Semantic search is a powerful feature in Junie that enhances code search functionality by leveraging code embeddings created from your project. Below are some frequently asked questions to help you better understand how it works.
Where does the semantic search store the indexes?
The indexes are stored on JetBrains servers. It is important to note that the code itself is not stored — only the embeddings are. All operations with the code occur on the user’s machine.
Why was it decided to store the indexes remotely?
Storing the indexes remotely allows for saving the tokens by indexing open-source repositories and makes it possible to share indexes within organizations.
What is currently indexed?
The content of the git repository is indexed. Files listed under .gitignore are not included in the indexing process.
How often does Junie perform the indexing?
Currently, indexing is performed only once, and only when the user explicitly agrees to it. An update mechanism to perform indexing every day is currently being developed.
Can the results of a semantic search be outdated?
Yes, the results of semantic search can become outdated. If this happens, the regular search feature will also be used in addition to the semantic search to ensure accurate results.
How do I disable the semantic search?
You can disable semantic search via Junie settings. Navigate to Settings | Tools | Junie | Project Settings and uncheck the Allow Semantic Indexing option. After doing so, the regular search functionality will replace semantic search.
Comments
0 comments
Please sign in to leave a comment.