Uploaded by
Published on

A parser for the PureScript concrete syntax tree.


The supported parsers are exported from PureScript.CST. The parsers support some error recovery, which is reflected in the RecoveredParserResult type. The CST types in PureScript.CST.Types are indexed by an error type, which is fixed to RecoveredError in the case of failures. Parses that succeed without failures have the error type fixed to Void.

import PureScript.CST (RecoveredParserResult(..), parseModule)

example = case parseModule myModuleSource of
  ParseSucceeded cst ->
    -- `cst` is type `Module Void` to indicate no errors
  ParseSucceededWithErrors cst errors ->
    -- `cst is type `Module RecoveredError` and contains error nodes at points of failure.
  ParseFailed error ->
    -- An unrecoverable error was encountered.


PureScript.CST.Traversal contains traversals for analyzing and rewriting the CST. These folds take a language visitor record with functions for handling the primary types in the CST. Default records are provided that do nothing for the cases you don't care about.

For example, if you wanted to quickly gather all the identifiers used in the expressions of a module along with locations, you might use foldMapModule and only provide a case for onExpr.

import Prelude
import Data.Map (SemigroupMap(..))
import Data.Map as Map
import Data.Set (Set)
import Data.Set as Set
import Data.Tuple (Tuple(..))
import PureScript.CST.Traversal (foldMapModule, defaultMonoidalVisitor)
import PureScript.CST.Types as CST

type QualifiedIdent = Tuple (Maybe CST.ModuleName) CST.Ident
type UsageMap = SemigroupMap QualifiedIdent (Set CST.SourceRange)

getExprIdents :: forall a. CST.Module a -> UsageMap
getExprIdents = foldMapModule $ defaultMonoidalVisitor
  { onExpr = case _ of
      CST.ExprIdent (CST.QualifiedName ident) ->
          $ Map.singleton (Tuple ident."module"
          $ Set.singleton ident.token.range
      _ -> mempty


The provided integration test attempts to parse a provided package set, and will report any errors it encounters as well as listing the fastest and slowest parse times along with the mean parse time for the set.

npm run parse-package-set

You can also benchmark or parse a single file:

npm run bench-file MyModule.purs
npm run parse-file -- MyModule.purs --tokens