| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
Upgrade htrie. | 1 年前 | |
Remove line length restriction on INI reader. | 1 年前 | |
String sort: allow missing values to be sent to either end. | 4 年前 | |
update: query events to fully as search events (#2565) * update: query events to fully as search events * fix: ci errors * fix: ci | 11 个月前 | |
Parameterize configs used by remote embedding API. | 7 个月前 | |
add prefix auth_key in access logs (#2915) * add prefix auth_key in access logs * Log resolved API key prefixes for scoped and multi-search requests * add empty query check --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
add: personalization model APIs (#2001) * add: basic recommendations endpoint * add: POST `/recommendations/models` recommendations endpoint * add: get specific model endpoint * add: DELETE `/recommendations/models/:id` and GET all models recommendations endpoint * add: ONNX model upload endpoint * add: save the file as a .tar.gz instead of ONNX * add: verify_tar_gz function * add: support for model upload * add: PUT `/recommendations/model/:id` to update the model, name, collection * refactor: recommendations -> personalization * add: personalization tests * add: personalization_model_manager_tests * fix: review comments * fix: review comments | 1 年前 | |
Fixed a bug with hlighlighting in upsert import. | 5 年前 | |
Always use facet hash index. | 3 年前 | |
Add `ArrayUtils::skip_index_to_id`. | 3 年前 | |
num_tree iterator (#1538) * Add `num_tree_t::iterator_t`. * Add `num_tree_t::iterator_t` tests. * Add `bool_iterator` in `filter_result_iterator_t`. * Fix `filter_result_iterator_t::compute_iterators`. | 2 年前 | |
add prefix auth_key in access logs (#2915) * add prefix auth_key in access logs * Log resolved API key prefixes for scoped and multi-search requests * add empty query check --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
Enable proper stack trace with symbols. | 3 年前 | |
Join Import (#2868) * Serialize requests of related collections based on `is_complete` and `latest_chunk_log_index`. * Add `reference_cluster_follower_node_restart.test.ts`. * Review changes. * Remove log. * Remove log. * Increase timeout. | 4 个月前 | |
Fix health resource status logic. | 2 年前 | |
Fix access logging cmdline flag. | 4 年前 | |
Rebind reference field to the new swapped collection alias. (#2963) * Rebind reference field to the new swapped collection alias. * Fix a scenario where swapping alias would break references when new collection did not exist yet. * Fix alias target swap rollback after async ref backfill failure. * Split async reference backfill into verification and application steps. * Update reference alias swap, deferred symlink resolution and create collection to use the uniform logic of staging, verification and rollback. * Stage referenced_ins replay and deferred alias resolution together during collection creation. * Fix staged async backfills to patch helper fields only. * Obtain shared_ptr of Collection during backfilling to avoid dangling pointer edge case. * Skip document which has a new reference value in between staging and applying changes. | 2 个月前 | |
Rebind reference field to the new swapped collection alias. (#2963) * Rebind reference field to the new swapped collection alias. * Fix a scenario where swapping alias would break references when new collection did not exist yet. * Fix alias target swap rollback after async ref backfill failure. * Split async reference backfill into verification and application steps. * Update reference alias swap, deferred symlink resolution and create collection to use the uniform logic of staging, verification and rollback. * Stage referenced_ins replay and deferred alias resolution together during collection creation. * Fix staged async backfills to patch helper fields only. * Obtain shared_ptr of Collection during backfilling to avoid dangling pointer edge case. * Skip document which has a new reference value in between staging and applying changes. | 2 个月前 | |
Add streaming support for conversations | 1 年前 | |
