This PR builds on top of #603 (dataset styling)

It looks like there were some profiling capabilities in the ProcessingProfiler, but it seems limited to logs and I don't know whether there is any tool to visualize it.

Interfaces

I've been thinking a bit about this, and I think I came up with a nice interface that is extremely flexible while also providing very fine grained access to items of interest. It currently lives in the dataset module, but once it's stable I'd like to turn the base into a separate project so it's usable in non-chart projects as well. It consists of 3 interfaces:

a Profiler is something that creates named measurements

interface Profiler {
    DurationMeasure newDuration(String tag);
}

a DurationMeasure can measure time by using start and stop. An appropriate clock is chosen internally, but there is also a fallback to record a raw value for cases where timestamps are passed around (e.g. in event sourcing timestamps are often kept with the events)

interface DurationMeasure {
    void start();
    void stop();

   TimeUnit getClockUnit();
   void recordRawValue(long duration);
}

Anything that needs to provide profiling information implements the Profileable interface

interface Profileable {
    void setProfiler(Profiler profiler);
}

Each metric of interest gets a dedicated measure. The field initialize to a DISABLED implementation with empty methods that should be able to be removed by dead code elimination.

class SpecialAxis implements Profileable {

    public void layoutChildren() {
        benchLayoutChildren .start();
        // do work...
        benchLayoutChildren.stop();
    }

    public void drawAxis() {
        benchDrawAxis.start();
        // do work...
        benchDrawAxis.stop();
    }

    @Override
    public void setProfiler(Profiler profiler) {
        benchLayoutChildren = profiler.newDuration("specialAxis-layoutChildren");
        benchDrawAxis = profiler.newDuration("specialAxis-drawAxis");
    }

    private DurationMeasure benchLayoutChildren = DurationMeasure.DISABLED;
    private DurationMeasurem benchDrawAxis = DurationMeasure.DISABLED;

}

Examples

Debug printer

The simplest profiler is a debug printer that prints start/stop tags and some timing information to a log.

chart.setProfiler(Profiler.printProfiler(System.out::println, boolean printStartedInfo));

chart-runPreLayout - started
chart-lockDataSets - started
chart-lockDataSets - finished (0.01 ms)
chart-updateAxisRange - started
chart-updateAxisRange - finished (0.39 ms)
chart-runPreLayout - finished (0.53 ms)
chart-cssAndLayout - started
chart-layoutChildren - started
chart-layoutChildren - finished (0.34 ms)
chart-cssAndLayout - finished (0.57 ms)
chart-runPostLayout - started
chart-drawAxes - started
chart-drawAxes - finished (0.09 ms)
chart-drawCanvas - started
chart-drawCanvas - finished (1.16 ms)
chart-runPostLayout - finished (1.36 ms)

Tag selection

Via default methods we can create a fluent API for filtering tags of interest and modify the tags with meta-information for better disambiguation.

var profiler = Profiler.debugPrinter(System.out::println);
chart.getXAxis().setProfiler(profiler.addPrefix("x-").contains("computePrefSize"));
chart.getYAxis().setProfiler(profiler.addPrefix("y-").contains("computePrefSize"));

y-axis-computePrefSize - started
y-axis-computePrefSize - finished (8.85 ms)
x-axis-computePrefSize - started
x-axis-computePrefSize - finished (5.67 ms)
y-axis-computePrefSize - started
y-axis-computePrefSize - finished (0.02 ms)
x-axis-computePrefSize - started
x-axis-computePrefSize - finished (0.01 ms)

HdrHistogram output

HdrHistogram is an established and very low-overhead way to measure production behavior in low-latency environments (banking, real-time systems etc.). It records latencies in high-dynamic range histograms (i.e. resolution depends on the absolute value, so it's very space efficient) and provides a file format that can be read by various tools. I've used it a few times, so I had some code sitting around.

