- Main feature of this release is an API for Kotlin libraries
that simplifies a kernel integration for them (#99).
Special Gradle plugin
is provided for this purpose. It enables the dependency on
kotlin-jupyter-api
artifact which allows a library writer to add a kernel integration by implementing one ofLibraryDefinition
orLibraryDefinitionProvider
interfaces. You may find a minimal example in docs. For more "real-world" example see the integration for Kotlin dataframe library - Add handling clients with
allow_stdin=false
(#124) - Update library descriptors for Lets-Plot, Kravis (#118) and kaliningraph
- Add enhanced support of typed dataframes (
dataframe
library) - Add jDSP library descriptor (#114)
- Add possibility of kernel embedding (#102)
- Provide completion and errors analysis for commands
- Add support for minKernelVersion field in library descriptors
- Formalise kernel versions format and ordering
- Add possibility to provide shutdown hooks in library descriptors (#87)
- Switch to the latest development version of Kotlin 1.5
- Fix completion bug in Notebook client (#113)
- Fix irrelevant error popups in Notebook client (#109)
- Improve and fix parsing of
%use
magic (#110) - Add resolution of transitive dependencies with runtime scope (previously only compile dependencies were resolved)
- Fix "leaking" of kernel stdlib into script classpath (#27)
- Fix added repositories ordering (#107)
- Add parallel testing
- Upgrade Gradle
- Add ktlint check
- Add README generation from library descriptors and magics/commands descriptions
- Distribute
kotlin-jupyter-shared-compiler
artifact which may be used for building scripting compilers with Jupyter dialect, including compilers inside IDEA