tabviz supports embedded visualizations within table columns. This section covers the visualization column types and their common features.

Choosing the Right Visualization

Question Recommended Example Use Case
Comparing point estimates with uncertainty? viz_forest() Meta-analysis, treatment effects, regression coefficients
Comparing single values across rows? viz_bar() Scores, counts, percentages
Showing distribution spread and outliers? viz_boxplot() Response distributions, measurement variability
Showing full distribution shape? viz_violin() Survey responses, bimodal data, comparing shapes
Just need a simple inline bar? col_bar() Weight indicators, progress bars
Showing trends over time? col_sparkline() Stock prices, metrics over time
TipWhen to Use What
  • Forest plots are ideal when you have point estimates with confidence/credible intervals and want to visually compare effect sizes
  • Bar charts work best when comparing single numeric values where the magnitude matters
  • Box plots summarize distributions when you care about median, spread, and outliers
  • Violin plots reveal distribution shapes—use when you suspect bimodality or skewness

Visualization Column Types

Focal Visualizations (viz_*)

These column types display data with their own axes and scales, designed for direct visual comparison across rows:

Column Purpose Guide
viz_forest() Forest plots with confidence intervals Forest Plots
viz_bar() Horizontal bar charts with shared axis Distribution Plots
viz_boxplot() Box-and-whisker plots Distribution Plots
viz_violin() Violin/density plots Distribution Plots

Inline Visualizations (col_*)

Compact visualizations that fit within cells without dedicated axes:

Column Purpose Guide
col_bar() Simple inline bars Columns
col_sparkline() Mini line/area/bar charts Columns
NoteNaming Convention
  • viz_* functions create focal visualization columns with their own axes and scales
  • col_* functions create inline cell content (text, numbers, simple graphics)

Common Features

All visualization columns share these capabilities:

  • Axis configuration — Range, ticks, gridlines
  • Interactivity — Hover, tooltips, zoom
  • Theming — Colors adapt to the active theme
  • Export — SVG, PNG, PDF output
TipTheme Effect Colors

When you have multiple effects per row (e.g., multiple bar series, boxplot groups, or forest plot effects), colors are assigned from the theme’s effect_colors palette. Each built-in theme has a curated 5-color palette designed to work well together.

You can customize with set_effect_colors():

web_theme_default() |>
  set_effect_colors(c("#2563eb", "#dc2626", "#16a34a"))

Individual effects can still override theme colors via their color parameter.

Interactivity

NoteInteractive by Default

All tabviz widgets include these interactions out of the box:

  • Hover highlighting on rows
  • Click to select rows
  • Column sorting via header clicks
  • Column resizing via drag
  • Group collapsing via chevron clicks
  • Download button on hover (top-right)

Hover Tooltips

Tooltips are opt-in. Specify fields to show when hovering over markers:

Code 
data(glp1_trials)

glp1_trials |>
  filter(row_type == "data", group == "Main Trials") |>
  head(4) |>
  tabviz(label = "study",
    interaction = web_interaction(tooltip_fields = c("n", "events", "pvalue")),
    footnote = "Hover over markers to see tooltip",
    columns = list(
      viz_forest(point = "hr", lower = "lower", upper = "upper",
                 scale = "log", null_value = 1)
    )
  )

Interaction Controls

web_interaction(

  tooltip_fields = NULL,    # Column names for tooltip
  enable_sort = TRUE,       # Click headers to sort
  enable_collapse = TRUE,   # Click group headers to collapse
  enable_select = TRUE,     # Click rows to select
  enable_hover = TRUE,      # Highlight on hover
  enable_resize = TRUE,     # Drag column borders
  enable_export = TRUE,     # Download button
  enable_themes = "default" # Theme selection menu
)

Interaction Presets

Preset Description
web_interaction() Full interactivity (default)
web_interaction_minimal() Hover only—no sorting/selection
web_interaction_publication() Fully static (for print)
Code 
# Publication-ready static output
tabviz(data, label = "study",
  interaction = web_interaction_publication(),
  columns = list(viz_forest(...))
)

Zoom & Sizing

Control how visualizations fit within their container:

Argument Options Description
zoom 0.5 to 2.0 Initial zoom level (default 1.0)
auto_fit TRUE/FALSE Shrink to fit container if too large (default TRUE)
max_width pixels or NULL Maximum container width
max_height pixels or NULL Maximum container height (enables scrolling)
Code 
# Constrain to specific dimensions
tabviz(data, label = "study",
  max_width = 800,
  max_height = 400,
  columns = list(...)
)

Exporting Visualizations

tabviz supports exporting to SVG, PNG, and PDF.

Using save_plot()

# Create a visualization
p <- tabviz(data, label = "study", columns = list(
  viz_forest(point = "hr", lower = "lower", upper = "upper",
             scale = "log", null_value = 1)
))

# Save as SVG (vector)
save_plot(p, "figure.svg")

# Save as PNG (raster, high-res)
save_plot(p, "figure.png", width = 1200, scale = 2)

# Save as PDF
save_plot(p, "figure.pdf", width = 800)

Output Formats

Format Extension Use Case Dependencies
SVG .svg Vector graphics, web, editing None
PNG .png Documents, slides rsvg package
PDF .pdf Print, publications rsvg package

Browser Export

Interactive plots include a download button (visible on hover in top-right):

  1. Hover over plot to reveal button
  2. Click to open format menu
  3. Choose SVG or PNG

See Also

  • Forest Plots — Confidence intervals with viz_forest()
  • Distribution Plots — Bars, boxplots, and violins with viz_bar(), viz_boxplot(), viz_violin()
  • Columns — All column types including col_bar() and col_sparkline()
  • Export — Advanced export options
  • Shiny Integration — Interactive applications