feat(conversation): enhance SSE handling and fix async write callbacks (#2837) | 5 个月前 | |
Refactor conversation model. | 2 年前 | |
add prefix auth_key in access logs (#2915) * add prefix auth_key in access logs * Log resolved API key prefixes for scoped and multi-search requests * add empty query check --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
feat(api): add option to return deleted documents and IDs - Add `return_doc` and `return_id` parameters to document deletion API - Store deleted document content and IDs when requested - Include deleted items in API response when parameters are enabled - Add tests for both single and batch document deletion scenarios | 1 年前 | |
fix: make count min sketch hash seeding deterministic (#2897) * fix: make count min sketch hash seeding deterministic - seed hash generation with a fixed prime for repeatable behavior * refactor(count-min-sketch): use local mt19937 instead of rand/srand | 4 个月前 | |
Support synonyms with curations (#2708) * add stemming support for curations * update test * support synonyms with curations | 8 个月前 | |
update: curation refactor (#2574) * add: v30 overrides * update: UTs * fix: all UTs * add: migration support and integration tests * add: migration support and integration tests * fix: CI * rename: curation * rename: curation * rename: curation * fix: CI * add: curation_sets API tests * fix: unintentional override changes | 11 个月前 | |
fix: delete synonyms and curations during cluster rotation (#2719) * fix: delete during cluster rotation * fix: CI * fix: check for is_live_req | 8 个月前 | |
Single leaf addition (cvt). | 5 年前 | |
`vector_distance` similarity method for MMR diversity. (#2706) * Refactor `void process_results_bruteforce()`. * Support `vector_distance` similarity method. * Refactor `similarity_t::calculate`. | 8 个月前 | |
refactor: analytics (#2321) * add: doc analytics * add: query analytics * add: capture_search_request (internal query event) * add: analytics manager files renaming * feat: delayed write analytics write queue * add: migrate v29 to v30 * add: mising UTs * fix: app metrics UTs * update: POST /analytics/rules endpoint * add: POST /flush and GET /status for analytics * update: integration with timeout removed * rm: event_type in POST /events * fix: CI * update: no-phase test * update: CI with download artifact * fix: CI * fix: old rule migration bug * update: api_tests .gitignore * fix: PATCH rule with PUT (upsert) rule * fix: add removed tests * update: DELETE rule api response | 1 年前 | |
Add support for audio query | 2 年前 | |
fix: include num_dim in remote embedder cache key (port of #2721) (#2724) * fix: include num_dim in remote embedder cache key for matryoshka models * fix(embedder): include num_dims when updating remote embedder api key - pass num_dims to remote embedder key generation to ensure correct lookup - prevent missing or mismatched embedder entries * test(embedding): add tests for multi-dimension embedders and persistence - add tests for collections sharing the same model with different dimensions - verify indexing, search, restart, snapshot, and multi-node behavior for matryoshka embeddings --------- Co-authored-by: Reda El Issati <reda.elissati@datadoghq.com> | 8 个月前 | |
Fix exectimer. | 5 年前 | |
Add `diversity` override parameter. (#2572) * Add `diversity` override parameter. * Fix failing test. | 11 个月前 | |
Fix reference initialization bug. (#2944) * Fix reference initialization bug in case of reference to an alias collection name which is created later on. * Fix failing test. * PR review changes: * Call `update_reference_info()` in `CollectionManager::resolve_deferred_references_for_symlink` since we have already obtained a lock in `CollectionManager::upsert_symlink`. * Add `CollectionManager::validate_deferred_references_for_symlink`. * Add `Collection::backfill_async_reference_helpers`. * Validate async references before committing the referenced collection alias name. * Refactor filter string building logic in `Collection::async_reference_helper_backfill`. * Avoid deadlock scenario. * Refacto `CollectionManager::resolve_deferred_references_for_symlink()`. * Handle references to id field in `async_reference_helper_backfill`. * Handle a scenario where alias is created before referenced collection. * Drop the collection if an error is encountered while validating async_references. * Persist the initialized referenced_ins correctly on restart. * Return an error in `CollectionManager::resolve_deferred_references_for_symlink` if referenced field is not found. | 3 个月前 | |
