Reflex Platform is a curated package set and set of tools that let you build Haskell packages so they can run on a variety of platforms. Reflex Platform is built on top of the nix package manager.
There are five main reasons to use Reflex Platform:
-
It's curated: the core packages in Reflex Platform are known to work together and are tested together.
-
It's cached: the core packages in Reflex Platform are cached so you can download prebuilt binaries from the public cache instead of building from scratch.
-
It's consistent: nix locks down dependencies even outside the Haskell ecosystem (e.g., versions of C libraries that the Haskell code depends on), so you get completely reproducible builds.
-
It's cross-platform: Reflex Platform is designed to target iOS and Android on mobile, JavaScript on the web, and Linux and macOS on desktop. It's Haskell, everywhere.
-
It's convenient: Reflex Platform comes packaged with tools to make development easier, like a hoogle server that you can run locally to look up definitions.
To get started with Reflex development, follow the instructions below.
Try Reflex lets you set up an environment from which you can use Reflex with GHC or GHCJS.
To use Reflex Platform as a build/development system for your own projects, refer to HACKING.md
.
To see what platforms and build targets we support, refer to platform-support.md
To see what has changed since a previous version of Reflex Platform, see ChangeLog.md
.
The default branch for this git repo is develop
, but the latest release is master
.
End-users should prefer to use master
, not develop
.
The default branch is just develop
to ensure new PRs go to the right place by default.
If you're using one of these platforms, please take a look at notes before you begin:
If you encounter any problems that may be specific to your platform, please submit an issue or pull request so that we can add a note for future users.
Reflex Platform will not work on Windows because we rely on Nix to define and construct our environment. You may have some success by using the Windows Subsystem for Linux, but we do not provide support for this platform.
GHCJS uses a lot of memory during compilation. 16GB of memory is recommended, with 8GB being pretty close to bare minimum.
This process will install the Nix package manager. If you prefer to install Nix yourself, you may do so any time prior to step 2.
-
Clone this repository:
git clone https://github.com/reflex-frp/reflex-platform
-
Navigate into the
reflex-platform
folder and run thetry-reflex
command. This will install Nix, if you don't have it already, and use it to wrangle all the dependencies you'll need and drop you in an environment from which you can use Reflex. Be warned, this might take a little while the first time (but it shouldn't take more than a few minutes, if your binary cache is configured properly):cd reflex-platform ./try-reflex
-
From this nix-shell, you can compile any haskell source files you like. Replace
your-source-file.hs
with the name of the file you'd like to compile. For the most part, ghcjs supports the same options as ghc:-
GHC
ghc --make your-source-file.hs ./your-source-file
Compilation will produce a
your-source-file
native executable via WebkitGtk. Simply run it to launch your app. Developer tools are available viaInspect Element
in the right-click context menu. -
GHCJS
ghcjs --make your-source-file.hs
Compilation will produce a
your-source-file.jsexe
folder containing anindex.html
file. Open that in your browser to run your app.
-
Don't use cabal install
to install libraries while inside the try-reflex shell - the resulting libraries may not be found properly by ghc or ghcjs.
Using Cabal to configure, build, test, and run a particular package, however, should work just fine.
try-reflex
and ghcjs --make
are not recommended for real-world projects — just as a quick and easy way to install Nix and experiment with reflex-dom
.
If you need to use additional Haskell libraries (e.g. from Hackage), we recommend using the tools described in project-development.rst instead.
If you've already set up nix, haddock documentation for the versions pinned by your current reflex-plaftorm can be browsed by running
./scripts/docs-for reflex
./scripts/docs-for reflex-dom
In this example, we'll be following Luite Stegemann's lead and building a simple functional reactive calculator to be used in a web browser.
Reflex's companion library, Reflex-DOM, contains a number of functions used to build and interact with the Document Object Model. Let's start by getting a basic app up and running.
> {-# LANGUAGE OverloadedStrings #-}
> import Reflex.Dom
> main = mainWidget $ el "div" $ text "Welcome to Reflex"
Save this file as source.hs
and compile it by running ghcjs source.hs
.
If you've entered everything correctly, this will produce a folder named source.jsexe
in the same directory as source.hs
.
Navigate to this folder in your file manager and open index.html
using your browser.
The browser should show a page with the text "Welcome to Reflex".
Most Reflex apps will start the same way: a call to mainWidget
with a starting Widget
.
A Widget
is some DOM wrapped up for easy use with Reflex.
In our example, we are building the argument to mainWidget
, (in other words, our starting Widget
) on the same line.
el
has the type signature:
el :: DomBuilder t m => Text -> m a -> m a
The first argument to el
is a Text
, which will become the tag of the html element produced.
The second argument is a Widget
, which will become the child of the element being produced.
We turned on the OverloadedStrings
extension so that the literal string in our source file would be interpreted as the appropriate type (Text
rather than String
).
FRP-enabled datatypes in Reflex take an argument
t
, which identifies the FRP subsystem being used. This ensures that wires don't get crossed if a single program uses Reflex in multiple different contexts. You can think oft
as identifying a particular "timeline" of the FRP system. Because most simple programs will only deal with a single timeline, we won't revisit thet
parameters in this tutorial. As long as you make sure yourEvent
,Behavior
, andDynamic
values all get theirt
argument, it'll work itself out.
In our example, el "div" $ text "Welcome to Reflex"
, the first argument to el
was "div"
, indicating that we are going to produce a div element.
The second argument to el
was text "Welcome to Reflex"
.
The type signature of text
is:
text :: DomBuilder t m => Text -> m ()
text
takes a Text
and produces a Widget
.
The Text
becomes a text DOM node in the parent element of the text
.
Of course, instead of a Text
, we could have used el
here as well to continue building arbitrarily complex DOM.
For instance, if we wanted to make a unordered list:
> {-# LANGUAGE OverloadedStrings #-}
> import Reflex.Dom
> main = mainWidget $ el "div" $ do
> el "p" $ text "Reflex is:"
> el "ul" $ do
> el "li" $ text "Efficient"
> el "li" $ text "Higher-order"
> el "li" $ text "Glitch-free"
Of course, we want to do more than just view a static webpage. Let's start by getting some user input and printing it.
> {-# LANGUAGE OverloadedStrings #-}
> import Reflex.Dom
> main = mainWidget $ el "div" $ do
> t <- inputElement def
> dynText $ _inputElement_value t
Running this in your browser, you'll see that it produces a div
containing an input
element.
When you type into the input
element, the text you enter appears inside the div as well.
inputElement
is a function with the following type:
inputElement :: DomBuilder t m
=> InputElementConfig er t (DomBuilderSpace m)
-> m (InputElement er (DomBuilderSpace m) t)
It takes a InputElementConfig
(given a default value in our example), and produces a Widget
whose result is a InputElement
.
The InputElement
exposes the following functionality:
data InputElement er d t
= InputElement { _inputElement_value :: Dynamic t Text
, _inputElement_checked :: Dynamic t Bool
, _inputElement_checkedChange :: Event t Bool
, _inputElement_input :: Event t Text
, _inputElement_hasFocus :: Dynamic t Bool
, _inputElement_element :: Element er d t
, _inputElement_raw :: RawInputElement d
, _inputElement_files :: Dynamic t [RawFile d]
}
Here we are using _inputElement_value
to access the Dynamic Text
value of the InputElement
.
Conveniently, dynText
takes a Dynamic Text
and displays it.
It is the dynamic version of text
.
A calculator was promised, I know. We'll start building the calculator by creating an input for numbers.
> {-# LANGUAGE OverloadedStrings #-}
> import Reflex
> import Reflex.Dom
> import Data.Map (Map)
> import qualified Data.Map as Map
> main = mainWidget $ el "div" $ do
> t <- inputElement $ def
> & inputElementConfig_initialValue .~ "0"
> & inputElementConfig_elementConfig . elementConfig_initialAttributes .~ ("type" =: "number")
> dynText $ _inputElement_value t
The code above overrides some of the default values of the InputElementConfig
.