The profiler records histograms and continuously writes them to disk in set intervals. The histograms are stored with a tag so they can be disambiguated again later. Profiling an axis and loading it in HdrHistogramVisualizer (a tool I wrote a long time ago when testing the normal JavaFX Charts) looks as below

chart.getXAxis().setProfiler(Profiler.hdrHistogramProfiler("bench/axis.hlog"));

The top chart shows the maxima for each interval, and the lower chart shows a percentile plot with an exponential x axis.

Live chart

Given that this is a charting library, I figured it'd be appropriate to also have a live chart of the real-time generated profiling data:

var profiler = Profiler.chartProfiler("RollingBufferSample");
chart.setProfiler(profiler.matches(".*[run|draw].*"));
chart.getXAxis().setProfiler(profiler.addPrefix("x-").contains("draw"));
chart.getYAxis().setProfiler(profiler.addPrefix("y-").contains("draw"));
beamIntensityRenderer.setProfiler(profiler.addPrefix("beamIntensity-").contains("render"));
dipoleCurrentRenderer.setProfiler(profiler.addPrefix("dipoleCurrent-").contains("render"));

I already had some code for HdrHistogram-like visualizations, so I put together an initial demo of it profiling itself. Here is a video.

Pull Request Review

Hey there! 👋 I've summarized the previous results for you. Here's a markdown document for your pull request review:

Changes

Added a new dependency in the pom.xml file at line 99.
Added new private DurationMeasurement variables in the Chart.java file at lines 105-118.
Added a new method ensureJavaFxPulse() in the Chart.java file at lines 124-137.
Added a new method getCssMetaData() in the Chart.java file at lines 143-148.
Removed the XYChartCss.java file.
Added new imports in the AbstractAxis.java file at lines 6-9.
Implemented the Profileable interface in the AbstractAxis.java file at line 11.
Added new private DurationMeasurement variables in the AbstractAxis.java file at lines 32-35.
Added new method setProfiler() in the AbstractAxis.java file at lines 123-134.
Removed the DataSet parameter from the updateLegend() method in the Legend.java file.

Suggestions

In Chart.java, line 100, change protected final Pane canvasForeground = FXUtils.createUnmanagedPane(); to protected final Pane canvasForeground = StyleUtil.addStyles(new FullSizePane(), "chart-canvas-foreground"); for consistent styling.
In Chart.java, line 164, change protected void invalidated() { to protected void invalidated() { for better code readability.
In Chart.java, line 268, change var canvasArea = StyleUtil.addStyles(new PlotAreaPane(canvas, canvasForeground,...

Bugs

Added a new method updateDataSetIndices() to update the global indices of the datasets.
Modified the rendererChanged() method to call updateDataSetIndices() after adding or removing a renderer.
Modified the datasetsChanged() method to call updateDataSetIndices() after setting the global indices.
Modified the updateLegend() method to accept only the list of renderers instead of both datasets and renderers.
Removed the unused forEachDataSet() method.

Improvements

In Chart.java, line 19, instead of commenting 'Update data ranges', it would be better to use a more descriptive comment like 'Update data ranges and trigger layout updates'.
In Chart.java, line 28, instead of commenting 'Make sure the datasets won't be modified', it would be better to use a more descriptive comment like 'Lock the datasets to prevent modification during layout updates'.
In Chart.java, line 36, instead of commenting 'Update data ranges etc. to trigger anything that might need a layout', it would be better to use a more descriptive comment like 'Update data ranges and trigger layout updates for components that depend on them'.
In Chart.java, line 47, instead of commenting 'Update other components', it would be better to use a more descriptive comment like 'Update other components (renderers, plugins) before layout'.
In Chart.java, line 57, instead of commenting 'nothing to do', it would be better to use a more descriptive comment like 'No layout updates required if the state...

Rating

Please provide a rating from 0 to 10 based on the following criteria:

Readability
Performance
Security

Feel free to add a brief explanation for each criterion.

That's it! Good luck with your pull request! 🚀

By the way, have you heard about our premium plan? It's perfect for analyzing big pull requests like this one. Just thought you might be interested! 😉

Codecov Report

Patch coverage: 42.42% and project coverage change: -0.45% :warning:

Comparison is base (38f34f6) 48.51% compared to head (38c4a47) 48.06%.

