enso/gui/CHANGELOG.md

Next Release

Visual Environment

• Statusbar reports connectivity issues. IDE maintains a connection to the Language Server. If this connection is lost, any unsaved and further work will be lost. In this build we added notification in statusbar to signalize that the connection was lost and IDE must be restarted. In future IDE will try to automatically reconnect.
  • Visualization can be extended to the whole screen by selecting the node and pressing space twice. To quit this view, press space again.
• Visualization preview on output port hover. There is now a quick preview for visualizations and error descriptions. Hovering a node output will first show a tooltip with the type information and then after some time, will show the visualization of the node. The preview visualization will be located above other nodes, whereas the normal view, will be shown below nodes. Errors will show the preview visualization immediately. Nodes without type information will also show the visualization immediately. You can enter a quick preview mode by pressing ctrl (or command on macOS), which will show the preview visualization immediately for any hovered node output.
• Database Visualizations. Visualizations for the Database library have been added. The Table visualization now automatically executes the underlying query to display its results in a table. In addition, the SQL Query visualization allows the user to see the query that is going to be run against the database.
• Histogram and Scatter Plot now support Dataframes. The Table and Column datatypes are properly visualized. Scatter Plot can display points of different color, shape and size, all as defined by the data within the Table.
• [onHide callback for visualizations][1383]. JavaScript visualizations can now implement a method onHide()` that will be called whenever the visualization is hidden or closed on screen.

Visual Environment

• Not adding spurious imports. Fixed cases when IDE was adding unnecessary library imports when selecting hints from node searcher. This makes the generated textual code easier to read and reduces likelihood of accidental name collision.
• [Hovering over an output port shows a pop-up with the result type of a node] 1312. This allows discovering the result type of a node which can help with debugging and development.
• Visualizations can define context for preprocessor evaluation. Users can now decide what module's context should be used for visualization preprocessor. This allows providing visualization with standard library functionalities or defining utilities that are shared between multiple visualizations.
• Fix issue with multiple instances of the IDE running. This fixes an issue where multiple instances of the IDE (or even other applications) could lead to the IDE not working.
• Allow JS to log arbitrary objects. Previously using console.log in a visualisation or during development would crash the IDE. Now it correctly logs the string representation of the object.
• Fix mouse cursor offset on systems with fractional display scaling. Instead of there being an offset between visible cursor and cursor selection this works now with any display scaling.
• Disable area selection. The area selection was visible despite being non-functional. To avoid confusion, area selection has been disabled until it is correctly implemented.
• Fix error after adding a node. Sometimes, after picking a suggestion the inserted node was annotated with "The name could not be found" error.
• Handle syntax errors in custom-defined visualizations. The IDE is now able to run properly, even if some of the custom-defined visualisations inside a project contain syntax errors.
• Fix issues with pasting multi-line text into single-line text fields. The first copied line will be inserted and all additional lines will be ignored.
• Users can opt out of anonymous data gathering. This can be done with the --no-data-gathering command-line flag during the startup of the IDE.
• Provide a theming API for JavaScript visualizations. It is now possible to use the Enso theming engine while developing custom visualizations in JavaScript. You can query it for all IDE colors, including the colors used to represent types.
• You can now start the IDE service without window again. The command line argument --no-window now starts all the required backend services again, and prints the port on the command line, allowing you to open the IDE with a browser of your choice.
• JS visualizations have consistent gestures with the IDE. Panning and zooming now works just as expected on trackpad and mouse.
• Running watch command works on first try.. Running the build command run watch would fail if it was run as the first command on a clean repository. This now works.
• [The inputType field of visualizations is actually taken into consideration] 1384. The visualization chooser shows entries accepting the node's output's type only.
• Fix applying selected node output to the expression of new node. For example, having selected node with Table output and adding a new node with expression at "x" == "y" the selected node was applied to the right side of ==: at "x" == operator1."y" instead of operator1.at "x" == "y".
• Enso_Project.data is visible in searcher.
• Geo Map visualization recognizes columns regardless of their name letter case. This allows visualizing tables with columns like LONGITUDE or Longitude, where previously only longitude was recognized.
• It is possible now to switch themes. Also, theme manager was integrated with the FRP event engine, which was a long-standing issue in the IDE. Themes management was exposed to JavaScript with the window.theme variable. It is even possible to change and develop themes live by editing theme variables directly in the Chrome Inspector. Use the following command to start: theme.snapshot("t1"); theme.get("t1").interactiveMode().
• The active visualization is highlighted. Now it is clearly visible when the mouse events are passed to the visualization.

EnsoGL (rendering engine)

• Unified shadow generation. Added a toolset to create shadows for
  arbitrary UI components.

Enso Compiler

If you're interested in the enhancements and fixes made to the Enso compiler, you can find their release notes here.

Enso 2.0.0-alpha.2 (2020-03-04)

This is a release focused on bug-fixing, stability, and performance. It improves the performance of workflows and visualizations, and improves the look and feel of the graphical interface. In addition, the graphical interface now informs the users about errors and where they originate.

