scaling.Rmd

# (PART\*) Best practices {.unnumbered}

# Introduction {#scaling-intro .unnumbered}

When you start using Shiny, it'll take you a long time to make even small apps, because you have to learn the fundamentals.
Over time, however, you'll become more comfortable with the basic interface of the package and the key ideas of reactivity, and you'll be able to create larger, more complex applications.
As you start to write larger apps, you'll encounter a new set of challenges: keeping a complex and growing code-base organized, stable, and maintainable.
This will include problems like:

-   "I can't find the code I'm looking for in this huge file."

-   "I haven't worked on this code in 6 months and I'm afraid I'm going to break it if I make any changes."

-   "Someone else started working with me on the application and we keep standing on each others toes."

-   "The app works on my computer but doesn't work on my collaborator's or in production."

In this, the "best practices", part of the book, you'll learn some key concepts and tools from software engineering that will help you overcome these challenges:

-   In Chapter \@ref(best-practices), I'll briefly introduce you to the big ideas of software engineering.

-   In Chapter \@ref(scaling-functions), I'll show you how to extract code out of your Shiny app in to independent apps, and discuss why you might want to do so.

-   In Chapter \@ref(scaling-modules), you'll learn about Shiny's module system, which allows you to extract coupled UI and server code into isolated and reusable components.

-   In Chapter \@ref(scaling-packaging), I'll show you how to turn your app in R package, and motivate why that investment will pay off for bigger apps.

-   In Chapter \@ref(scaling-testing), you'll learn how to turn your existing informal tests into automated test that can easily be re-run whenever your app changes.

-   In Chapter \@ref(performance), you'll learn how to identify and resolve performance bottlenecks in your apps, ensuring they remain speedy even when used by hundreds of users.

Of course you can't learn everything about software engineering in one part of one book, so I'll also point you to good places to learn more.