SDKs
Python SDK (Beta) v1.X
Reference
Changelog

Changelog

Release history for the DealCloud Python SDK (dealcloud-sdk). The package is available on PyPI (opens in a new tab).

[Unreleased]

Added

  • Opt-in error enrichment (dealcloud-sdk 1.0.11): get_field_lookup, build_field_lookup, enrich_row_errors, format_error_with_metadata, format_error_message_english, and row-level wrappers. Not wired into standard write paths. See Error Enrichment.

Added (prior)

  • DealCloudConfig.raiseOnRowErrors: client default for write methods when raise_on_row_errors is omitted.
  • upsert_data: error_handling, output="write_result", BatchResult on ErrorHandling.COLLECT.
  • write_cells: error_handling, output="write_result", BatchResult on COLLECT.
  • delete_cells: error_handling, raise_on_row_errors, BatchResult on COLLECT (mirrors delete_data).
  • Row-level error surfacing for Rows writes: raise_on_row_errors, BatchResult.row_errors, RowsWriteResult (output="write_result"), and exported helpers split_row_results / finalize_list_response.
  • delete_data: raise_on_row_errors and BatchResult.row_errors on ErrorHandling.COLLECT.

Changed

  • Parallel pagination: object reads (read_data, read_data_streaming, and async variants) now fetch pages concurrently by default, bounded by read_concurrency (config concurrencyLimits.read, default 4). Pages were previously always fetched sequentially. Pages are still yielded in order; large objects read up to ~read_concurrencyx faster. See Streaming and Performance.
  • resolve on all outputs: the legacy resolve parameter now applies to list, polars, and polars_lazy in addition to pandas. Previously it was silently ignored for non-pandas outputs. Resolution is local (no extra API calls); reference_format still takes precedence. See Reference Fields.
  • raise_on_row_errors default is now None (inherit config; effective default remains False).
  • Removed legacy throw_error alias on process_errors.
  • Cells write error handling aligned with Rows write contract.
  • upsert_data: returns the full API row list on HTTP 200 (including rows with "Errors"), consistent with insert_data / update_data.
  • Documentation for ErrorHandling.COLLECT: examples use BatchResult.results and BatchResult.row_errors instead of iterating the result as a list.
  • Row error decision matrix and FAIL_FAST documented as transport-only across write docs.

Changed (prior)

  • Narrative docs under /sdk/python-1.X (data, files, users, backups, history, publications, merge, RI, reference) aligned with dealcloud-sdk 1.x and intapp-rest-client behavior.
  • Package releases continue to follow your internal PyPI process; this site tracks the same changelog as the repo when published.

[1.0.0] — 2026

First 1.x line on PyPI (dealcloud-sdk). Built on intapp-rest-client (see pyproject.toml for the pinned version), pandas, pydantic, numpy, and fsspec.

Added

  • DealCloud client with schema, data (rows), cells, history, files, users, publications, backups, relationship intelligence, merge, and related APIs.
  • Optional extras: polars, yaml, s3, azure, gcs, all.
  • Documentation aligned with this site under /sdk/python-1.X.

Format

Releases follow Keep a Changelog (opens in a new tab). Types of changes:

  • Added – new features
  • Changed – changes in existing functionality
  • Deprecated – soon-to-be removed features
  • Removed – removed features
  • Fixed – bug fixes
  • Security – vulnerability fixes
ExceptionsLicense