* Fix: Remove all nodes for multi-chunk documents in update_llm_index incremental path
The existing_nodes dict comprehension keyed on document_id silently dropped all
but the last node per document, so only that one node was deleted when a
modified document was re-indexed, leaving all other chunks as ghost vectors in
the FAISS index. Switch to a defaultdict(list) that collects every node per
document_id, then iterate and delete all of them before inserting fresh nodes.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix: Wire document_updated signal to LLM index update handler
Connect document_updated to add_or_update_document_in_llm_index in
DocumentsConfig.ready() so REST API edits (PATCH /api/documents/{id}/)
enqueue an LLM vector store update, matching the existing
document_consumption_finished behavior.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix: Add file lock around FAISS index mutations to prevent concurrent write corruption
Two concurrent Celery workers calling llm_index_add_or_update_document or
llm_index_remove_document each loaded the same on-disk index independently,
made their own change, and the last writer silently overwrote the first's
update. Wrap both functions and the rebuild/persist body of update_llm_index
in a filelock.FileLock keyed on LLM_INDEX_DIR/index.lock. Add a TOCTOU
comment on queue_llm_index_update_if_needed explaining the residual risk
(duplicate rebuild tasks are wasteful but not corrupting because the lock
serialises the actual write).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix: Apply _normalize() in extract_unmatched_names to prevent duplicate suggestions
extract_unmatched_names was using .lower() while _match_names_to_queryset
uses _normalize() (which also strips punctuation). A name like "J. Smith"
matched to existing correspondent "J Smith" would still appear in the
unmatched list, causing duplicate object creation.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix: Skip LLM index update gracefully when document has no indexable content
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix: Persist empty index when all documents are deleted to clear stale FAISS vectors
The early-return guard in update_llm_index fired before persist() when no
documents existed, leaving a stale on-disk FAISS index that returned phantom
hits for deleted document IDs. Now the guard only returns early for the
incremental (rebuild=False) path when no index exists on disk; the rebuild
path always continues through to persist(), producing an empty clean index.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Chore: Simplify incremental index update — use docs.values() and deduplicate node extend
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix: Validate and limit chat question input in ChatStreamingView
Add max_length=4000 to ChatStreamingSerializer.q and replace the bare
request.data["q"] read with proper serializer.is_valid(raise_exception=True)
so oversized or missing questions are rejected with HTTP 400 before
reaching the LLM.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix: Add defensive prompt framing to mark document content as untrusted
* Also adds a system prompt which is treated higher that this is untrusted stuff
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Implements and tests a retry with backoff + jitter for aquring the index update lock. If we still can't get it, dispatch a celery task to handle it later instead (also with retry)
Signed-off-by: stumpylog <797416+stumpylog@users.noreply.github.com>
* Replaces loaddata with streaming bulk_create
Replaces call_command('loaddata') with a streaming implementation that
reads manifest records one at a time via ijson, accumulates per-model
batches up to --batch-size, and flushes via bulk_create. This reduces
peak memory and no longer scales directly with the size of the import.
* fix(importer): avoid guardian lru_cache poisoning; include M2M through tables in check_constraints
clear_cache() inside the import transaction emptied Django's ContentType
manager cache while fixture PKs were live, causing downstream ContentType
lookups to repopulate guardian's separate @lru_cache(None) with
fixture-PK objects. After the TestCase transaction rolled back to
original PKs, guardian's lru_cache held stale fixture ContentType
objects, causing MixedContentTypeError in unrelated subsequent tests.
Remove clear_cache() since it was defending against a theoretical
stale-cache scenario that doesn't occur in a proper same-install restore.
Fix check_constraints() to explicitly include auto-created M2M through
tables (populated by .set() after bulk_create) alongside the model tables,
addressing the gap where join-table FK violations would have gone
undetected.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Excludes the consumer and AnonymousUser from any models which might have a FK relation to it. This prevents orphan things like UI setting, which have a relation to no existing user
* Splits into more sub functions for Sonar
* Improvements to the typing of the new functions
* Coverage for some error cases, and removes handling for pk only models. No need to support these
* Final coverage gaps
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* Updates code to use a FileResponse for streaming and unlink the file, but keep a handle to it
* Transitions the rest of the code to use FileResponse instead of a basic response, fixes up tests which assumed .content exists
* While here, let's add schema for it
shamoon
Trenton H
Matt Van Horn
Gaëtan GOUZI