John C Fitzpatrick 85bd6bc012 Fix multi-dimensional vector hybrid search functions (#681) ...

Fixes critical bug where hybrid search functions referenced non-existent
cp.embedding and ce.embedding columns instead of dimension-specific columns.

Changes:
- Add new multi-dimensional hybrid search functions with dynamic column selection
- Maintain backward compatibility with existing legacy functions
- Support all embedding dimensions: 384, 768, 1024, 1536, 3072
- Proper error handling for unsupported dimensions

Resolves: #675 - RAG queries now work with multi-dimensional embeddings

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-authored-by: Claude <noreply@anthropic.com>

2025-09-18 10:06:10 -07:00

..

add_hybrid_search_tsvector.sql

Fix multi-dimensional vector hybrid search functions (#681)

2025-09-18 10:06:10 -07:00

add_priority_column_to_tasks.sql

feat: decouple task priority from task order (#652)

2025-09-17 13:44:25 +03:00

add_source_url_display_name.sql

Fix race condition in concurrent crawling with unique source IDs (#472)

2025-08-29 14:54:16 +03:00

backup_database.sql

feat: Ollama Integration with Separate LLM/Embedding Model Support (#643)

2025-09-15 06:38:02 -07:00

complete_setup.sql

Fix multi-dimensional vector hybrid search functions (#681)

2025-09-18 10:06:10 -07:00

DB_UPGRADE_INSTRUCTIONS.md

feat: Ollama Integration with Separate LLM/Embedding Model Support (#643)

2025-09-15 06:38:02 -07:00

RESET_DB.sql

Enhanced the hybrid search strategy with tsvector keyword matching (#539)

2025-09-10 14:23:49 +03:00

upgrade_database.sql

feat: Ollama Integration with Separate LLM/Embedding Model Support (#643)

2025-09-15 06:38:02 -07:00

validate_migration.sql

feat: Ollama Integration with Separate LLM/Embedding Model Support (#643)

2025-09-15 06:38:02 -07:00