jan.mrna/semantic_doc_search
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Serve by file index, not full path

Browse Source
This commit is contained in:
Jan Mrna
2025-11-06 10:45:42 +01:00
parent e624db1ce7
commit 788eebc916
3 changed files with 29 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
main.py
View File

@@ -166,7 +166,7 @@ def query(db_path: str, query_text: str):
for i, res in enumerate(results, 1):
print(f"\n{i}. Distance: {res.distance:.4f}")
print(f" Document: {res.document.name}")
print(f" Document: {res.document_name}")
print(f" Page: {res.record.page}, Chunk: {res.record.chunk}")
# Replace all whitespace characters with regular spaces for cleaner display
clean_text = ' '.join(res.record.text[:200].split())
@@ -201,18 +201,13 @@ def start_web_server(db_path: str, host: str = "127.0.0.1", port: int = 5000):
def index():
return render_template("index.html", results=None)
@app.route('/file/<path:document_path>')
def serve_file(document_path):
@app.route('/file/<int:document_index>')
def serve_file(document_index):
"""Serve PDF files directly."""
try:
file_path = Path(document_path)
file_path = db.get_document_path(db_file, document_index)
if not file_path.exists():
return jsonify({'error': 'File not found'}), 404
# Check if it's a PDF file for security
if file_path.suffix.lower() != '.pdf':
return jsonify({'error': 'Only PDF files are allowed'}), 403
return send_file(file_path, as_attachment=False)
except Exception as e:
return jsonify({'error': str(e)}), 500
@@ -236,13 +231,12 @@ def start_web_server(db_path: str, host: str = "127.0.0.1", port: int = 5000):
for res in results:
formatted_results.append({
'distance': float(res.distance),
'document': res.document.name,
'document_path': str(res.document), # Full path for the link
'document_name': res.document_name,
'document_index': res.record.document_index,
'page': res.record.page,
'chunk': res.record.chunk,
'text': ' '.join(res.record.text[:300].split()) # Clean and truncate text
})
return jsonify({'results': formatted_results})
except Exception as e: