Introduce new APIs for managing focus and using focus to inform delivery of keyboard events.
Use new APIs to implement the following behavior:
Focus:
- If the component browser is opened, its initial state is *focused*.
- If the node input area's text component is clicked, the component browser's state becomes *blurred*.
- If a click occurs anywhere in the component browser, the component browser's state becomes *focused*.
Event dispatch:
- When the component browser is in the *focused* state, it handles certain keyboard events (chiefly, arrow keys).
- If the component browser handles an event, the event is not received by other components.
- If an event occurs that the component browser doesn't handle, the node input area's text component receives the event.
[vokoscreenNG-2023-06-29_10-55-00.webm](https://github.com/enso-org/enso/assets/1047859/f1d9d07c-8c32-4482-ba32-15b6e4e20ae7)
# Important Notes
Changes to display object interface:
- **`display::Object` can now be derived.**
- Introduce display object *focus receiver* concept. Many components, when receiving focus, should actually be focused indirectly by focusing a descendant.
- For example, when the CB Panel receives focus, its descendant at `self.model().grid.model().grid` should be focused, because that's the underlying Grid View, which has its own event handlers. By allowing each level of the hierarchy to define a `focus_receiver`, focus can reach the right object without the CB panel having to know structural details of its descendants.
- When delegating to a field's `display::Object` implementation, the derived implementation uses the child's `focus_receiver`, which will normally be the correct behavior.
**Changes to `shortcut` API**:
- New `View::focused_shortcuts()` is a focus-aware alternative to `View::default_shortcuts()` (which should now only be used for global shortcuts, i.e. shortcuts that don't depend on whether the component is focused). It's based on the *Keyboard Event* API (see below), so events propagate up the focus hierarchy until a shortcut is executed and `stop_propagation()` is called; this allows sensible resolution of event targets when more than one component is capable of handling the same keypress.
Keypress dataflow overview:
DOM -> KeyboardManager -> FrpKeyboard -> KeyboardEvents -> Shortcut.
Low-level keyboard changes to support Focus:
- New `KeyboardManager`: Attaches DOM event handlers the same way as `MouseManager`.
- New *Keyboard Event* API: `on_event::<KeyDown>()`. Events propagate up the focus hierarchy. This API is used for low-level keyboard listeners such a `Text`, which may need complex logic to determine whether a key is handled (rather than having a closed set of bindings, which can be handled by `shortcut`).
- FRP keyboard: Now attaches to the `KeyboardManager` API. It now serves primarily to produce Keyboard Events (it still performs the role of making `KeyUp` events saner in a couple different ways). The FRP keyboard can also be used directly as a global keyboard, for such things as reacting to modifier state.
Misc:
- Updated the workspace `syn` to version 2. Crates still depending on legacy `syn` now do so through the workspace-level `syn_1` alias.
Fixes performance problems observed when creating/resolving errors (#6674):
|before|after|
|---|---|
|||
This also helps with #6637, although I haven't been able to reproduce the degree of slowness shown there so I can't confirm that this resolves that issue.
# Important Notes
- Disable visualizations until shown. [Faster startup, and all graph changes.]
- 6x faster message deserialization. [Saves 400ms when making a change with many visualizations open.]
- Fast edge recoloring. [Saves 100-150ms when disconnecting an edge in Orders.]
- Add a checked implementation of a `profiler` data structure, used instead of the fast `unsafe` version when `debug-assertions` are enabled.
This PR changes build script's `ide watch` and `ide start` commands, so they don't use `electron-builder` to package. Instead, they invoke `electron` directly, significantly reducing time overhead.
`ide watch` will now start Electron process, while continuously rebuilding gui and the client in the background. Changes can be puilled by reloading within the electron, or closing the electron and letting it start once again. To stop, the script should be interrupted with `Ctrl+C`.
This PR implements HTML generation from documentation IR for all suggestion database entries and replaces the old documentation panel with a newer one.
Additional adjustments to the looks of the documentation would be applied separately in a future PR. This PR focuses on the fastest possible delivery of a usable documentation panel. We want to test it in real-world use cases and gather feedback for future improvements.
Documentation demo scene with mocked data:
https://user-images.githubusercontent.com/6566674/213436313-88753ed8-346f-423e-956e-7db39f5dc266.mp4
Component browser with actual engine-provided data:
https://user-images.githubusercontent.com/6566674/213436375-d0ec074b-f7a6-4deb-a7de-3adee999cc86.mp4
# Important Notes
- Fixed language protocol data structures.
- Scrolling to the selected method is also implemented here.
- Also, the selected item is highlighted with yellow.
- Only some pieces of information we have are displayed. For example, we don't display return types for methods or types of arguments.
- A bunch of code related to previous implementation is removed, but probably not all of it.
