Histogram

A histogram is a BarChart with touching bars, one per bin. make_histogram_chart() bins raw samples and returns an ordinary chart, so titles, colors, data labels and every export format (including native vector SVG/PDF) work as usual.

#include "insightdock/charts.h"
using namespace insightdock::charts;

std::vector<double> latencies = /* thousands of samples */;

HistogramOptions opts;
opts.bins = 20;                       // 0 (default) = Sturges' rule: 1 + log2(n), at most 100
BarChart hist = make_histogram_chart(latencies, opts, "Requests");
hist.options().title = "Latency distribution (ms)";
hist.options().legend.enabled = true;  // off by default (single series)
ChartExporter::save(hist, "latency.png");

Binning only

compute_histogram() returns the edges and counts without building a chart — useful for feeding another chart type or your own table:

Histogram h = compute_histogram(latencies, opts);
// h.edges.size() == h.counts.size() + 1

HistogramOptions

Field	Default	Purpose
bins	0	Number of equal-width bins; 0 chooses automatically.
min_value, max_value	0, 0	Fixed range when min_value < max_value; samples outside it are ignored. Otherwise the data's own min/max.

NaN and Inf samples are skipped. The last bin includes its upper edge. Bars are labelled lo-hi; on a dense chart the usual label skipping keeps the axis readable (see x-axis label modes).