This is an alternative site for discovering Elm packages. You may be looking for the official Elm package site instead.
elm-html-parser
version 1.0.0
license BSD-3-Clause
native-modules False
elm-version 0.18.0 <= v < 0.19.0
Tag 1.0.3
Committed At 2018-02-10 23:43:08 UTC
elm-lang/core 5.0.0 <= v < 6.0.0 5.1.1

README

Elm Parser

General purpose parser library for Elm

elm-package install jaredramirez/elm-parser

Usage

module Sample exposing (..)

import Parser exposing (Parser, (|=), (|*))
import Parser.Char as Char
import Parser.Number as Number


tuple : Parser ( Int, Int )
tuple =
    Parser.parse (\x y -> ( x, y ))
        |* Char.char '('
        |= Number.naturalNumber
        |* Char.char ','
        |= Number.naturalNumber
        |* Char.char ')'

run1 =
  Parser.run tuple "(3,3)"

run2 =
  Parser.run tuple "(35)"

run1 will produce the value:

Ok (3,3)

run2 will produce the value:

Err
    ( { source = "(35)"
      , offset = 3
      , row = 1
      , col = 4
      }
    , ExpectedSymbol ","
    )

In the run2 example, you are givin the state of the parser at the moment of failure and the Problem with the parse operation.

You can look all the possible values of Problem's here.

How it works

Take the parser from the above example.

transform : Int -> Int -> ( Int, Int )
transform x y =
    ( x, y )


tuple : Parser ( Int, Int )
tuple =
    Parser.succeed transform
        |* Char.char '('
        |= Number.naturalNumber
        |* Char.char ','
        |= Number.naturalNumber
        |* Char.char ')'

First, we create the function transform that takes two Ints, and puts them in a tuple. Then, in the first line of the tuple function, we take the function transform and lifting it into a "parser". In the subsequent parts of the pipline, we are applying parsers to the function transform. If the parser in the pipline begins with

  • |* it means "run is parser and make sure it is successful, then throw away the result".
  • |= it means "run is parser and make sure it is successful, then apply the result to the function".

This relies on partial function application. In this case, transform has two arguements, so in our pipeline we must have two |= to get a result, otherwise transform won't have all of it's arguements applied!

For more on the pipeline parser concept, see this.

You can also use this library to chain parsers with andThen. This is generally not as readable and can be more confuisng than pipeline-style parsing. It can be helpeful to have though. Say you only want tuples where the values are equal. That is, (1,1) would succeed but (1,2) would not. You can refactor tuple it include that functionality easily with andThen:

tuple : Parser ( Int, Int )
tuple =
    let
        tupleParser =
            Parser.succeed transform
                |* Char.char '('
                |= Number.naturalNumber
                |* Char.char ','
                |= Number.naturalNumber
                |* Char.char ')'
    in
        tupleParser
            |> Parser.andThen
                (\( x, y ) ->
                    if x == y then
                        Parser.succeed ( x, y )
                    else
                        Parser.fail <|
                            Parser.Bad "I expected the values to match"
                )

Thanks

While I wrote all of the code in this package, most of it was heavily influenced/inspired by others (with the exception of Parser.Html). This package was written for the purpose of my learning, and I figured I'd publish it for kicks. So, a big thanks to the following as they taught me a lot about parsers in a functional language, and are great resources.