Logging: Replace tracing with an efficient logging implementation, with 0-runtime cost for disabled log levels. (https://www.pivotaltracker.com/story/show/183755412)
Profiling: Support submitting `profiler` events to the User Timing Web API, so that measurements can be viewed directly in the browser. (https://www.pivotaltracker.com/story/show/184003550)
# Important Notes
Logging interface:
- The macros (`warn!`, etc.) now take standard `format_args!` arguments (the tracing implementations accepted a broader syntax).
- Compile-time log levels can now be set through the CLI, like so:
`./run ide start --log-level=trace --uncollapsed-log-level=info`
Profiling:
- The hotkey Ctrl+Alt+Shift+P submits all `profiler` events logged since the application was loaded to the Web API, so that they can then be viewed with the browser's developer tools. Note that standard tools are not able to represent async task lifetimes or metadata; this is a convenient interface to a subset of `profiler` data.
- As an alternative interface, a runtime flag enables continuous measurement submission. In the browser it can be set through a URL parameter, like http://localhost:8080/?emit_user_timing_measurements=true. Note that this mode significantly impacts performance.
When running the profiling run-graph and flamegraph demo scenes, if a profile file is not found in the directory served over http, fall back to generating demo data.
Remove a `Symbol`from the `SymbolRegistry` when its `SpriteSystem` is dropped.
This fixes the remaining buffer leak (after #3504) in https://www.pivotaltracker.com/story/show/181943457
# Important Notes
- The `SymbolRegistry` now assigns unique `SymbolId`s, so that we can tell if a `SymbolId` refers to a `Symbol` that has already been unregistered (this shouldn't happen, but it's not statically-obvious that it doesn't, so if it occurs we shouldn't misbehave).
- Also fix a bug in how `buffer_count` was tracked (we were decrementing more than incrementing!).
[ci no changelog needed]
This PR implements a new helper for the future Component Browser - `component_group::multi::Wrapper`. It propagates FRP events from multiple component groups and ensures that only a single component group is focused at all times.
See the updated component group demo scene (console logs shows propagated FRP events from all component groups):
https://user-images.githubusercontent.com/6566674/172359141-8ea6f1ba-e357-4c1b-852a-adb4d5207e03.mp4
- Fixed a `define_endpoints_2!` macro. FRP endpoints for `focus` events weren't connected properly.
- List View now uses an overlay shape to catch mouse events, it allows much easier implementation of `is_header_selected` in the component group.
- Change dev profile settings. Improves build performance; will not affect anything else. Details below.
- Introduce script for benchmarking various incremental builds. Usage is explained in the script comments.
- Add a line to `intervals` showing total main-thread CPU work logged in a profile; this can be used to compare the results of optimizations (I'll be starting a discussion informed by that data separately; this change just enables the tooling to report it).
Define some workflows for batch-mode profiling.
Implemented:
- collapse nodes
- create node
- enter collapsed node
- new project
- open visualization
They can currently be built and run with a command like:
`./run.sh ide build --profiling-level=debug && dist/ide/linux-unpacked/enso --entry-point profile --workflow create_node --save-profile out.json`
And the data can be displayed with:
`dist/ide/linux-unpacked/enso --entry-point profiling_run_graph --load-profile out.json`
Demo of recording and viewing a profile with a command-line one-liner:
https://user-images.githubusercontent.com/1047859/169954795-2d9520ca-84f9-45d2-b83a-5063ebe6f718.mp4
See: https://www.pivotaltracker.com/story/show/182195399.
# Important Notes
- When defining workflows, two helpers are enough to allow us to tell when the action is really done: `Fixture::compile_new_shaders`, and `Fixture::backend_execution`. Often, it is appropriate to await both, but it depends on the task.
- The shader compiler is now driven by a `Controller`; while the `Compiler` is reset if context is lost, the `Controller`'s state survives context loss.
- A new `--load-profile` option supports specifying a profile by path when running `profiling_run_graph`.
- Drop the `with_same_start` profiler interface; we ended up preferring a child profiler convention, and this interface was not implemented compatibly with the stricter data model we've had since the introduction of `profiler::data`.
- Fix the noisy `rustfmt` output.
[ci no changelog needed]
[Task link](https://www.pivotaltracker.com/story/show/181725003)
This PR implements a fully visible component group header while scrolling the group (using the ScrollArea).
The header moves in sync with scrolling movements (using new `set_header_pos` FRP input), so it looks like the component group is scrolled. ScrollArea masks the "scrolled" entries above the header. This design allows a fully visible header even though our renderer doesn't support nested layers masking yet.
