---
title: "process_data()"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{02 process_data()}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(
  collapse  = TRUE,
  comment   = "#>",
  message   = FALSE,
  warning   = FALSE
)
library(ggpop)
library(dplyr)
```

`process_data()` converts a count-based data frame to one row per icon.
`group_var` and `sum_var` are unquoted (tidy eval); `high_group_var` takes a
character string.

```{r basic}
df_raw <- data.frame(
  sex = c("Female", "Male"),
  n   = c(55, 45)
)

df_plot <- process_data(
  data        = df_raw,
  group_var   = sex,
  sum_var     = n,
  sample_size = 20
)

head(df_plot, 4)
```

The result has one row per icon. The `type` column carries the original group
label and is what you map to `color` and `icon` in `geom_pop()`.

## With `high_group_var`

Use `high_group_var` when your data has a faceting variable. It samples
independently within each panel so proportions are correct per group.

```{r facet}
df_region_raw <- data.frame(
  region = c("North", "North", "South", "South"),
  sex    = c("Female", "Male", "Female", "Male"),
  n      = c(30, 20, 25, 25)
)

df_region <- process_data(
  data           = df_region_raw,
  group_var      = sex,
  sum_var        = n,
  sample_size    = 20,
  high_group_var = "region"
)

head(df_region, 4)
```

The `group` column in the output contains the panel label — pass it to
`facet_wrap(~group)` and `geom_pop(facet = group)`.