Releases: caltechlibrary/dataset
Release list
v2.5.1
- Split SQLite driver by build target: normal builds use glebarez/go-sqlite (pure-Go, no WASM runtime overhead, eliminates spurious stderr warnings); WASM/wasip1 builds use ncruces/go-sqlite3 with the SQLite binary embedded
- Fixed nil pointer dereference in dsquery when querying a pairtree collection with --pt-index
Full Changelog: v2.5.0...v2.5.1
v2.5.0
- Added libdataset WASM module (GOOS=wasip1) replacing the old cgo-based C shared library, with Python (wasmtime-py) and TypeScript/Deno wrappers
- Named SQL queries pre-configured via settings.yaml-style config; new release artifact dataset-vVERSION-libdataset.zip
- Switched SQLite driver to ncruces/go-sqlite3 (SQLite 3.53.1, up from 3.41.2), enabling WASM builds while retaining FTS5 and full feature set
- Fixed datasetd request body size limits not being enforced (http.MaxBytesReader now applied)
- Fixed datasetd error redirect using HTTP 304 instead of HTTP 303
- Replaced deprecated io/ioutil throughout
Full Changelog: v2.4.0...v2.5.0
v2.4.1-rc2
- Removed MySQL support
- Added schema-based validation for dataset collections using the models package
- Added support for nested object and list structures in schema definitions
- Added identifier types (ISBN, ISSN, DOI, ORCID, ROR, ISNI, PMID, PMCID, FundRef, LCNAF, VIAF, SNAC, ArXiv, EAN) for CrossRef/DataCite record validation
- Added schemas configuration to settings.yaml for defining reusable validation schemas
- Added schema_name and validate fields to collection configuration for enabling per-collection validation
- API now returns X-Validation-Errors header with detailed validation error information on create/update failures
- Added validation tests, fixed generator and validation sequence bug.
Full Changelog: v2.4.1-rc1...v2.4.1-rc2
v2.4.1-rc1
- Removed MySQL support
- Added schema-based validation for dataset collections using the models package
- Added support for nested object and list structures in schema definitions
- Added identifier types (ISBN, ISSN, DOI, ORCID, ROR, ISNI, PMID, PMCID, FundRef, LCNAF, VIAF, SNAC, ArXiv, EAN) for CrossRef/DataCite record validation
- Added schemas configuration to settings.yaml for defining reusable validation schemas
- Added schema_name and validate fields to collection configuration for enabling per-collection validation
- API now returns X-Validation-Errors header with detailed validation error information on create/update failures
Full Changelog: v2.4.0...v2.4.1-rc1
v2.4.0
- Removed MySQL support
Full Changelog: v2.3.4...v2.4.0
v2.4.0-rc1
v2.3.4
- Deprecated MySQL support in documentation, will depreciate MySQL support in future release
- Improved loading large JSON objects from jsonl files
- Fixed issue #164 where the queries in COLD would work in v2.2.0 but fail in v2.3.x.
- Removed support for SQL parameters in dsquery due to encoding issues and lack of practical use cases
- Removed duplicated code from dsquery.go and api_routes.go in favor of collection.go's implementation of query functionality.
- Added tailing semi-colon removal for SQL queries due to changes in behavior of SQLite3 driver
Full Changelog: v2.3.3...v2.3.4
v2.3.3
- Issue #161 fix for handling GET with query were data is passed via URL parameters.
- Removed support for frame, clone, sample, sync and join support removed. The dsimporter cli removed (use jsonl dump and load instead).
- Improved loading large JSON objects from jsonl files
What's Changed
- Bump filippo.io/edwards25519 from 1.1.0 to 1.1.1 by @dependabot[bot] in #162
Full Changelog: v2.3.2...v2.3.3
v2.3.2
Issue #161 fix for handling GET with query were data is passed via URL parameters.
Removed support for frame, clone, sample, sync and join support removed. The dsimporter cli removed (use jsonl dump and load instead).
Full Changelog: v2.3.1...v2.3.2
v2.3.1
Issue #152 has started. The frames, sample and clone help has been removed as those features are depreciated. The frames supprt was remove from datasetd. Help is being reorganizaed. api_doc.go and api_cmd.go was removed as it was dead code with changes in help implementaiton.
The -help option can pull up help on a topic by providing a keyword as a separate parameter. This let's you pull up help on the related commands.
Example:
dataset -help api
dastaet -help dsqueryHelp text is now maintain inside a single file, helptext.go.
This while the code for frames, clone and sample remains in the dataset cli it'll be removed in an upcoming release before the transition to v2.4.
Full Changelog: v2.3.0...v2.3.1