The screencast:
https://user-images.githubusercontent.com/6566674/168320360-2c2017b2-0ef5-42ce-9c79-82b9641c1d73.mp4
The most recent one, with the updated demo scene from develop:
https://user-images.githubusercontent.com/6566674/168555268-8552c4b0-f887-4388-89a1-e65ddf668be6.mp4
# Important Notes
- I fixed the API of the list view so now it supports non-hardcoded scene layers (previously it did not). I also believe it was implemented incorrectly.
- I've found a [pretty weird bug](https://www.pivotaltracker.com/story/show/182193824): the component group inside the ScrollArea is invisible unless I add some arbitrary shape to the scroll area content. I use a `transparent_circle` for this purpose in the demo scene. The bug is probably related to masking the sublayers, though I wasn't able to reproduce it properly on a simpler example.
- The selection box is removed from the demo scene as agreed with @farmaazon . The correct implementation has proven to be much harder than I expected, and we will implement another approach in a separate PR.
- I also modified the `shadow::Parameters` so that it uses `Var`s instead of plain values.
* Extends the instrumentation of the code base and upgrades some FRPs to the newer API macro.
* Extends the run-graph demo scene to specify a profile via URL without recompilation.
* Fixes labels in the flame graph demo scene.
* Fixes an issue with loading profiles that contains escaped characters.
# Important Notes
* no longer contains the upgrade of the `text::View` to `define_endpoints_2`. This should be fixed as part of the text rendering rewrite.
[ci no changelog needed]
* Extends the instrumentation of the code base and upgrades some FRPs to the newer API macro.
* Extends the run-graph demo scene to specify a profile via URL without recompilation.
* Fixes labels in the flame graph demo scene.
* Fixes an issue with loading profiles that contains escaped characters.
# Important Notes
[ci no changelog needed]
* New JSON profile format.
* Use string-table optimization for labels in JSON format.
* Use TimeOffset header to render beanpoles
* Log RPC messages sent to the backend.
* Display RPC requests on graph
* Simplify metadata-logging interface.
Implements a visualization that is integrated with our GUI profiling visualization for the multiprocess data implemented in #3395https://user-images.githubusercontent.com/1428930/165915395-c850c7b2-1cc5-4eb0-8f21-37565d113b1e.mp4
The visualization shows a horizontal line for Engine, Language Server and GUI and renders arrows for each message passed between them. Information about the message is revealed on hover.
# Important Notes
* this PR refactors the tooltip mechanism. Note that this has not been in active use anywhere else, as tooltips for node received a custom implementation and the tooltip that was previously implemented was used nowhere else yet.
[ci no changelog needed]
See: [#181837344](https://www.pivotaltracker.com/story/show/181837344).
I've separated this PR from some deeper changes I'm making to the profile format, because the changeset was getting too complex. The new APIs and tools in this PR are fully-implemented, except the profile format is too simplistic--it doesn't currently support headers that are needed to determine the relative timings of events from different processes.
- Adds basic support for profile files containing data collected by multiple processes.
- Implements `api_events_to_profile`, a tool for converting backend message logs (#3392) to the `profiler` format so they can be merged with frontend profiles (currently they can be merged with `cat`, but the next PR will introduce a merge tool).
- Introduces `message_beanpoles`, a simple tool that diagrams timing relationships between frontend and backend messages.
### Important Notes
- All TODOs introduced here will be addressed in the next PR that defines the new format.
- Introduced a new crate, `enso_profiler_enso_data`, to be used by profile consumers that need to refer to Enso application datatypes to interpret metadata.
- Introduced a `ProfileBuilder` abstraction for writing the JSON profile format; partially decouples the runtime event log structures from the format definition.
- Introducing the conversion performed for `ProfilerBuilder` uncovered that the `.._with_same_start!` low-level `profiler` APIs don't currently work; they return `Started<_>` profilers, but that is inconsistent with the stricter data model that I introduced when I implemented `profiler_data`; they need to return profilers in a created, unstarted state. Low-level async profilers have not been a priority, but once #3382 merges we'll have a way to render their data, which will be really useful because async profilers capture *why* we're doing things. I'll bring up scheduling this in the next performance meeting.
Add logging of EnsoGL performance stats to the profiling framework. Also extends the visualization in the debug scene to show an overview of the performance stats. We now render a timeline of blocks that indicate by their colour the rough FPS range we are in:
https://user-images.githubusercontent.com/1428930/162433094-57fbb61a-b502-43bb-8815-b7fc992d3862.mp4
# Important Notes
[ci no changelog needed]
Needs to be merged after https://github.com/enso-org/enso/pull/3382 as it requires some changes about metadata logging from there. That is why this PR is currently still in draft mode and based on that branch.
* Profiling: intervals tool
* devtools profile generator
* docs
* ignore open intervals, because devtools does
* intervals tool: show interval counts
* Re-add a doc comment. Was accidentally deleted.
* Implement review.
