Pure C++20 plotting API with a gnuplot renderer for figures.

• Palette/color-cycle API (Default, Tab10, Viridis, Grayscale)
• Secondary-axis (y2) series support
• Shared-axis layout controls for multi-panel figures
• Interactive preview script generation (interactive_preview)
• Panel labels + figure caption metadata (panel_labels, caption)
• Confidence bands (add_band)
• Histogram, KDE, ECDF, percentile-band, and heatmap helpers
• Fan-chart and violin uncertainty helpers
• QQ plot, box summary, and confidence ellipse helpers
• Rich legend controls (position, columns, box, opacity, font)
• Tick/format controls (major step, minor count, format strings)
• Text mode selection (Enhanced, Plain, LaTeX toggle)
• Journal-like presets (IEEE_Tran, Nature_1Col, Elsevier_1Col)
• Style profiles (Science, IEEE_Strict, AIAA_Strict, Presentation, DarkPrintSafe, Tufte_Minimal)
• Typed annotations/objects (labels, arrows, rectangles)
• Typed equation/callout annotations (equations, callouts)
• Fluent FigureBuilder API
• Quick-start helpers (make_quick_figure_spec, make_quick_figure, make_quick_axes)
• One-shot publication helper (make_publication_figure)
• Theme JSON save/load (save_theme_json / load_theme_json)
• Data transforms (moving_average, downsample_uniform, autocorrelation)
• Transform pipeline (TransformPipeline, z-score/clip/rolling)
• Statistical model overlays (linear_fit, add_linear_fit_overlay)
• CSV ingestion and unit-aware labels (read_csv_numeric, label_with_unit)
• Faceting helpers (facet_grid, apply_facet_axes)
• Composition helpers (apply_panel_titles, apply_shared_legend, apply_shared_colorbar_label)
• Auto legend placement heuristic (auto_place_legend)
• YAML declarative spec loading (load_yaml_figure_spec)
• Template gallery and quick templates (apply_plot_template, write_template_gallery_yaml)
• DataTable column plotting adapters (add_line, add_scatter)
• Versioned theme presets (ThemePreset::*_v1)
• CLI renderer (gnuplotpp_cli --spec ... --out ...)
• Font fallback chain for cross-format typography consistency
• Reproducibility manifest export (manifest.json)

• CMake >= 3.20
• C++20 compiler
• gnuplot (required for final figure generation)
• yaml-cpp (required for YAML spec loading)

Install gnuplot:

# macOS
brew install gnuplot

# Ubuntu/Debian
sudo apt-get update && sudo apt-get install -y gnuplot libyaml-cpp-dev

# Fedora
sudo dnf install -y gnuplot yaml-cpp-devel

cmake --preset dev-debug
cmake --build --preset build-debug
ctest --preset test-debug

• Documentation Index
• API Contracts
• Getting Started
• Best Practices Quick Reference
• Styling and Presets
• Statistical Plot Guide
• Examples Cookbook
• Troubleshooting
• Full Plot Controls

Refresh all README figure assets from current examples:

cmake --build --preset build-debug --target refresh-readme-assets
# or:
./scripts/refresh_readme_assets.sh

flowchart LR
  subgraph Inputs
    Y[YAML Spec]
    C[CSV / Arrays]
    CPP[C++ API]
  end

  Y --> CLI[gnuplotpp_cli]
  C --> CLI
  CPP --> Core[FigureSpec / AxesSpec / SeriesSpec]
  CLI --> Core

  Core --> Ops[Transforms + Model Overlays + Statistics]
  Ops --> Fig[Figure / Axes]

  Fig --> Backend[IPlotBackend]
  Backend --> Gnuplot[GnuplotBackend]

  Gnuplot --> Dat[tmp/*.dat]
  Gnuplot --> Script[tmp/figure.gp]
  Gnuplot --> Out[figure.pdf / figure.svg / figure.eps / figure.png]

1. Configure a publication preset (IEEE_SingleColumn, IEEE_DoubleColumn, AIAA_Column, AIAA_Page).
2. Set output formats to vector (Pdf, Svg, optional Eps).
3. Render via GnuplotBackend from your C++ executable.
4. Use PDF as primary submission asset.

• Cairo terminals with enhanced text
• Explicit line widths and rounded joins
• Tick formatting and outward tics
• Grid styling suitable for print
• Escaped plot labels/titles for robust script generation

When using IEEE_SingleColumn or IEEE_DoubleColumn, renderer behavior is tightened to print-safe defaults:

• 8.5 pt Times-style text defaults
• monochrome plotting by default (set monochrome)
• dashed line-style differentiation per series (dt 1..N)
• per-series color/opacity overrides disable global monochrome automatically
• SciencePlots-like axis defaults: visible minor ticks, thin 0.5 axis/grid strokes
• vector-first outputs (PDF, SVG, EPS)

./build/dev-debug/two_window_example --out out/two_window
./build/dev-debug/layout_2x2_example --out out/layout_2x2
./build/dev-debug/three_line_ieee_example --out out/three_line_ieee
./build/dev-debug/monte_carlo_alpha_example --out out/monte_carlo_alpha
./build/dev-debug/feature_rich_showcase --out out/feature_rich_showcase
./build/dev-debug/interactive_facet_example --out out/interactive_facet_example
./build/dev-debug/yaml_spec_example --out out/yaml_spec_example
./build/dev-debug/gnuplotpp_cli --spec examples/specs/minimal.yaml --out out/cli_run
./build/dev-debug/stats_plot_examples --out out/stats_plot_examples
./build/dev-debug/tufte_minimal_example --out out/tufte_minimal_example

Generated from:

./build/dev-debug/three_line_ieee_example --out out/three_line_ieee_readme

Generated from:

./build/dev-debug/stats_plot_examples --out out/stats_plot_examples

Normal Q-Q:

Violin with raw samples:

Confidence ellipse:

Autocorrelation:

Generated from:

./build/dev-debug/tufte_minimal_example --out out/tufte_minimal_example

Direct labels, minimal framing, and low-ink reference guides:

Generated from:

./build/dev-debug/monte_carlo_alpha_example --out out/monte_carlo_alpha_example --npaths 1000 --lw 2 --alpha 0.3

Expected outputs:

• out/<name>/figures/figure.pdf
• out/<name>/figures/figure.svg
• out/<name>/figures/figure.eps
• out/<name>/figures/tmp/figure.gp
• out/<name>/figures/tmp/ax*_series*.dat

Detailed controls and recipes: Plot Controls

RenderResult.status values:

• Success
• InvalidInput
• IoError
• ExternalToolMissing
• ExternalToolFailure
• UnsupportedFormat