Add create dir file util method. | 4 年前 | |
perf(index): split grouped second-pass string filters into chunks (#2925) * perf(index): split grouped second-pass string filters into chunks * fix(filter): add validate_max_ops flag to parse_filter_query - gate filter_by_max_ops check behind the new flag (default true) - pass false for the internally generated grouped second-pass filter, which is chunked into many ORed sub-filters and can exceed the limit even when the user's filter_by would not | 3 个月前 | |
feat: add existence filter for optional fields (#2777) * feat: add existence filter for optional fields * feat: add optional_index field to collection tests * refactor: use only missing index for existence check and update related tests * docs: clarify comment on _exists logic in filter_result_iterator * feat: add lazy evaluation for existence filter * clarify logic for applying not equals in filter initialization * fix bugs and rename to `track_missing_values`exist * Add tests for missing filter on dynamic fields * Add track_missing_values support for the fallback field * use id_list_t::iterator * Refactor test by removing redundant validity checks * Refactor filter_result_iterator by removing unused methods and simplifying logic * Fix validity assignment in filter_result_iterator reset method * remove unncessary indent --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 4 个月前 | |
refactor using s2regiontermindexer | 1 年前 | |
Pre-emptively log long running queries. | 6 个月前 | |
add ssl wrapper for remote model http calls (#2907) * add ssl wrapper for model http api calls * Fix SSE proxy handling for SSL verification failures --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
add prefix auth_key in access logs (#2915) * add prefix auth_key in access logs * Log resolved API key prefixes for scoped and multi-search requests * add empty query check --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
add ssl wrapper for remote model http calls (#2907) * add ssl wrapper for model http api calls * Fix SSE proxy handling for SSL verification failures --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
add prefix auth_key in access logs (#2915) * add prefix auth_key in access logs * Log resolved API key prefixes for scoped and multi-search requests * add empty query check --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
Add sampling for value based faceting. | 2 年前 | |
Add sampling for value based faceting. | 2 年前 | |
feat: add support for SigLIP tokenizer and image processor (#2795) | 6 个月前 | |
feat: add support for SigLIP tokenizer and image processor (#2795) | 6 个月前 | |
feat: Infix filtering support #2685 (#2813) * feat: Infix filtering support [#2685](https://github.com/mdhaduk/typesense/issues/2685) * refactor(infix-filter): lazy evaluation + remove multi-token infix support Changes: - include/index.h, src/index.cpp: add search_infix_leaves(); refactor search_infix() to call it - src/filter_result_iterator.cpp: lazy infix path via search_infix_leaves() + posting_list_iterators; 400 error for multi-token; remove has_infix_results block; remove #include <algorithm> - test/filter_test.cpp: add FilterTest.InfixLazyEvaluation (5 cases including multi-token 400 error) Assumptions: - Infix filter values must be single tokens. A value that tokenises to N > 1 tokens returns 400. This is safe: every existing test uses single-token patterns (*ris*, *pta*, etc.); no documented behaviour depended on multi-token AND semantics. * test(infix-filter): verbatim error assertions + AND-of-two-infix test case - Replace substring find() checks with verbatim ASSERT_EQ on full error messages in FilterTest.InfixLazyEvaluation - Add FilterTest.InfixLazyEvaluation Test 5: cast:*ris* && cast:*in* * test(infix-filter): add enable_lazy_filter and exact-match operator tests Note: Tests 10 and 11 use collectionManager.do_search() with req_params to exercise the enable_lazy_filter HTTP parameter, which is not exposed through the coll->search() | 3 个月前 | |
