Code

History

Commits on fix-git-https-support to crates/semantic_index210+
  1. update request timeout for open ai embeddingsKCaverly committed 3 years ago265d02a
  2. update token_count for OpenAIEmbeddings to accomodate for truncationKCaverly committed 3 years ago17237f7
  3. Rename `Document` to `Span`Antonio Scandurra committed 3 years agoce62173
  4. Ensure `SemanticIndex::search` waits for indexing to completeAntonio Scandurra committed 3 years agode0f53b
  5. Clip ranges returned by `SemanticIndex::search`Antonio Scandurra committed 3 years agoc802680
  6. Re-index project when a worktree is registeredAntonio Scandurra committed 3 years ago95b72a7
  7. Simplify `SemanticIndex::index_project`Antonio Scandurra committed 3 years ago3c70b12
  8. Retrieve embeddings based on pending filesAntonio Scandurra committed 3 years ago6b1dc63
  9. Move retrieval of embeddings from the db into `reindex_changed_files`Antonio Scandurra committed 3 years ago7b5a41d
  10. :art:Antonio Scandurra committed 3 years agod4cff68
  11. update pragma settings for improved database performanceKCaverly committed 3 years ago8dbc0fe
  12. updated embeddings background delay to 5 minutesKCaverly committed 3 years ago54235f4
  13. optimize insert file in vector databaseKCaverly committed 3 years agoe86964e
  14. flush embeddings queue when no files are parsed for 250 millisecondsKCaverly committed 3 years ago524533c
  15. move embeddings queue to use single hashmap for all changed pathsKCaverly committed 3 years agoc4db914
  16. WIP: work towards wiring up a embeddings_for_digest hashmap that is stored for all indexed filesKCaverly committed 3 years agoafa59ab
  17. fill embeddings with database values and skip during embeddings queueKCaverly committed 3 years ago50cfb06
  18. WIPAntonio Scandurra committed 3 years ago220533f
  19. Rename `Sha1` to `DocumentDigest`Antonio Scandurra committed 3 years ago2503d54
  20. Reify `Embedding`/`Sha1` structs that can be (de)serialized from SQLAntonio Scandurra committed 3 years ago3001a46
  21. Write to and read from the database in a transactional wayAntonio Scandurra committed 3 years agoc763e72
  22. Abstract away how database transactions are executedAntonio Scandurra committed 3 years ago35440be
  23. fix bug for truncation ensuring no valid inputs are sent to openaiKCaverly committed 3 years ago7d4d6c8
  24. moved semantic index to use embeddings queue to batch and managed for atomic database writesKCaverly committed 3 years ago5abad58
  25. move queuing to embedding_queue functionality and update embedding provider to include trait items for max tokens per batch"KCaverly committed 3 years ago76ce52d
  26. move truncation to parsing step leveraging the EmbeddingProvider traitKCaverly committed 3 years ago9781047
  27. add should_truncate to embedding providersKCaverly committed 3 years ago76caea8
  28. added token count to documents during parsingKCaverly committed 3 years agoe377ada
  29. reindex files in the background after they have not been edited for 10 minutesKCaverly committed 3 years agoa7e6a65
  30. add proper handling for open ai rate limit delaysKCaverly committed 3 years ago4f8b95c
  31. Format let-else statementsMax Brunsfeld committed 3 years ago404f767
  32. cleaned up warningsKCaverly committed 3 years agoee97bc5
  33. add handling for Added file events to semantic indexKCaverly committed 3 years ago131950f
  34. update initialize project call to accomodate for test scenariosKCaverly committed 3 years agoa892a51
  35. move semantic search project intialization to a subscribe event for workspace createdKCaverly committed 3 years agoa1519e4