Tags: weaviate/weaviate-python-client
Tags
Support `1.37` features (#1988) * Refactor hybrid queries to use `alpha_param` and remove `0.7` default in favour of `None` * Remove mistakenly commited local change to `regen.sh` * Update logic to use new proto message * Change formatting * Tidy version check code * Parse correct default for BC if server < 1.36 * Update CI image * Fix wrong version comparison * Fix typo in ci * Update CI image * Remove client-side default from aggregate queries * Update ver check and CI tags * Remove test of lazy loading shards * Refactor client test for new server lazy shard loading * Debug failing ci test * Remove outdated lazy shard load test * Update CI images * Add per-test timeouts and stack dump on timeout * Reduce per-test timeout to 5 mins * Fix inc backups test * Add server version check for incremental backups * Remove comment * Hard kill the process on timeout detection * Timeout putting the sentinel to avoid deadlocking * Handle the timeout of sentinel pushing gracefully * feat: add TextAnalyzerConfig for ASCII folding in text properties * refactor: ruff format * feat: add min version check * feat: update TextAnalyzerConfig docstring for ascii_fold attributes * feat: add asciiFold check in _text_analyzer_from_config function * test: fix ASCII folding tests * feat: add support for stopword presets in inverted index configuration and text analyzer * test: added live and config tests * refactor: improve docstrings for stopword presets and asciiFold tests * refactor: simplify _any_property_has_text_analyzer function using _property_has_text_analyzer * test: remove redundant insertion ascii fold tests from test_collection_config * test: add stopwords roundtrip test for collection configuration * feat: add model validator to enforce asciiFoldIgnore constraints in TextAnalyzerConfig * feat: add factory class for text analyzer configurations with ASCII folding support * refactor: update TextAnalyzerConfig usage to new Configure class methods * test: remove redundant line in stopword presets merge test * refactor: use factory pattern * Add MCP permission * refactor: format text analyzer configuration for better readability * refactor: remove server side behavior tests * test: add stopword presets roundtrip tests for Weaviate collections * Fix formatting * refactor: remove unnecessary stopword preset coercion from _TextAnalyzerConfigCreate * refactor: replace custom text analyzer method with a direct function call * chore: remove unused deprecated import from config.py * chore: update WEAVIATE_137 version to 1.37.0-rc.1-578c4eb in workflow * refactor: update text analyzer method to use new static method in Configure * test: add stopwords roundtrip test with ASCII folding configuration * Add query profiling * Reformatted * Skip test for lower versions * feat: add tokenizer module with sync and async support, including integration tests * Add support for collection export endpoint * Small cleanup after review * Rename ENum * adapt to latest version * Update UX * Remove export path parameter * Self-review of changes * Review fixes * Add version guard for export integration tests * Update to latest image * Lowercase export ID * Enforce kwargs for export * Fix tests * Fix tests * Add negative assertions * refactor: names don't shadow existing * fix: add version gate * refactor: update tokenization type to use Tokenization enum in TokenizeResult and related tests * refactor: models * refactor: move tokenize property to class config * fix: remove trailing whitespace in __init__.py * test: add version gate for Weaviate >= 1.37.0 in tokenization tests * feat: add support for blobHash property type * Add full_with_profile * Refactor RBAC permissions * Bump Weaviate version * refactor: tokenization executor and models to support stopword configurations and improve response handling * fix: update Weaviate 1.37.1 version to include specific build identifier * Reduce timeouts in batch tests * fix: update Weaviate 1.37.1 version to include architecture suffix * fix: refactor tokenization tests to use parameterized cases for improved readability and maintainability * fix: update Weaviate 1.37.1 version and enhance tokenization tests with new fixtures * refactor: ruff format * test: refactor output types and tests to config * refactor: remove unused imports in tokenization models and format * Use public classes for .text endpoint * Add overloads for exclusivity of stopwrods * Accept collection config classes as stopwords * Improve docstring * Hook up tokenization and clean up model * Move property back to tokenization * Add changelog --------- Co-authored-by: André Mourão <andre.mourao@weaviate.io> Co-authored-by: Dirk Kulawiak <dirk@semi.technology> Co-authored-by: André Mourão <andre.b.mourao@gmail.com> Co-authored-by: Ivan Despot <66276597+g-despot@users.noreply.github.com> Co-authored-by: Dirk Kulawiak <dirk@weaviate.io> Co-authored-by: Marcin Antas <antas.marcin@gmail.com>
Merge pull request #2010 from weaviate/mcp-rbac Add MCP permission
Merge pull request #1848 from weaviate/rob/spfresh Add Hfresh index type
PreviousNext