unsupported file type: {file_extension}")
|
| 68 |
+
|
| 69 |
text_file_path = tempfile.mktemp(suffix='.txt')
|
| 70 |
with open(text_file_path, 'w', encoding='utf-8') as f:
|
| 71 |
f.write(text)
|
| 72 |
+
|
| 73 |
+
# Assume this function is defined elsewhere
|
| 74 |
formatted_output, json_path, txt_path = analyze_document(
|
| 75 |
+
text, gemini_api_key, language, content_type
|
|
|
|
|
|
|
|
|
|
| 76 |
)
|
| 77 |
+
|
| 78 |
return f"Document processed successfully", text_file_path, formatted_output, txt_path, json_path
|
| 79 |
except Exception as e:
|
| 80 |
error_message = f"Error processing document: {str(e)}"
|
| 81 |
+
return error_message, None, error_message, None, None
|