Commit graph

21 commits

Author SHA1 Message Date
Nicolò Boschi
7accac94b2
fix: migrate mental_models.embedding dimension alongside memory_units (#526)
ensure_embedding_dimension() now also checks and migrates mental_models.embedding,
fixing silent failures when changing embedding model dimensions. Extracted shared
per-table logic into _migrate_table_embedding_dimension() to avoid duplication.
Adds test coverage for the mental_models dimension migration path.

Fixes #523
2026-03-09 12:23:50 +01:00
Chris Bartholomew
f88b50a45e
fix: serialize alembic upgrades in-process (#521) 2026-03-09 11:22:24 +01:00
Nicolò Boschi
7942f181c2
fix(performance): improve recall and retain performance on large banks (#469) 2026-03-03 13:35:22 +01:00
Anton Evseev
8cd65b9896
fix: raise error when embedding dimensions exceed pgvector HNSW limit (#361)
Instead of silently skipping HNSW index creation for embeddings > 2000
dimensions, raise a RuntimeError with an actionable message suggesting
pgvectorscale/DiskANN as an alternative.

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 10:24:17 +01:00
Nicolò Boschi
476726c2a2
feat: support azure pg_diskann (#381) 2026-02-16 16:08:54 +01:00
Nicolò Boschi
95c4220477
feat: support for pgvectorscale (DiskANN) (#378)
* feat: support for pgvectorscale (DiskANN)

* feat: support for pgvectorscale (DiskANN)
2026-02-16 14:19:56 +01:00
Nicolò Boschi
d871c3009d
feat: support timescale pg_textsearch as text search extension (#359)
* feat: support timescale pg_textsearch as text search extension

* refactor: deduplicate text search query in retrieve_semantic_bm25_combined

Instead of maintaining 3 complete query copies (native, vchord, pg_textsearch),
now we:
- Build backend-specific parts (score_expr, order_by, where_filter)
- Use a single query template with injected backend-specific parts

This makes maintenance easier - changes to the semantic CTE or overall structure
only need to be made once.
2026-02-12 17:38:13 +01:00
Nicolò Boschi
c029807add
feat: support for other text and vector search pg extensions (#355)
* feat: support for other text and vector search pg extensions

* test: increase timeout for test_batch_chunking_behavior to account for VectorChord BM25 tokenization overhead

* feat: support for other text and vector search pg extensions
2026-02-12 14:13:04 +01:00
Damien
a713b68b1f
Implement the vchord / pgvector support (#350)
* Implement the vchord / pgvector support

* feat(alembic): detect vector extension and create appropriate index
2026-02-12 10:47:07 +01:00
Nicolò Boschi
c2607d7699
fix(helm): improve appVersion usage (#326) 2026-02-09 11:35:08 +01:00
Van Vuong Ngo
c568094b8c
fix: do not log db user/password (#312)
* fix: security vulnerability - exposed sensitve database credentials in logs

* add comment

* fix: mask credentials of the postgeSQL connection string
2026-02-09 10:15:03 +01:00
Nicolò Boschi
382550690a
fix: custom pg schema is not reliable (#278)
* fix: custom pg schema is not reliable

* fix

* fix

* fix: WorkerPoller now always has tenant extension

Ensures WorkerPoller follows same pattern as MemoryEngine - always
creates a DefaultTenantExtension if none is provided, preventing
NoneType errors when calling list_tenants().

Fixes test failures in test_worker.py

* fix: DefaultTenantExtension honors explicit schema parameter

Allows WorkerPoller's schema parameter to be passed through to
DefaultTenantExtension via config dict, maintaining backward
compatibility for tests that use schema parameter without
providing a tenant extension.

Fixes test_poller_with_custom_schema test failure.
2026-02-02 15:33:50 +01:00
Nicolò Boschi
59913086be
fix: batch queries on recall (#149)
* fix: batch queries on recall

* fix: batch queries on recall
2026-01-13 13:20:22 +01:00
Nicolò Boschi
70de23ed85
feat: configurable embedding dimensions + OpenAI Embeddings (#101)
* feat: configurable embedding dimensions + OpenAI Embeddings

* fix tests
2026-01-05 14:43:05 +01:00
Nicolò Boschi
76fd052b3a
misc: add mcp integration tests and increase test coverage (#98)
* misc: add mcp integration tests and increase test coverage

* misc: add mcp integration tests and increase test coverage

* misc: add mcp integration tests and increase test coverage
2026-01-05 11:16:55 +01:00
Chris Latimer
1a620697b1
Feature/graph viz (#85)
* Improve graph visualization on the UI

* Fix double animation when loading the graph visualization

* Fix typescript issues

* CI test changes for temporal scenarios

* Fix typescript errors

* Fix animation issue on opinions and experiences
2026-01-02 16:27:29 +01:00
Nicolò Boschi
2a0c490c9e
feat: extensions (#54) 2025-12-22 11:05:23 +01:00
Nicolò Boschi
9394cf92f2
fix: doc build and lint files (#34)
* fix doc build

* fix doc build
2025-12-16 13:49:09 +01:00
Nicolò Boschi
f42476bf94
fix: make sure openai provider works + docs updates (#23)
* fix: make sure openai provider works

* fix: make sure openai provider works

* fix
2025-12-10 16:10:10 +01:00
Nicolò Boschi
040eb33ea6
improve ci and tests (#22)
* improve ci and tests

* add more tests

* fixes

* fix tests

* fix more

* fix

* fix

* fix

* fix

* fix for real

* tests and doc

* fix cp

* fix link pg0

* fix pg0

* fix pg0

* fix pg0

* even better

* more

* fix
2025-12-09 19:16:00 +01:00
Nicolò Boschi
b1ff2e8823
rename to hindsight (#2) 2025-11-25 19:28:26 +01:00
Renamed from memora/memora/migrations.py (Browse further)