Handle concurrency of ja tokenizer. | 3 年前 | |
Fix race conditions in processing concurrent requests of related collections. (#2815) * Remove reference helper fields in response. * Refactor `Index::update_async_references` logic. * Implement `Collection::reset_referencing_documents()`. * Refactor `Collection::reset_referencing_documents`. * Refactor cascade delete logic. * Refactor serialization logic of related requests. * Refactor `BatchedIndexer::get_requests_to_wait_on` to use request's log_index, when available. * Add `reference_cascade_delete_concurrent.test.ts`. * Fix api-test failing due to collision of collection name since the test state is shared. * Save `waiting_on_requests` in snapshot and load on restart. Re-introduce `BatchedIndexer::coll_to_references` to hold the references of every collection. Return `last_index`, `known_applied_index`, `applying_index` in `/status` response. | 5 个月前 | |
Add support for image embeddings using CLIP | 2 年前 | |
Group key should be actual values. | 6 年前 | |
Group by optimization. (#2146) * Collect top-k groups in the first pass. * Filter on the group_by field values in the second pass. * Collect document ids where group_by values are missing in the first pass. * Use hyperloglog_hip algorithm to approximate found value during group_by. * `found` count in case of `group_by` should be at least equal to the `grouped_hits` size. * With `sort_by: _group_found`, gather approximate group size in the first pass. * Refactor union sorting logic. * Fix rebase error. * Fix concurrent insertion in `group_by_missing_value_ids`. * Count of group elements can't exceed `UINT32_MAX`. * Refactor group_by first pass logic to get facet values from index instead of documents. Remove `numeric_range_trie_test.h`. * Reuse the facet value iterators during first pass. * Refactor `Index::get_group_by_values` and `Topster::sort`. * Replace `hyperloglog_hip` with `LogLogBeta`. * Refactor `Index::search_wildcard` first group_by pass logic. * Fix `found_docs` value mismatch in CollectionGroupingTests. | 1 年前 | |
Handle bad data gracefully. | 5 年前 | |
update: synonyms proximity logic (#2530) * update: synonyms proximity logic * update: simplify synonyms test * add: relative scoring for num_match tokens based on synonyms * add: relative scoring only for proximity * add: relative scoring and prioritize_synonym_match flag for search * fix: all UTs * add: old assertions in collection_synonyms_test.cpp * update: demote_synonym_match * update: demote_synonym_match * fix: text_match_scores * fix: DemoteSynonymMatch name | 11 个月前 | |
feat(nl): add azure openai support for natural language search (#2478) | 1 年前 | |
add: natural language search (#2357) * Framework for natural language search * refactor: move logic out of core_api * add: tests for natural_language_search_model.cpp * add: tests for natural_language_search_model_manager.cpp * add: tests for cloudflare and tests for schema prompts cache * add: tests for query parsing and results addition * trigger: ci * fix: flakey UTs * fix: model incompetency in returning a wrong parsable response. * fix: ci * update: search_time_ms * fix: review commets --------- Co-authored-by: Jason Bosco <mail@jasonbos.co> Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 1 年前 | |
Numeric iterator (#1709) * Numeric filter iterator. * Add test cases. * Refactor. * Float field filter iterator. | 2 年前 | |
Ditch use of number_t for sorting. | 6 年前 | |