• Learn how to define custom data visualizations in Enso.
• Learn how to use Java libraries in Enso, to build a webserver.
• Learn how to use Javascript libraries in Enso, to build custom server-side website rendering.
• Discover why Enso Compiler is so fast and how it was built to support a dual-representation language.
• Learn more about the vision behind Enso and about its planned future.

Visual Environment

• Errors in workflows are now displayed in the graphical interface. Previously, these errors were silently skipped, which was non-intuitive and hard to understand. Now, the IDE displays both dataflow errors and panics in a nice and descriptive fashion.
• Added geographic map support for Tables (data frames). Tables that have latitude, longitude, and optionally label columns can now be shown as points on a map.
• Added a shortcut for live reloading of visualization files. This drastically improves how quickly new visualizations can be tested during their development. This is currently limited in that, after reloading visualization definitions, the currently visible visualizations must be switched to another and switched back to refresh their content. See the video podcast about building custom visualizations to learn more.
• Added a visual indicator of the ongoing standard library compilation. Currently, each time IDE is started, the backend needs to compile the standard library before it can provide IDE with type information and values. Because of that, not all functionalities are ready to work directly after starting the IDE. Now, there is a visible indication of the ongoing background process.
• Added the ability to reposition visualisations. There is now an icon in the visualization action bar that allows dragging the visualization away from a node. Once the visualization has been moved, another icon appears that can pin the visualization back to the node.
• There is now an API to show Version Control System (like Git) status for nodes.

Visual Environment

• You can now use the table visualization to display data frames. Please note, that large tables will get truncated to 2000 entries. This limitation will be lifted in future releases.
• Performance improvements during visual workflow. Nodes added with the searcher will have their values automatically assigned to newly generated variables, which allows the Enso Engine to cache intermediate values and hence improve visualization performance.
• Minor documentation rendering fixes. Fixed cases where text would be misinterpreted as a tag, added support for new tag types, added support for more common characters, properly renders overflowing text.
• Improved handling of projects created with other IDE versions. The IDE is now better at dealing with incompatible metadata in files, which stores node visual position information, the history of chosen searcher suggestions, etc. This will allow IDE to correctly open projects that were created using a different IDE version and prevent unnecessary loss of metadata.
• Pressing and holding up and down arrow keys make the list view selection move continuously.
• The shortcuts to close the application and to toggle the developer tools at runtime now work on all supported platforms.
• The loading progress indicator remains visible while IDE initializes. Previously the loading progress indicator completed too quickly and stopped spinning before the IDE was ready. Now it stays active, giving a visual indication that the initialization is still in progress.
• Fixed visual glitch where a node's text was displayed as white on a white background. Most notably this occurred with the output node of a function generated using the node collapse refactoring.
• Many visual glitches were fixed, including small "pixel-like" artifacts appearing on the screen.
• Several parser improvements. The parser used in the IDE has been updated to the latest version. This resolves several issues with language constructs like import, lambdas, and parentheses, whereupon typing certain text the edit could be automatically reverted.
• The auto-import functionality was improved. Libraries' Main modules are omitted in expressions inserted by the searcher. For example, the point method of Geo library will be displayed as Geo.point and will insert import Geo instead of Geo.Main.
• Cursors in text editors behave correctly now (they are not affected by scene pan and zoom). This was possible because of the new multi-camera management system implemented in EnsoGL.
• Fixed method names highlighted in pink. There was a bug introduced after one of the latest Engine updates, that sent Unresolved_symbol types, which made all methods pink. This is fixed now.

EnsoGL (rendering engine)

• A new multi-camera management system, allowing the same shape systems to be rendered on different layers from different cameras. The implementation automatically caches the same shape system definitions per scene layer in order to minimize the amount of WebGL draw calls and hence improve performance.
• A new depth-ordering mechanism for symbols and shapes. It is now possible to define depth order dependencies between symbols, shapes, and shape systems.
• Various performance improvements, especially for the text rendering engine.
• Display objects handle visibility correctly now. Display objects are not visible by default and need to be attached to a visible parent to be shown on the screen.

Enso Compiler

If you're interested in the enhancements and fixes made to the Enso compiler, you can find their release notes here.

Enso 2.0.0-alpha.1 (2020-01-26)

This is the first release of Enso, a general-purpose programming language and environment for interactive data processing. It is a tool that spans the entire stack, going from high-level visualization and communication to the nitty-gritty of backend services, all in a single language.

Anonymous Data Collection

Please note that this release collects anonymous usage data which will be used to improve Enso and prepare it for a stable release. We will switch to opt-in data collection in stable version releases. The usage data will not contain your code (expressions above nodes), however, reported errors may contain brief snippets of out of context code that specifically leads to the error, like "the method 'foo' does not exist on Number". The following data will be collected:

• Session length.
• Graph editing events (node create, dele, position change, connect, disconnect, collapse, edit start, edit end). This will not include any information about node expressions used.
• Navigation events (camera movement, scope change).
• Visualization events (visualization open, close, switch). This will not include any information about the displayed data nor the rendered visualization itself.
• Project management events (project open, close, rename).
• Errors (IDE crashes, WASM panics, Project Manager errors, Language Server errors, Compiler errors).
• Performance statistics (minimum, maximum, average GUI refresh rate).