---
title: "Shiny integration"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{shiny-integration}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>", 
  eval = FALSE
)
library(apexcharter)
```

```{r setup}
library(apexcharter)
```

## Charts

### Create and update (or destroy and re-create)

When a graph has been generated in Shiny, if the values change (via a reactive function), the graph is *not regenerated*, only the data is *updated*. If you have changed specific options in the graphic (such as maximum y axis value, chart's title, ...) these will not be updated. This behavior can be controlled with `auto_update` argument (available in `apexchart()` and `apex()`) :

By default, `auto_update` is `TRUE` :

```{r}
apex(..., auto_update = TRUE)
```

If you want to re-create the whole chart, set the option to `FALSE`:

```{r}
apex(..., auto_update = FALSE)
```

You can also use `config_update()` to specify what to update : 

```{r}
apex(..., auto_update = config_update(update_options = TRUE))
```


### Proxy 

A proxy is also implemented to update charts manually in oberserver. You can update data:

```{r}
output$my_chart <- renderApexchart({
  apex(data = isolate(data_reactive()), ..., auto_update = FALSE)
})

observeEvent(input$update, {
  apexchartProxy("my_chart") %>% 
    ax_proxy_series(data_reactive())
})
```

Be sure to use `shiny::isolate()` to block any reactivity in `renderApexchart` function and to set `auto_update` to FALSE te prevevnt updating twice.
Then you can use in an observe function (or any reactive function) `apexchartProxy()` with the output id to get the chart instance and `ax_proxy_series()` to update data.

If you want to update chart's options, use : 

```{r}
observeEvent(input$update, {
  apexchartProxy("my_chart") %>% 
    ax_proxy_options(list(
      title = list(
        text = "New title"
      ),
      xaxis = list(
        max = NEW_VALUE
      )
    ))
})
```



## Interactions

### Click

Click on a chart to select a data point and retrieve value server side with `set_input_click()` :

```{r, eval=TRUE}
data.frame(
  month = month.abb,
  value = sample(1:100, 12)
) %>% 
  apex(aes(month, value), height = "250px") %>% 
  ax_title("Click a bar:") %>% 
  set_input_click("month_click")
```

Value server-side will be available through `input$month_click`.

Depending on the type of graphic, you can retrieve :

 * **bar and column:** category (x-axis).
 * **pie and donut:** label.
 * **time-series:** retrieve x-axis value, you have to display markers
  with size > 0 and set tooltip's options `intersect = TRUE` and `shared = FALSE`.
 * **scatter:** retrieve XY coordinates.



Multiple slection is possible and you can change the darken effect of selected bars :


```{r, eval=TRUE}
data.frame(
  month = month.abb,
  value = sample(1:100, 12)
) %>% 
  apex(aes(month, value), height = "250px") %>% 
  ax_title("Click several bars:") %>% 
  set_input_click(
    inputId = "month_click", 
    multiple = TRUE, 
    effect_value = 0.1
  )
```



More examples are available with:

```{r}
run_input_demo("click")
```




### Zoom

Retrieve the coordinates of the axes when the graph is zoomed in:

```{r, eval=TRUE}
data("economics", package = "ggplot2")
apex(economics, aes(date, psavert), type = "line", height = "250px") %>% 
  set_input_zoom("zoom_ts")
```

Value server-side will be available through `input$zoom_ts` unde the form :

```{r, echo=FALSE, eval=TRUE}
list(
  x = list(
    min = "1981-10-24 15:41:16 UTC",
    max = "1992-01-24 06:40:22 UTC"
  ),
  y = list(
    min = NULL,
    max = NULL
  )
)
```

Here values for `y` are `NULL` because zoom is only possible on x-axis, but for a scatter chart for example you can zoom on both axis.


More examples are available with:

```{r}
run_input_demo("zoom")
```