We provide a Map Text Text
value for the inputElementConfig_elementConfig
's elementConfig_initialAttributes
, specifying the html input element's type
attribute to number
.
Next, we override the default initial value of the InputElement
.
We gave it "0"
.
Even though we're making an html input
element with the attribute type=number
, the result is still a Text
.
We'll convert this later.
Let's do more than just take the input value and print it out. First, let's make sure the input is actually a number:
> {-# LANGUAGE OverloadedStrings #-}
> import Reflex.Dom
> import Data.Map (Map)
> import qualified Data.Map as Map
> import Data.Text (pack, unpack)
> import Text.Read (readMaybe)
> main = mainWidget $ el "div" $ do
> x <- numberInput
> let numberString = fmap (pack . show) x
> dynText numberString
> numberInput :: DomBuilder t m => m (Dynamic t (Maybe Double))
> numberInput = do
> n <- inputElement $ def
> & inputElementConfig_initialValue .~ "0"
> & inputElementConfig_elementConfig . elementConfig_initialAttributes .~ ("type" =: "number")
> return . fmap (readMaybe . unpack) $ _inputElement_value n
We've defined a function numberInput
that both handles the creation of the InputElement
and reads its value.
Recall that _inputElement_value
gives us a Dynamic Text
.
The final line of code in numberInput
uses fmap
to apply the function readMaybe . unpack
to the Dynamic
value of the InputElement
.
This produces a Dynamic (Maybe Double)
.
Our main
function uses fmap
to map over the Dynamic (Maybe Double)
produced by numberInput
and pack . show
the value it contains.
We store the new Dynamic Text
in numberString
and feed that into dynText
to actually display the Text
Running the app at this point should produce an input and some text showing the Maybe Double
.
Typing in a number should produce output like Just 12.0
and typing in other text should produce the output Nothing
.
Now that we have numberInput
we can put together a couple inputs to make a basic calculator.
> {-# LANGUAGE OverloadedStrings #-}
> import Reflex
> import Reflex.Dom
> import Data.Map (Map)
> import qualified Data.Map as Map
> import Data.Text (pack, unpack)
> import Text.Read (readMaybe)
>
> main = mainWidget $ el "div" $ do
> nx <- numberInput
> text " + "
> ny <- numberInput
> text " = "
> let result = zipDynWith (\x y -> (+) <$> x <*> y) nx ny
> resultString = fmap (pack . show) result
> dynText resultString
> numberInput :: DomBuilder t m => m (Dynamic t (Maybe Double))
> numberInput = do
> n <- inputElement $ def
> & inputElementConfig_initialValue .~ "0"
> & inputElementConfig_elementConfig . elementConfig_initialAttributes .~ ("type" =: "number")
> return . fmap (readMaybe . unpack) $ _inputElement_value n
numberInput
hasn't changed here.
Our main
function now creates two inputs.
zipDynWith
is used to produce the actual sum of the values of the inputs.
The type signature of zipDynWith
is:
zipDynWith :: Reflex t => (a -> b -> c) -> Dynamic t a -> Dynamic t b -> Dynamic t c
You can see that it takes a function that combines two pure values and produces some other pure value, and two Dynamic
s, and produces a Dynamic
.
In our case, zipDynWith
is combining the results of our two numberInput
s (with a little help from <$>
and <*>
) into a sum.
We use fmap
again to apply pack . show
to result
(a Dynamic (Maybe Double)
) resulting in a Dynamic Text
.
This resultText
is then displayed using dynText
.
Next, we'll add support for other operations.
We're going to add a dropdown so that the user can select the operation to apply.
The function dropdown
has the type:
dropdown :: (DomBuilder t m, MonadFix m, MonadHold t m, PostBuild t m, Ord k) => k -> Dynamic t (Map k Text) -> DropdownConfig t k -> m (Dropdown t k)
The first argument is the initial value of the Dropdown
.
The second argument is a Dynamic (Map k Text)
that represents the options in the dropdown.
