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

DB Record: use document index instead of Path

Browse Source
This commit is contained in:
Jan Mrna
2025-11-05 14:31:36 +01:00
parent c0c80142be
commit 9bc39ccea8
2 changed files with 80 additions and 108 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
main.py
View File

@@ -164,12 +164,12 @@ def query(db_path: str, query_text: str):
print(f"\nFound {len(results)} results:")
print("=" * 60)
for i, (distance, record) in enumerate(results, 1):
print(f"\n{i}. Distance: {distance:.4f}")
print(f" Document: {record.document.name}")
print(f" Page: {record.page}, Chunk: {record.chunk}")
for i, res in enumerate(results, 1):
print(f"\n{i}. Distance: {res.distance:.4f}")
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(record.text[:200].split())
clean_text = ' '.join(res.record.text[:200].split())
print(f" Text preview: {clean_text}...")
if i < len(results):
print("-" * 40)
@@ -233,14 +233,14 @@ def start_web_server(db_path: str, host: str = "127.0.0.1", port: int = 5000):
# Format results for JSON response
formatted_results = []
for distance, record in results:
for res in results:
formatted_results.append({
'distance': float(distance),
'document': record.document.name,
'document_path': str(record.document), # Full path for the link
'page': record.page,
'chunk': record.chunk,
'text': ' '.join(record.text[:300].split()) # Clean and truncate text
'distance': float(res.distance),
'document': res.document.name,
'document_path': str(res.document), # Full path for the link
'page': res.record.page,
'chunk': res.record.chunk,
'text': ' '.join(res.record.text[:300].split()) # Clean and truncate text
})
return jsonify({'results': formatted_results})