* DURATION_FLOOR_MS
* Rename methods that apply a profile's contents to self
* Description of chrome format
* Link to profile.json format documentation is design doc
* Implement profiling-level switch described in design doc
* lint
[ci no changelog needed]
This PR adds a few simple unit tests for GraphEditor, that can be used as an example of native Unit Tests.
Covered:
1. Creating nodes
- By internal API
- By using a TAB shortcut
- By using (+) button
- By dropping edge
2. Connecting two nodes with an edge
Some APIs were extended to allow their testing.
Usage of `glyph::System` in `text/component/area` was disabled by conditional compilation, as this code can't be used in native code due to JS dependencies.
* Profiling application details
Add enough profiling to account for every missed frame during startup.
See https://www.pivotaltracker.com/story/show/181499507
* Build ActiveInterval hierarchy in profiler_data
* update doctests / await_!
* docs/formatting/naming
* more graph modes
* increase WASM size
Due to new render-profile-flamegraph scene. We should remove these from the main release WASM blob one way or another.
* lint
* fix a test
* Organization (feedback)
* Add @wdanilo to Cargo.lock CODEOWNERS
As discussed after my previous PR got stuck waiting for Cargo.lock review.
* fix doctests
* Update docs. Removed a limitation.
* Creating a new node with the (+) button (#3278)
[The Task](https://www.pivotaltracker.com/story/show/180887253)
A new (+) button on the left-bottom corner appeared. It may be clicked to open searcher in the middle of the scene, as an alternative to tab key.
https://user-images.githubusercontent.com/3919101/154514279-7972ed6a-0203-47cb-9a09-82dba948cf2f.mp4
* The window_control_buttons::common was extracted to separate crate `ensogl-component-button` almost without change.
* This includes a severe refactoring of adding nodes in general in the Graph Editor. The whole responsibility of adding new nodes (and starting their editing) was moved to Graph Editor - the Project View only reacts for GE events to show searcher properly.
* The status bar was moved from the bottom-left corner to the middle-top of the scene. It does not collide with (+) button, and plays "notification" role anyway.
* The `interface` debug scene was buggy. The problem was with one expression's span-tree. When I replaced it, the scene works.
* I've removed "new searcher" API, as it is completely outdated.
* I've changed code owners of integration tests to GUI team, as it is the team writing mostly the integration tests (int rust)
* Fix regression #181528359
* Add docs & remove unused function
* Fix & enable native Rust tests
* Fix formatting
Co-authored-by: Adam Obuchowicz <adam.obuchowicz@enso.org>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Add an API to create a flame graph from profiling data. Also adds a demo scene showcasing the functionality that generates some profiling data by measuring dummy function calls and rendering a flame graph for the dummy data (see video for the result).
Not that the functionality is not yet exposed user-facing in the GUI itself, but only as API and demo scene, therefore [ci no changelog needed]
https://user-images.githubusercontent.com/1428930/155118977-ecac0628-777c-48bd-9aa7-30ee6aef1976.mp4
# Important Notes
* Change from the initial design: labels are shown on the flame graph instead of as a tooltip. This is because tooltips are currently only implemented in the graph editor and would require some additional refactoring (probably taking the better part of a day).
* re-instated the behaviour that logs are shown in the JS console if development mode is active.
API for storing metadata.
See: https://www.pivotaltracker.com/story/show/181149277
# Important Notes
**New APIs**:
- Storing metadata is implemented with `profiler::MetadataLogger`.
- A full metadata storage/retrieval example is in [the top-level doctests](https://github.com/enso-org/enso/blob/wip/kw/profiling-metadata-api/lib/rust/profiler/data/src/lib.rs) for profiler::data, a crate which implements an API for profiling data consumers (it abstracts away the low-level details of the event log, and checks its invariants in the process) [after review of this new API here I'll open a PR to add it to the design doc].
**Implementation**:
- `profiler::Event` is parameterized by a metadata type, so that different types of metadata can be dependency-injected into it.
- A data consumer defines its metadata type as an enum of all the kinds of metadata it is interested in.
- Producing the metadata enum is accomplished without defining its type (which would require dependencies from around the app): A `MetadataLogger` internally use a serialization helper `Variant` to serialize its variant of the metadata enum without knowledge of the other possible variants.
**Performance impact**: still in the low ns/measurement range, comparable to pushing to a vec.
*Note*: `LocalVecBuilder` is currently present under the name `Log`, which is accurate but probably too overloaded. I'd like to find the right name for it, document it with examples, and move it to its own crate under data-structures, but I don't want doing that to hold up this PR.
* profiling instrumentation
* Support native testing with mock impl of `mod js`
* Add benchmarks
* Wrapper: support methods.
* `#[profile]`: work in any context
* feature-gate lineno info that breaks IDE
* Support async; more docs; add perf analysis
* docs & formatting