The Text
values of the Map
are the strings that will be displayed to the user.
If the initial key is not in the Map
, it is added and given a Text
value of ""
.
The final argument is a DropdownConfig
.
Our supported operations will be:
data Op = Plus | Minus | Times | Divide deriving (Eq, Ord)
ops = Map.fromList [(Plus, "+"), (Minus, "-"), (Times, "*"), (Divide, "/")]
We'll use this as an argument to dropdown
:
d <- dropdown Times (constDyn ops) def
We are using constDyn
again here to turn our Map
of operations into a Dynamic
.
Using def
, we provide the default DropdownConfig
.
The result, d
, will be a Dropdown
.
We can retrieve the Dynamic
selection of a Dropdown
by using _dropdown_value
.
> {-# LANGUAGE OverloadedStrings #-}
> import Reflex
> import Reflex.Dom
> import Data.Map (Map)
> import qualified Data.Map as Map
> import Data.Text (pack, unpack, Text)
> import Text.Read (readMaybe)
>
> main = mainWidget $ el "div" $ do
> nx <- numberInput
> d <- dropdown Times (constDyn ops) def
> ny <- numberInput
> let values = zipDynWith (,) nx ny
> result = zipDynWith (\o (x,y) -> runOp o <$> x <*> y) (_dropdown_value d) values
> resultText = fmap (pack . show) result
> text " = "
> dynText resultText
>
> numberInput :: DomBuilder t m => m (Dynamic t (Maybe Double))
> numberInput = do
> n <- inputElement $ def
> & inputElementConfig_initialValue .~ "0"
> & inputElementConfig_elementConfig . elementConfig_initialAttributes .~ ("type" =: "number")
> return . fmap (readMaybe . unpack) $ _inputElement_value n
>
> data Op = Plus | Minus | Times | Divide deriving (Eq, Ord)
>
> ops :: Map Op Text
> ops = Map.fromList [(Plus, "+"), (Minus, "-"), (Times, "*"), (Divide, "/")]
>
> runOp :: Fractional a => Op -> a -> a -> a
> runOp s = case s of
> Plus -> (+)
> Minus -> (-)
> Times -> (*)
> Divide -> (/)
This is our complete program.
We've added an uninteresting function runOp
that takes an Op
and returns an operation.
The keys of the Map
we used to create the Dropdown
had the type Op
.
When we retrieve the value of Dropdown
, we'll use runOp
to turn the Dropdown
selection into the function we need to apply to our numbers.
After creating the two numberInput
s, we combine them using zipDynWith
applying (,)
, making a tuple of type Dynamic (Maybe Double, Maybe Double)
and binding it to values
.
Next, we call zipDynWith
again, combining the _dropdown_value
and values
.
Now, instead of applying (+)
to our Double
values, we use runOp
to select an operation based on the Dynamic
value of our Dropdown
.
Running the app at this point will give us our two number inputs with a dropdown of operations sandwiched between them. Multiplication should be pre-selected when the page loads.
Let's spare a thought for the user of our calculator and add a little UI styling. Our number input currently looks like this:
numberInput :: DomBuilder t m => m (Dynamic t (Maybe Double))
numberInput = do
n <- inputElement $ def
& inputElementConfig_initialValue .~ "0"
& inputElementConfig_elementConfig . elementConfig_initialAttributes .~ ("type" =: "number")
return . fmap (readMaybe . unpack) $ _inputElement_value n
Let's give it some html attributes to work with:
numberInput :: DomBuilder t m => m (Dynamic t (Maybe Double))
numberInput = do
let initAttrs = (("type" =: "number") <> ("style" =: "border-color: blue"))
n <- inputElement $ def
& inputElementConfig_initialValue .~ "0"
& inputElementConfig_elementConfig . elementConfig_initialAttributes .~ initAttrs
return . fmap (readMaybe . unpack) $ _inputElement_value n
Here, we've used a (Map Text Text)
.
This Map
represents the html attributes of our inputs.
Static attributes are useful and quite common, but attributes will often need to change.
