byteit 1.1.2__tar.gz → 1.2.0__tar.gz

{byteit-1.1.2 → byteit-1.2.0}/CHANGELOG.md

@@ -1,115 +1,148 @@
-# Changelog
-
-All notable changes to this project will be documented in this file.
-
-The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
-and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
-
-## [Unreleased]
-
-## [1.1.2] - 2026-07-08
-- Fixed pruning of title fields in byteit schemas
-- Improved error handling for downloading result
-
-## [1.1.2] - 2026-06-08
-- Improved error handling for downloading result
-
-## [1.1.1] - 2026-06-05
-
-### Fixed
-- Restored `result_format` support for synchronous `parse()` so callers can
-  explicitly choose the returned output format.
-
-
-## [1.1.0] - 2026-05-06
-
-### Breaking Changes
-- Renamed parse-job public methods to explicit parse naming:
-  - `get_jobs()` -> `get_parse_jobs()`
-  - `get_job_details(job_id)` -> `get_parse_job_details(job_id)`
-  - `get_job_result(job_id)` -> `get_parse_job_result(job_id)`
-- Removed `result_format` parameter from `parse()` and `parse_async()`.
-  Parse job submission now defaults to JSON internally.
-
-### Added
-- Structured extraction workflow for completed parse jobs:
-  - `extract(parse_job_id, schema, output=None, extraction_complexity="medium")`
-  - `extract_async(parse_job_id, schema, extraction_complexity="medium")`
-  - `get_extract_jobs()`, `get_extract_job_details(job_id)`,
-    `get_extract_job_result(job_id)`
-- Optional download-time output selection for async parse results via
-  `get_parse_job_result(job_id, result_format=...)`.
-- New extraction domain models and schema support:
-  - `ExtractJob`, `ExtractJobList`, `ExtractionSchema`
-  - `ExtractionSchema` export via package `__init__` when available
-
-### Changed
-- Parse and extract operations now use dedicated endpoints under `/v1/jobs/`:
-  - parse jobs under `parse-jobs`
-  - extract jobs under `extract-jobs`
-- `get_job_status(job_id)` continues using the generic lightweight status path
-  while detail and result operations route to parse/extract-specific endpoints.
-- `result_format` validation accepts both enum names and backend value tokens
-  (for example `"excel"` and `"zip"`).
-- Dependency set now includes `pydantic` for extraction schema flows.
-
-
-## [1.0.1] - 2026-04-29
-
-### Added
-- Local release preflight checks for version consistency, changelog presence, and package integrity.
-- Commit-title validation for release commits (`chore(release): X.Y.Z`) through pre-commit hooks.
-- Release checklist skill document to standardize pre-commit and pre-publish verification.
-
-
-## [1.0.0] - 2026-03-25
-
-### Added
-- `parse_async()` method for non-blocking document submission.
-- `_submit_job()` internal helper to eliminate duplication between `parse()` and `parse_async()`.
-
-
-## [1.0.0] - 2026-03-25
-
-### Added
-- `parse_async()` method for non-blocking document submission.
-- `_submit_job()` internal helper to eliminate duplication between `parse()` and `parse_async()`.
-
-
-## [0.1.2] - 2026-01-31
-
-### Changed
-- Processing migrated to top-enhanced production servers, reducing processing time by ~50%.
-
-- Adaptive polling improved for smoother operation.
-
-- Progress bar and output display enhanced for notebooks.
-
-- Increased `DEFAULT_TIMEOUT` from 10 minutes to 30 minutes for large documents
-
-## [0.1.1] - 2026-01-24
-
-### Changed
-- Increased `DEFAULT_TIMEOUT` from 30 seconds to 10 minutes for large documents
-- Adaptive polling in `_wait_for_completion`: starts at 2s, increases to max 10s
-
-### Added
-- Progress logging during document parsing
-
-## [0.1.0] - 2026-01-18
-
-### Added
-- Initial release of the ByteIT Python SDK
-- `ByteITClient` for AI-powered document parsing
-- Multiple output formats: text, JSON, Markdown, HTML
-- Input connectors:
-  - `LocalFileInputConnector`
-  - `S3InputConnector`
-- Output connector:
-  - `LocalFileOutputConnector`
-- Job management (list jobs, check status, download results)
-- Support for PDF, Word, Excel, and other common document formats
-- Batch processing support
-- Environment variable configuration
-- Custom base URL support (testing & staging)
-- Python 3.8+ support
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [1.2.0] - 2026-07-28
+
+### Added
+- Custom processing jobs: `custom_job()`, `custom_job_async()`,
+  `get_custom_jobs()`, `get_custom_job_result()` and `CustomJob`, `CustomJobList`
+  models.
+- Saved schema management: `save_schema()`, `get_schemas()`, `delete_schema()`
+  and `SavedSchema`, `SavedSchemaList` models.
+- `parse_async()` now accepts a list of file paths (`list[str | Path]`) for
+  multi-file batch submission with automatic packing and per-file failure
+  reporting.
+- `force_image_annotations` option on `ProcessingOptions`.
+- `batch_request_delay` parameter on `ByteITClient` to pace consecutive uploads.
