Skip to main content
Glama

MCP Codebase Insight

by tosin2013
# MCP Codebase Insight Test Fixes ## Identified Issues 1. **Package Import Problems** - The tests were trying to import from `mcp_codebase_insight` directly, but the package needed to be imported from `src.mcp_codebase_insight` - The Python path wasn't correctly set up to include the project root directory 2. **Missing Dependencies** - The `sentence-transformers` package was installed in the wrong Python environment (Python 3.13 instead of 3.11) - Had to explicitly install it in the correct environment 3. **Test Isolation Problems** - Tests were failing due to not being properly isolated - The `component_test_runner.py` script needed fixes to properly load test modules 4. **Qdrant Server Issue** - The `test_vector_store_cleanup` test failed due to permission issues in the Qdrant server - The server couldn't create a collection directory for the test ## Applied Fixes 1. **Fixed Import Paths** - Modified test files to use `from src.mcp_codebase_insight...` instead of `from mcp_codebase_insight...` - Added code to explicitly set `sys.path` to include the project root directory 2. **Fixed Dependency Issues** - Ran `python3.11 -m pip install sentence-transformers` to install the package in the correct environment - Verified all dependencies were properly installed 3. **Created a Test Runner Script** - Created `run_test_with_path_fix.sh` to set up the proper environment variables and paths - Modified `component_test_runner.py` to better handle module loading 4. **Fixed Test Module Loading** - Added a `load_test_module` function to properly handle import paths - Ensured the correct Python path is set before importing test modules ## Results - Successfully ran 2 out of 3 vector store tests: - ✅ `test_vector_store_initialization` - ✅ `test_vector_store_add_and_search` - ❌ `test_vector_store_cleanup` (still failing due to Qdrant server issue) ## Recommendations for Remaining Issue The `test_vector_store_cleanup` test is failing due to the Qdrant server not being able to create a directory for the collection. This could be fixed by: 1. Checking the Qdrant server configuration to ensure it has proper permissions to create directories 2. Creating the necessary directories beforehand 3. Modifying the test to use a collection name that already exists or mock the collection creation The error message suggests a file system permission issue: ``` "Can't create directory for collection cleanup_test_db679546. Error: No such file or directory (os error 2)" ``` A simpler fix for testing purposes might be to modify the Qdrant Docker run command to include a volume mount with proper permissions: ```bash docker run -d -p 6333:6333 -p 6334:6334 -v $(pwd)/qdrant_data:/qdrant/storage qdrant/qdrant ``` This would ensure the storage directory exists and has the right permissions.

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/tosin2013/mcp-codebase-insight'

If you have feedback or need assistance with the MCP directory API, please join our Discord server