To be updated every time code is pushed to master.
- CSS fixes for running in Jupyterlab
- Better handling of waiting for Wrattler to load
- Protection against NULL dataframes in R service
- Add new R packages to base docker image.
- Enable visualization of the dependency graph.
- Button for fullScreen view of cell outputs
- Fixes to R and python services for loading e.g. function definitions from files when running in Binder.
- Changes and fixes to enable interactive maps in Wrattler.
- Bugfixes in client related to adding cells of the same language.
- Enable D3 visualizations.
- Update to logic for getting URLs in Jupyterlab, in order to integrate with Binderhub.
- Bugfixes for R service (avoid writing NULL dataframes to data store)
- Modifications to structure of python service and datastore (convert both into python packages) in preparation for Binderhub integration.
- Change base docker image for client.
- Changes to facilitate html output from python and R services.
- Improvements to Jupyterlab experience, in terms of caching results, not forgetting cell outputs when page is reloaded.
- Improve static source code analysis in Python to detect modified data frames
- Adding more documentation on writing your own plugins and API docs
- Mount
resources
when running Wrattler inside Docker - Improvements in the Arrow data store handling
- Initial work on adding spinner when evaluating code
- Improve evaluation UI: Shift+Enter now evaluates
- Add caching of /eval requests, so we do not revaluate when hashes match
- Added jupyterlab support - jupyterlab runs in a docker container and has a wrattler extension enabling it to open files with a .wrattler extension using wrattler-app.js served by the client.
- Added support for python and R services to read files containing e.g. function definitions or import statements using the
%local <filename>
or%global <filename>
statements. - Logic for re-binding modified - will now re-bind when moving to a new cell.
- Added script
build_release.sh
to simplify building docker images and pushing to dockerhub. - Removed F# code and Fable from the client docker image - now uses node base image rather than dotnet.
- Added
docs
folder withtypedoc
generated Wrattler API documentation
- Added Racket service.
- Datastore is now python Flask app with support for local storage or Azure blob storage.
- R and Python service can use Apache Arrow format for input and output dataframes.
- Various bugfixes and bits of tidying up.
Initial working version, with Python, R, Javascript services.