Optimize `NumericTrie::Node::search_range_helper`. (#2390) * Optimize `NumericTrie::Node::search_range_helper`. * Refactor `NumericTrie::Node` methods. | 1 年前 | |
Add `filter_result_iterator_t`. | 3 年前 | |
Fix typo threshold for curated hits. | 1 年前 | |
Fix perf in filtering results. | 4 年前 | |
fix: num_dims (#2529) * fix: num_dims * update: assertions | 1 年前 | |
add: personalization model manager disposer for using manager across tests | 1 年前 | |
Fix highlight race condition. | 6 个月前 | |
fix token offsets when prioritized (#2740) * fix token offsets when prioritized * normalize and clamp single-token posting list offsets | 5 个月前 | |
add: IPv6 support (#2397) * Enable the http server to listen in dual-stack mode and covert obvious ipv4-only code to dual-stack enabled code. * IPv6 fixes for raft communication * Small fixes for ipv6 * Change dynamic ip detection for integration tests so that test generated nodes file always matches peering address * Removed commented out code. Reduce unhelpful error messages. Extract ReplicationState::hostname2ipstr from ReplicationState::resolve_node_hosts and unit test * Update get_api_key_and_ip to handle ipv6 and add unit test for it --------- Co-authored-by: Jonathan Crapuchettes <jonathan@redballoon.work> Co-authored-by: Jonathan Crapuchettes <jcrapuchettes@gmail.com> Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 1 年前 | |
Rate limit API updates & improvements (#846) * Rate limit procedure updates & improvements * Review changes * Update for adding support to delete temp throttles * Review Changes II * Review Changes III | 3 年前 | |
update: query events to fully as search events (#2565) * update: query events to fully as search events * fix: ci errors * fix: ci | 11 个月前 | |
Use UUID for conversation ids | 2 年前 | |
Basics of a block based posting list container. | 5 年前 | |
Added an ART search token cache. To cache previous searches so that we don't repeatedly call ART search as we iterate through the correction. | 9 年前 | |
Patches for ARM. | 4 年前 | |
Fix stability of stack traces on Linux | 5 年前 | |
Normalize plurals (#2062) * initial commit * allow multiple plurals set * acess dictionary word from stemmerManager directly * use find over count * make dictionary post request async * add endpoints to get dictionary * update function names * update core_api function names * fix the test * update end-point names * reload dictionaries on restart --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 1 年前 | |
return error response when stopword is not found (#1545) * return error response when stopword is not found * use shared_lock | 2 年前 | |
rockdb param flags (#2371) * add rocksdb param flags custom tuning * make indexing concurrency configurable * update param name * update funciton name * add environment config | 1 年前 | |
Allow multiple joins to the same collection in `filter_by`. (#2398) * Refactor `StringUtils::tokenize_filter_query`. * Allow multiple joins to the same collection. * Refactor filtering related methods out of `StringUtils`. | 1 年前 | |
Improve synonym index concurrency. | 2 个月前 | |
Improve synonym index concurrency. | 2 个月前 | |
Implement cgroups conditions memory handling (#2906) * Implement cgroups conditions memory handling Signed-off-by: stafot <stafot@gmail.com> * fix: resolve container cgroup path via /proc/self/cgroup The initial implementation read memory.max and memory.current from fixed paths under /sys/fs/cgroup/, which only works when the container runtime sets up a cgroup namespace (cgroupns) — remounting the pod's own cgroup as the root inside the container. Without cgroupns the container sees the host cgroup hierarchy, so the pod's memory files live under a nested path such as /sys/fs/cgroup/kubepods.slice/.../scope/memory.max rather than at the root. The guard would silently fall back to node-level sysinfo, never firing for the container. Fix: read /proc/self/cgroup at runtime to resolve the process's actual cgroup path (0::<rel_path> for v2, memory:<rel_path> for v1), then construct the full file paths as cgroup_root + rel_path + /memory.max etc. This works correctly both with and without cgroupns. get_cgroup_memory signature updated to accept injectable proc_self_cgroup and cgroup_root paths for testing. Test fixtures restructured to simulate a real cgroup filesystem layout under test/resources/cgroupfs/. Signed-off-by: stafot <stafot@gmail.com> * Revert unneeded changes Signed-off-by: stafot <stafot@gmail.com> * Reorg tests Signed-off-by: stafot <stafot@gmail.com> * Fix test sandbox fixtures Signed-off-by: stafot <stafot@gmail.com> --------- Signed-off-by: stafot <stafot@gmail.com> | 2 个月前 | |