:exclamation: Current head 38c4a47 differs from pull request most recent head 34f690b. Consider uploading reports for the commit 34f690b to get more accurate results

Additional details and impacted files

```diff @@ Coverage Diff @@ ## ennerf/dataset-styling #604 +/- ## ============================================================ - Coverage 48.51% 48.06% -0.45% + Complexity 6187 6179 -8 ============================================================ Files 385 397 +12 Lines 37659 38071 +412 Branches 6065 6078 +13 ============================================================ + Hits 18270 18300 +30 - Misses 18224 18616 +392 + Partials 1165 1155 -10 ``` | [Files Changed](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc) | Coverage Δ | | |---|---|---| | [...t/src/main/java/io/fair\_acc/chartfx/axes/Axis.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvYXhlcy9BeGlzLmphdmE=) | `0.00% <ø> (ø)` | | | [.../java/io/fair\_acc/chartfx/plugins/ChartPlugin.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcGx1Z2lucy9DaGFydFBsdWdpbi5qYXZh) | `78.20% <ø> (ø)` | | | [.../java/io/fair\_acc/chartfx/plugins/EditDataSet.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcGx1Z2lucy9FZGl0RGF0YVNldC5qYXZh) | `23.48% <ø> (ø)` | | | [.../io/fair\_acc/chartfx/plugins/UpdateAxisLabels.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcGx1Z2lucy9VcGRhdGVBeGlzTGFiZWxzLmphdmE=) | `0.00% <0.00%> (ø)` | | | [...rtfx/plugins/measurements/DataSetMeasurements.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcGx1Z2lucy9tZWFzdXJlbWVudHMvRGF0YVNldE1lYXN1cmVtZW50cy5qYXZh) | `76.32% <ø> (ø)` | | | [...va/io/fair\_acc/chartfx/profiler/ChartProfiler.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcHJvZmlsZXIvQ2hhcnRQcm9maWxlci5qYXZh) | `0.00% <0.00%> (ø)` | | | [...\_acc/chartfx/profiler/CircularDoubleDataSet2D.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcHJvZmlsZXIvQ2lyY3VsYXJEb3VibGVEYXRhU2V0MkQuamF2YQ==) | `0.00% <0.00%> (ø)` | | | [...fair\_acc/chartfx/profiler/HdrHistogramDataSet.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcHJvZmlsZXIvSGRySGlzdG9ncmFtRGF0YVNldC5qYXZh) | `0.00% <0.00%> (ø)` | | | [...air\_acc/chartfx/profiler/HdrHistogramProfiler.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcHJvZmlsZXIvSGRySGlzdG9ncmFtUHJvZmlsZXIuamF2YQ==) | `0.00% <0.00%> (ø)` | | | [...a/io/fair\_acc/chartfx/profiler/PercentileAxis.java](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc#diff-Y2hhcnRmeC1jaGFydC9zcmMvbWFpbi9qYXZhL2lvL2ZhaXJfYWNjL2NoYXJ0ZngvcHJvZmlsZXIvUGVyY2VudGlsZUF4aXMuamF2YQ==) | `0.00% <0.00%> (ø)` | | | ... and [60 more](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc) | | ... and [3 files with indirect coverage changes](https://app.codecov.io/gh/fair-acc/chart-fx/pull/604/indirect-changes?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=fair-acc)

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

After review I think that the measurements are benchmarks rather than profiling info, so I changed the names to reflect that.

Example

var recorder = MeasurementRecorders.newLiveDisplay("RollingBufferSample").matches(".*(all).*");
beamIntensityRenderer.setRecorder(recorder.contains("all").addPostfix("error"));

The PR also adds includes a BasicDataSetRenderer that is more efficient for rendering standard datasets without errors. It is based on a renderer that we use internally.

There are also a few classes like CircularDoubleDataSet2D and PercentileAxis that are currently package private that should maybe become public at some point.