v0.11.0

What's Changed

• Introduce more Ruff rules to improve code quality by @corafid in #69
• Add type support to update_group_authorization by @corafid in #70

Full Changelog: v0.10.1...v0.11.0

v0.10.1

What's Changed

• fix: set the default value of DocumentStatus.updated_at to None. by @luca-cohere in #68

New Contributors

• @luca-cohere made their first contribution in #68

Full Changelog: v0.10.0...v0.10.1

v0.10.0

What's Changed

• Add asset_type into the DocumentChunkAsset struct by @artemiyatcohere in #63

Full Changelog: v0.9.4...v0.10.0

v0.9.4

What's Changed

• fix: Allows clients to pass in an HTTP Session by @coessiane in #59
• Propagate the 400 error codes by @benrules3 in #61
• Adjusting fsspec to fix a conflict with other libraries used in load testing by @dbaxter-cohere in #65
• Pyproject version bump 0.9.3 to 0.9.4 by @artemiyatcohere in #67

New Contributors

• @coessiane made their first contribution in #59
• @dbaxter-cohere made their first contribution in #65

Full Changelog: v0.9.1...v0.9.4

v0.9.1

What's Changed

• Move modules under "cohere.compass" + More refactoring by @corafid in #56
• Update document name in models by @benrules3 in #57
• Renames related to the recent API renames by @corafid in #58

Full Changelog: v0.7.0...v0.9.1

v0.8.0

What's Changed

• Move modules under "cohere.compass" + More refactoring by @corafid in #56

Full Changelog: v0.7.0...v0.8.0

v0.7.0

What's Changed

• General refactoring and code quality improvement by @corafid in #54
• Support more datasources APIs by @corafid in #55

Full Changelog: v0.5.1...v0.7.0

v0.5.1

What's Changed

• require python >= 3.11 by @d-rau in #37
• Add optional content_type to parsing call by @javier-cohere in #38
• Add assets to the Chunk model by @corafid in #32
• Fix bug when adding errors in insert_docs. by @javier-cohere in #39
• Removing root from permissions. As root user cannot be created from here by @ankush-cohere in #40
• Configurable heavy parsing strategy by @javier-cohere in #41
• Revert "Configurable heavy parsing strategy (#41)" by @javier-cohere in #43
• Add doc filename to errors and increase max retries by @javier-cohere in #46
• Remove tqdm by @javier-cohere in #47
• feat: add default timeout by @raphael-cohere in #49
• Update API endpoints by @corafid in #45
• Support datasources APIs by @corafid in #50
• Support search_chunk API by @corafid in #51
• Add models for search APIs by @corafid in #52
• [SEA-1683] Fix retry mechanism for 400 errors by @corafid in #53

New Contributors

• @d-rau made their first contribution in #37
• @corafid made their first contribution in #32
• @raphael-cohere made their first contribution in #49

Full Changelog: 0.3.2...v0.5.1

0.3.2

What's Changed

• Add group merging logic to update_doc calls by @benrules3 in #26
• Fix empty iterator due to wrong filtering expression by @benrules3 in #29
• Sync ParserConfig settings with the current Parser HTTP API by @artemiyatcohere in #30
• Remove apply_pdf_rough_text_extraction and add pdf_parsing_strategy by @javier-cohere in #33
• Cleanup batch processing by @ankush-cohere in #34
• Add parameters for the new pptx parser by @javier-cohere in #35
• Increment version for tip of main by @benrules3 in #36

New Contributors

• @artemiyatcohere made their first contribution in #30

Full Changelog: release.pangolin...0.3.2

Pangolin

Adapts to the latest changes in the compass-parser new release, release.pangolin.

• Removes the is_dataset parameter from CompassDocument and all the parser process file/files/folder functions.
• CompassDocumentChunk.origin is no longer a pydantic model, but a Dict[str, Any] so that we can add arbitrary non-numeric fields such as "tab_name": "some name" for spreadsheets.