+- `DocumentType` model with `is_supported_extension()` for file-type filtering.
+- Internal modules extracted: `_http.py` (HTTP helpers), `_polling.py` (shared
+  adaptive polling), `_rate_limit.py` (rate-limited submitter with backoff).
+- CI: standalone `pytest-unit.yml` workflow; AdolfoBOT failure notification on
+  publish.
+- HTTP API examples in `docs/api-curl-typescript-examples.md`.
+
+### Changed
+- CI/CD workflows consolidated to call `scripts/release_preflight.py --ci` for
+  all validation, removing inline release checks.
+- Release preflight script refactored with shared helpers and clearer error
+  messages.
+- Rate-limit handling and polling logic extracted from `ByteITClient` into
+  dedicated modules.
+
+### Fixed
+- `RateLimitError` now carries `retry_after_seconds` attribute from response
+  headers.
+
+### Removed
+- `queue_for_batch` parameter from synchronous `parse()` (was async-only).
+- Inline release validation logic from CI/CD workflows.
+- Stale test file `rate_limit_test/run_batch_async_test.py`.
+
+## [1.1.2] - 2026-07-08
+- Fixed pruning of title fields in byteit schemas
+- Improved error handling for downloading result
+
+## [1.1.1] - 2026-06-05
+
+### Fixed
+- Restored `result_format` support for synchronous `parse()` so callers can
+  explicitly choose the returned output format.
+
+## [1.1.0] - 2026-05-06
+
+### Breaking Changes
+- Renamed parse-job public methods to explicit parse naming:
+  - `get_jobs()` -> `get_parse_jobs()`
+  - `get_job_details(job_id)` -> `get_parse_job_details(job_id)`
+  - `get_job_result(job_id)` -> `get_parse_job_result(job_id)`
+- Removed `result_format` parameter from `parse()` and `parse_async()`.
+  Parse job submission now defaults to JSON internally.
+
+### Added
+- Structured extraction workflow for completed parse jobs:
+  - `extract(parse_job_id, schema, output=None, extraction_complexity="medium")`
+  - `extract_async(parse_job_id, schema, extraction_complexity="medium")`
+  - `get_extract_jobs()`, `get_extract_job_details(job_id)`,
+    `get_extract_job_result(job_id)`
+- Optional download-time output selection for async parse results via
+  `get_parse_job_result(job_id, result_format=...)`.
+- New extraction domain models and schema support:
+  - `ExtractJob`, `ExtractJobList`, `ExtractionSchema`
+  - `ExtractionSchema` export via package `__init__` when available
+
+### Changed
+- Parse and extract operations now use dedicated endpoints under `/v1/jobs/`:
+  - parse jobs under `parse-jobs`
+  - extract jobs under `extract-jobs`
+- `get_job_status(job_id)` continues using the generic lightweight status path
+  while detail and result operations route to parse/extract-specific endpoints.
+- `result_format` validation accepts both enum names and backend value tokens
+  (for example `"excel"` and `"zip"`).
+- Dependency set now includes `pydantic` for extraction schema flows.
+
+
+## [1.0.1] - 2026-04-29
+
+### Added
+- Local release preflight checks for version consistency, changelog presence, and package integrity.
+- Commit-title validation for release commits (`chore(release): X.Y.Z`) through pre-commit hooks.
+- Release checklist skill document to standardize pre-commit and pre-publish verification.
+
+
+## [1.0.0] - 2026-03-25
+
+### Added
+- `parse_async()` method for non-blocking document submission.
+- `_submit_job()` internal helper to eliminate duplication between `parse()` and `parse_async()`.
+
+
+## [1.0.0] - 2026-03-25
+
+### Added
+- `parse_async()` method for non-blocking document submission.
+- `_submit_job()` internal helper to eliminate duplication between `parse()` and `parse_async()`.
+
+
+## [0.1.2] - 2026-01-31
+
+### Changed
+- Processing migrated to top-enhanced production servers, reducing processing time by ~50%.
+
+- Adaptive polling improved for smoother operation.
+
+- Progress bar and output display enhanced for notebooks.
+
+- Increased `DEFAULT_TIMEOUT` from 10 minutes to 30 minutes for large documents
+
+## [0.1.1] - 2026-01-24
+
+### Changed
+- Increased `DEFAULT_TIMEOUT` from 30 seconds to 10 minutes for large documents
+- Adaptive polling in `_wait_for_completion`: starts at 2s, increases to max 10s
+
+### Added
+- Progress logging during document parsing
+
+## [0.1.0] - 2026-01-18
+
+### Added
+- Initial release of the ByteIT Python SDK
+- `ByteITClient` for AI-powered document parsing
+- Multiple output formats: text, JSON, Markdown, HTML
+- Input connectors:
+  - `LocalFileInputConnector`
+  - `S3InputConnector`
+- Output connector:
+  - `LocalFileOutputConnector`
+- Job management (list jobs, check status, download results)
+- Support for PDF, Word, Excel, and other common document formats
+- Batch processing support
+- Environment variable configuration
+- Custom base URL support (testing & staging)
+- Python 3.8+ support