Update examples.md

[Quafadas]

I think I've written this method a few times now, and keep having to scratch my head each time :-)...

So here it is - no pressure to merge if it isn't idiomatic :-).

[zetashift]

Actually I've needed this several times as well, thank you! :P. I like the pipe approach more personally.

I would like the examples to be a bit more "hey why you would do it and this is how" style, so I would phrase it more like this:

## Writing data to a CSV file

If you want to save a list of a case class into a CSV file, you could do the following:
@:select(scala-version)

@:choice(scala-3)

\```scala mdoc:reset:silent
import fs2.*
import fs2.data.csv.*
import fs2.io.file.Path

// create an encoder for your case class, by creating a `CsvRowEncoder[YourCaseClass, String]`
given CsvRowEncoder[YourCaseClass, String] = deriveCsvRowEncoder

def writeCaseClassToCsv[A](path: Path)(using CsvRowEncoder[A, String]): Pipe[IO, A, Nothing] =
    _.through(encodeUsingFirstHeaders(fullRows = true))
      .through(text.utf8Encode)
      .through(Files[IO].writeAll(path))
      .drain      
\```

@:choice(scala-2)

\```scala mdoc:reset:silent
import fs2._
import fs2.data.csv._
import fs2.io.file.Path

// create an encoder for your case class, by creating a `CsvRowEncoder[YourCaseClass, String]`
implicit val csvRowEncoder: CsvRowEncoder[YourCaseClass, String] = deriveCsvRowEncoder

def writeCaseClassToCsv[A](path: Path)(using CsvRowEncoder[A, String]): Pipe[IO, A, Nothing] =
    _.through(encodeUsingFirstHeaders(fullRows = true))
      .through(text.utf8Encode)
      .through(Files[IO].writeAll(path))
      .drain      

\```

[Quafadas]

With the benefit of hinsight, my original PR was rushed, and perhaps a little sloppy. Will tidy with the suggestions above - thanks for feedback - agreed on all points. Will revise in the next days when I have a few mins at keyboard to devote to it!

[Quafadas]

Thanks for all the feedback - incorporated and hopefully mergable now :-)

[zetashift]

Just wanted to say that even if you might think this PR looked rushed, it still helps a lot with just getting the ball rolling. And that is always valuable (especially in open source).

edit: fml github ui bamboozled me, I thought I submitted my suggestions, anyhow see blow

[zetashift]

@Quafadas just the last 2 suggestions and besides that it LGTM!

[Quafadas]

@zetashift I thought I already approved and committed them? Ah... I fear however, that I did not press the "resolve conversation" button ... gonna do that right now.

[Quafadas]

@zetashift Oh dear, I honestly thought I'd done them. I've had another go.

[zetashift]

Nice! Thank you for this :D

[TonioGela]

I've added a couple of comments in the code ;)
Thanks for your contribution!

[zetashift]

Here's a runnable example:

//> using toolkit typelevel:latest
//> using scala 3.3.1
import fs2.io.file.Files

import fs2.data.csv.*
import fs2.data.csv.generic.semiauto.*
import fs2.io.file.Path
import cats.effect.{IO, IOApp}
import fs2.{Pipe, Stream}
import java.time.LocalDate

/** Define your case class and derive an encoder for it */
case class YourCaseClass(n: String, i: Int)
given CsvRowEncoder[YourCaseClass, String] = deriveCsvRowEncoder

/** This is our helper function that writes a case class as a csv given a path. */
def writeCaseClassToCsv[A](
    path: Path
)(using CsvRowEncoder[A, String]): Pipe[IO, A, Nothing] =
  _.through(encodeUsingFirstHeaders(fullRows = true))
    .through(fs2.text.utf8.encode)
    .through(Files[IO].writeAll(path))
    .drain

/** Let's imagine we have a `Book` case class we would like to write to
  * a .csv file.
  */
object WriteBooksToCsv extends IOApp.Simple:
  case class Book(id: Long, name: String, isbn: String)
  given CsvRowEncoder[Book, String] = deriveCsvRowEncoder

  val input = Seq(
    Book(1, "Programming in Scala", "9780997148008"),
    Book(2, "Hands-on Scala Programming", "9798387677205"),
    Book(3, "Functional Programming in Scala", "9781617299582")
  )

  def run: IO[Unit] =
    Stream
      .emits(input)
      .through(writeCaseClassToCsv(Path("books.csv")))
      .compile
      .drain *> IO.println("Finished writing books to books.csv.")

@Quafadas if you don't have time to update the PR, is it okay if I add this in? Would hate for this to sit too long.

[TonioGela]

That's scala 3 only, can you add the scala 2 tab too?

[zetashift]

Continued here: #120 because completing this PR through Quafadas fork is getting too slow. Still @Quafadas thank you!!