Instead of just making the InputElement
blue, let's change it's color based on whether the input successfully parses to a Double
:
{-# LANGUAGE RecursiveDo #-}
import Control.Monad.Fix (MonadFix)
numberInput :: (DomBuilder t m, MonadFix m) => m (Dynamic t (Maybe Double))
numberInput = do
let initAttrs = ("type" =: "number") <> (style False)
color error = if error then "red" else "green"
style error = "style" =: ("border-color: " <> color error)
styleChange :: Maybe Double -> Map AttributeName (Maybe Text)
styleChange result = case result of
(Just _) -> fmap Just (style False)
(Nothing) -> fmap Just (style True)
rec
n <- inputElement $ def
& inputElementConfig_initialValue .~ "0"
& inputElementConfig_elementConfig . elementConfig_initialAttributes .~ initAttrs
& inputElementConfig_elementConfig . elementConfig_modifyAttributes .~ modAttrEv
let result = fmap (readMaybe . unpack) $ _inputElement_value n
modAttrEv = fmap styleChange (updated result)
return result
Note that we need to add a language pragma here to enable the RecursiveDo
language extension, and then we need to import MonadFix
.
Here style
function takes a Bool
value, whether input is correct or not, and it gives a Map
of attributes with green or red color respectively.
The next function styleChange
actually produces a Map
which tells which attribute to change.
If the value of a key in the Map
is a Just
value then the attribute is either added or modified.
If the value of key is Nothing
, then that attribute is removed.
An Event
of this Map
is specified in the elementConfig_modifyAttributes
.
In the first line of the rec
, we have supplied this Event
as argument modAttrEv
.
The Dynamic
value of the input is bound to result
.
The code for parsing this value has not changed.
After we bind result
, we use fmap
again to apply a switching function to the updated result
Event
.
The switching function checks whether the value was successfully parsed and gives the corresponding Event
to modify the attributes.
The complete program now looks like this:
> {-# LANGUAGE OverloadedStrings #-}
> {-# LANGUAGE RecursiveDo #-}
> import Reflex
> import Reflex.Dom
> import Data.Map (Map)
> import qualified Data.Map as Map
> import Data.Text (pack, unpack, Text)
> import Text.Read (readMaybe)
> import Control.Monad.Fix (MonadFix)
>
> main = mainWidget $ el "div" $ do
> nx <- numberInput
> d <- dropdown Times (constDyn ops) def
> ny <- numberInput
> let values = zipDynWith (,) nx ny
> result = zipDynWith (\o (x,y) -> runOp o <$> x <*> y) (_dropdown_value d) values
> resultText = fmap (pack . show) result
> text " = "
> dynText resultText
>
> numberInput :: (DomBuilder t m, MonadFix m) => m (Dynamic t (Maybe Double))
> numberInput = do
> let initAttrs = ("type" =: "number") <> (style False)
> color error = if error then "red" else "green"
> style error = "style" =: ("border-color: " <> color error)
> styleChange :: Maybe Double -> Map AttributeName (Maybe Text)
> styleChange result = case result of
> (Just _) -> fmap Just (style False)
> (Nothing) -> fmap Just (style True)
>
> rec
> n <- inputElement $ def
> & inputElementConfig_initialValue .~ "0"
> & inputElementConfig_elementConfig . elementConfig_initialAttributes .~ initAttrs
> & inputElementConfig_elementConfig . elementConfig_modifyAttributes .~ modAttrEv
> let result = fmap (readMaybe . unpack) $ _inputElement_value n
> modAttrEv = fmap styleChange (updated result)
> return result
>
> data Op = Plus | Minus | Times | Divide deriving (Eq, Ord)
>
> ops :: Map Op Text
> ops = Map.fromList [(Plus, "+"), (Minus, "-"), (Times, "*"), (Divide, "/")]
>
> runOp :: Fractional a => Op -> a -> a -> a
> runOp s = case s of
> Plus -> (+)
> Minus -> (-)
> Times -> (*)
> Divide -> (/)
The input border colors will now change depending on their value.