# Liquorice ![](logo.png "Liquorice Logo") *Liquorice* is a system for constructing maps for the [computer game Doom](https://doomwiki.org/wiki/Doom) by writing algorithms to describe the geometry. A *Liquorice* program is a series of instructions to move a virtual "turtle" or "pen", which define the walls, floors, rooms and monsters as it goes. More specifically, *Liquorice* is an embedded domain-specific language (eDSL) within the pure functional programming language [Haskell](https://www.haskell.org/). ## Quick example -- simple example, triangle (for orientation); unique texture per line import Liquorice.Monad import Liquorice.Render main = buildWad "example1.wad" $ runWadL $ do mid "ZZWOLF1" draw 128 0 mid "ZZWOLF2" draw 0 128 turnaround mid "ZZWOLF3" draw 128 128 rightsector 0 128 160 turnaround step 64 32 thing ## More detailed example See [examples/birds.hs](examples/birds.hs), for an example program that generates a complete playable map. The map targets Doom 1 / The Ultimate Doom (map slot E2M8), and requires a doom engine with raised engine limits, such as [Crispy Doom](https://www.chocolate-doom.org/wiki/Crispy_Doom). The generated PWAD, with nodes, ready to play: [birds.zip](https://redmars.org/liquorice/birds.zip) *birds.hs* is a re-implementation/transformation of ["Bird Cage" for WadC](https://redmars.org/wadc/examples/#_birds_wl). ## Getting Started ### For Doom novices Doom's engine source code was open-sourced in 1996, but you need a copy of the game's data files to use them. You can buy [The Ultimate Doom](https://www.gog.com/game/the_ultimate_doom) and [Doom II](https://www.gog.com/game/doom_ii_final_doom) at [Gog](http://www.gog.com/), amongst other places. Failing that, you could try [FreeDoom](https://freedoom.github.io/), a free-content game for Doom engines. Examples of powerful open source engines include [Crispy Doom](https://www.chocolate-doom.org/wiki/index.php/Crispy_Doom), [Eternity Engine](http://eternity.youfailit.net/wiki/Main_Page) and [GZDoom](https://zdoom.org/index). ### For Haskell novices … ## Evaluation The commands that a typical *Liquorice* program will use are predominantly monadic, and so make use of Haskell's "do-notation" for ordering. Internally, the majority of these monadic functions are wrappers around pure equivalents which transform an input `Context` type into an output, e.g.: xoff :: Int -> Context -> Context xoff x c = c { paletteXoff = x } place :: Int -> Int -> (Context -> Context) -> Context -> Context place x y fn c = c & step x y & fn & step (-1 * x) (-1 * y) The pure functions can be combined using the infix operator `&` (from `Data.Function`), or the more usual composition operator (which reads back-to-front). However, the monadic versions are probably more user- friendly, and so the separate pure implementations might go away or stop being explicitly exported at some point. We also probably need to use the monadic versions if we want to introduce randomness or debug IO in the middle of a program. A series of monadic Liquorice statements are converted into a final `Context` via `runWadL`. A `Context` is written to an output PWAD file via buildWad :: FilePath -> Context -> IO () Internally, `buildWad` first converts a `Context` into an intermediate data structure `WadMap`, which closely resembles the binary structure of a PWAD. (`WadMap` itself is a specialisation of `Wad`, imposing the presence of map specific lumps) ## Pros/Cons Pros: * Unlike WadC, you have full access to the Context at any point in your program, which you can inspect or transform as you wish. * You also have the full power of the Haskell programming language. * In particular simply infix numeric operators are hugely useful. Cons: * There's no GUI. * The Pure-versus-Monadic functions are a bit awkward and I might get rid of the pure versions altogether in the future * There are some slow parts (line splitting in particular) * If you don't know Haskell, this will be very hard to use and most error messages will make little sense. * You need a Haskell development environment installed to use *Liquorice*. ## Code overview The following three source files are the ones that you will want to import as modules to your Liquorice program: * `Liquorice.hs` — The main `Context` definition and most basic pure operations * `Liquorice/Monad.hs` — monadic wrappers around the above, + some more * `Liquorice/Render.hs` — exports `buildWad` for exporting a PWAD These are back-end implementation details: * `Liquorice/Pure.hs` — most of the "standard library", in pure functions * `Liquorice/Line.hs` — `Line` type, line splitting and tests * `Liquorice/Wad.hs` — Wad and WadMap definitions and binary serialisation * `TestMain.hs` — HTF test harness ## Author *Liquorice* was created by [Jonathan Dowland](https://jmtd.net/) and is distributed under the terms of the GNU Public License, version 3 (see [COPYING](COPYING)). The design of *Liquorice* is heavily influenced by [Wad Language (WadC)](https://jmtd.net/wadc/), which in turn owes a debt to [LOGO](http://el.media.mit.edu/logo-foundation/what_is_logo/logo_programming.html). ## See also * [Jonathan Dowland's homepage](https://jmtd.net/) * [WadC](https://jmtd.net/wadc/) * [The Doom Wiki](https://doomwiki.org/)