cavepedia/cavepediav2
1
0
Fork 0
Code Issues 1 Pull Requests Actions Packages Projects 1 Releases Wiki Activity

skip content < 100 chars

Browse Source
This commit is contained in:
Paul Walko
2025-12-26 03:36:28 +01:00
parent 499a3cb170
commit b6f01f9b3f
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
mcp/server.py
View File

@@ -76,7 +76,7 @@ def search_caving_documents(query: str, priority_prefixes: list[str] | None = No
top_n = 2 top_n = 2
candidate_limit = top_n * 4 candidate_limit = top_n * 4
rows = conn.execute( rows = conn.execute(
'SELECT * FROM embeddings WHERE embedding IS NOT NULL AND role = ANY(%s) ORDER BY embedding <=> %s::vector LIMIT %s', 'SELECT * FROM embeddings WHERE embedding IS NOT NULL AND LENGTH(content) > 100 AND role = ANY(%s) ORDER BY embedding <=> %s::vector LIMIT %s',
(roles, query_embedding, candidate_limit) (roles, query_embedding, candidate_limit)
).fetchall() ).fetchall()
@@ -96,6 +96,7 @@ def search_caving_documents(query: str, priority_prefixes: list[str] | None = No
sources_only = is_sources_only() sources_only = is_sources_only()
for result in rerank_resp.results: for result in rerank_resp.results:
row = rows[result.index] row = rows[result.index]
content = row['content'] or ''
score = result.relevance_score score = result.relevance_score
# Boost score if key starts with any priority prefix (e.g., 'nss/aca') # Boost score if key starts with any priority prefix (e.g., 'nss/aca')
@@ -107,7 +108,6 @@ def search_caving_documents(query: str, priority_prefixes: list[str] | None = No
if sources_only: if sources_only:
docs.append({'key': row['key'], 'relevance': round(score, 3)}) docs.append({'key': row['key'], 'relevance': round(score, 3)})
else: else:
content = row['content'] or ''
docs.append({'key': row['key'], 'content': content, 'relevance': round(score, 3)}) docs.append({'key': row['key'], 'content': content, 'relevance': round(score, 3)})
# Re-sort by boosted score and return top_n # Re-sort by boosted score and return top_n