Skip to contents

Describe the Runtime Inputs for a Parsed Formula

Source: R/paintr-parse.R
ptr_runtime_input_spec.Rd

Return the input ids consumed by ptr_exec() in a stable, documented data frame. This is the supported low-level discovery helper for tests, tooling, and advanced package authors who need to construct an input list programmatically without hard-coding internal ids in documentation or downstream code.

Usage

ptr_runtime_input_spec(ptr_obj)

Arguments

ptr_obj

A ptr_obj.

Value

A base data.frame with columns input_id, role, layer_name, keyword, param_key, and source_id.

Details

The returned rows follow the current UI/runtime order:

  • placeholder inputs in parsed layer order

  • derived upload dataset-name inputs immediately after each upload placeholder row

  • layer checkbox inputs for every non-ggplot layer, appended in layer order

Raw ids such as "ggplot+3+2" still appear in the returned data, but those ids remain implementation details. Call this helper instead of relying on the exact encoding scheme directly.

Examples

obj <- ptr_parse_formula(
  "ggplot(data = mtcars, aes(x = var, y = var)) + geom_point() + labs(title = text)"
)
ptr_runtime_input_spec(obj)
#>              input_id           role layer_name keyword param_key  source_id
#> 1          ggplot+3+2    placeholder     ggplot     var         x ggplot+3+2
#> 2          ggplot+3+3    placeholder     ggplot     var         y ggplot+3+3
#> 3              labs+2    placeholder       labs    text     title     labs+2
#> 4 geom_point+checkbox layer_checkbox geom_point    <NA>      <NA>       <NA>
#> 5       labs+checkbox layer_checkbox       labs    <NA>      <NA>       <NA>