app metrics percentile latencies (#2158) * add min,max,p70,p95,p99 latencies * add tdigest based approximated quantile * add test for estimated quantiles * update test for large dataset | 1 年前 | |
fix: align remote embedder query timeout default (#2840) | 5 个月前 | |
fix: align remote embedder query timeout default (#2840) | 5 个月前 | |
feat: add support for SigLIP tokenizer and image processor (#2795) | 6 个月前 | |
Enable slow searches logging by default (30s cutoff) | 3 年前 | |
Move thread pool exhaustion logging to separate func. | 2 年前 | |
Improve sorting during filtering. | 4 年前 | |
don't transliterate stopword tokens (#2781) * don't transliterate stopword tokens * add test * rename boolean to do_transliterate and make it default enabled | 6 个月前 | |
Filter out duplicates with Union (#2511) * add support for de-duplication with unions * update found count with duplicates * make remove_duplicates default enabled | 1 年前 | |
Transliterator objects pool (#2412) * pool transliterator objects * refactor transliterators pool * use condition variables, restructure code | 1 年前 | |
add ssl wrapper for remote model http calls (#2907) * add ssl wrapper for model http api calls * Fix SSE proxy handling for SSL verification failures --------- Co-authored-by: Kishore Nallan <kishorenc@gmail.com> | 3 个月前 | |
add param shutdown-delay-seconds (#2711) * add configuratble param shutdown_delay_seconds * update the signal for delayed shutdown | 6 个月前 | |
Merge branch 'v0.25-join' into v0.26-facets | 3 年前 | |
Vector query parse bugs (#1937) * fix _vector_query parsing in sort_by param * update model and make brace counter integer * _vector_query bug fixes * set default distance_threshold to max float value | 1 年前 | |
Fix voice query model references and error messages | 2 年前 | |
Add `ts` namespace for voice query models (#1549) * Add `ts` namespace for voice query models * Fix voice query test * Update error message in invalid voice query test | 2 年前 | |
Refactor facet search to use hash based mapping. Also fixed a bug where filter fields marked as facets were being searched only verbatim. | 6 年前 |
| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
| 1 年前 | ||
| 1 年前 | ||
| 4 年前 | ||
| 11 个月前 | ||
| 7 个月前 | ||
| 3 个月前 | ||
| 1 年前 | ||
| 5 年前 | ||
| 3 年前 | ||
| 3 年前 | ||
| 2 年前 | ||
| 3 个月前 | ||
| 3 年前 | ||
| 4 个月前 | ||
| 2 年前 | ||
| 4 年前 | ||
| 2 个月前 | ||
| 2 个月前 | ||
| 1 年前 | ||
| 5 个月前 | ||
| 2 年前 | ||
| 3 个月前 | ||
| 1 年前 | ||
| 4 个月前 | ||
| 8 个月前 | ||
| 11 个月前 | ||
| 8 个月前 | ||
| 5 年前 | ||
| 8 个月前 | ||
| 1 年前 | ||
| 2 年前 | ||
| 8 个月前 | ||
| 5 年前 | ||
| 11 个月前 | ||
| 3 个月前 | ||
| 4 年前 | ||
| 3 个月前 | ||
| 4 个月前 | ||
| 1 年前 | ||
| 6 个月前 | ||
| 3 个月前 | ||
| 3 个月前 | ||
| 3 个月前 | ||
| 3 个月前 | ||
| 2 年前 | ||
| 2 年前 | ||
| 6 个月前 | ||
| 6 个月前 | ||
| 3 个月前 | ||
| 3 年前 | ||
| 5 个月前 | ||
| 2 年前 | ||
| 6 年前 | ||
| 1 年前 | ||
| 5 年前 | ||
| 11 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 2 年前 | ||
| 6 年前 | ||
| 1 年前 | ||
| 3 年前 | ||
| 1 年前 | ||
| 4 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 6 个月前 | ||
| 5 个月前 | ||
| 1 年前 | ||
| 3 年前 | ||
| 11 个月前 | ||
| 2 年前 | ||
| 5 年前 | ||
| 9 年前 | ||
| 4 年前 | ||
| 5 年前 | ||
| 1 年前 | ||
| 2 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 2 个月前 | ||
| 2 个月前 | ||
| 2 个月前 | ||
| 1 年前 | ||
| 5 个月前 | ||
| 5 个月前 | ||
| 6 个月前 | ||
| 3 年前 | ||
| 2 年前 | ||
| 4 年前 | ||
| 6 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 3 个月前 | ||
| 6 个月前 | ||
| 3 年前 | ||
| 1 年前 | ||
| 2 年前 | ||
| 2 年前 | ||
| 6 年前 |