Extra slides

.title[
# Extra slides
]
.subtitle[
## <br/> Shiny app translation and using learnr
]
.author[
### Megsie Siple
]
.institute[
### NOAA AFSC - RACE/GAP
]
.date[
### 2020/12/7 (updated: 2022-06-15)
]

---

# Translating your Shiny app

For wider utility, you can translate your Shiny app into other languages.

`{shiny.i18n}` (from [Appsilon](https://appsilon.com/internationalization-of-shiny-apps-i18n/) ) provides translation for shiny apps

```r
devtools::install_github("Appsilon/shiny.i18n")

library(shiny)
library(shiny.i18n)

* i18n <- Translator$new(translation_json_path = "translations/translation.json")
```

---

# Translating your shiny app

You can provide either a .json file with all the translations, or a CSV with a column for each language.

The `i18n$t()` wrapper indicates text to be translated, e.g.:

```r
sliderInput("bins",
*            i18n$t("Number of bins:"), # translation wrapper
            min = 1,
            max = 50,
            value = 30)
```

---

class: center, middle
# Supercharge your remote teaching using Shiny tools: `learnr`

---

# Using `learnr`
`{learnr}` uses `runtime: shiny_prerendered` to turn Markdown files into live tutorials.

This is great for writing code in a 'controlled environment'
when you're:
- teaching coding concepts remotely
- demonstrating how a specific package is used
- any situation where you want to remotely set up the environment

---

# Getting started with `learnr`

```r
install.packages("learnr")
library(learnr)
```

To make a new tutorial from the RStudio IDE, go to New File --> R Markdown --> From Template --> Interactive Tutorial {learnr}

---

# `learnr` tutorials - 1
A learnr tutorial is an .Rmd script a few additional features:

1. The output is a `learnr::tutorial`

2. The runtime (in yaml header) is `shiny_prerendered`

3. There is some `learnr`-related setup code

4. There are `knitr` chunks which have custom options such as `exercise` and `solution`

(from Nischal Shrestha's, great blog post [Learning learnr](https://education.rstudio.com/blog/2020/07/learning-learnr/) )
---

# `learnr` tutorials - 2
You load the packages that people will need for the tutorial in the `setup` code chunk.

```r
library(tidyverse)
library(palmerpenguins) #load any packages here that you want to be part of the environment for everyone using the tutorial
library(learnr)

knitr::opts_chunk$set(echo = FALSE)
```

---
# `learnr` tutorials - 3
The `exercise = TRUE` code chunks in a `learnr` document contain exercises, hints, and solutions.

![learnrchunks](img/learnr_chunks.png)
---

# More `learnr` resources

Allison Horst's [post about learnr](https://education.rstudio.com/blog/2020/05/learnr-for-remote/) on RStudio Education

Angela Li's [post on publishing learnr tutorials](https://cran.r-project.org/web/packages/learnr/vignettes/shinyapps-publishing.html)

Desirée de Leon's [post on putting learnr tutorials in a package](https://education.rstudio.com/blog/2020/09/delivering-learnr-tutorials-in-a-package/)

---

# 04_learnr/learnr_basics.Rmd

---

# Shiny app publishing for NOAA affiliates

---

# What if I am a NOAA employee?

If you are building a tool for NOAA purposes, you should:

1. Host it on your NOAA GitHub account if it's a NOAA project. If not, host on your personal account and fork a copy to the NOAA account if necessary
2. Ensure good software design practices, including having documentation related to the code as well as a separate users guide
3.Structure your R code and folders consistent with a accepted best practices
4. Get a code review (this should ideally be someone who also can review the content, if there is a model in your `server.R`)
5. Include testing (and examples if it makes sense-- if it's only a shiny app and not a whole package, it may not be needed). Use a package like `testthat`.  You can also use github actions to automate this in your repository

*Modified from Corinne Bassin's recommendations to me*

---