spec.ctr

Broom memoryLimit: 1024 * 1024 * 1024 * 4.

import main: \*.
import
  Library/Test: 'describe'
.

#:language XFrozen

describe TokenAnalyser do: {:it:let
  it exists do: {
    { TokenAnalyser. } should not raiseError: String.
  }.

  let[Reflect thisContext] lexer { Lexer }.
  let[Reflect thisContext] analyser { TokenAnalyser with: lexer }.
  let[Reflect thisContext] glblctx { Reflect thisContext }.
  let[Reflect thisContext] outfile { File new: 'outfile', open: 'w+' }.
  let[Reflect thisContext] limit { 500 }.

  it can tokenise and detect spelling mistakes do: {:thisTest
    var allTargetTexts is (Generator from: 2 to: limit) fmap: (\:x File new: 'samples/sample$$x', read).
    # (
    #   File
    #     list: 'samples',
    #     filter: (\:_:descr (descr @ 'type' = 'file') & (descr @ 'file' startsWith: 'sample')),
    #     take: 1,
    #     asGenerator fmap: \:descr (File new: 'samples/' + (descr @ 'file'), read)
    # ).

    lexer collate: (File new: 'collation', read split: '\n', fmap: {:x ^x split: ' ' max: 1. }).

    thisTest tests: 'WordTree'.
    var tree is WordTree new: 'dict'.

    thisTest tests: 'Spellcheck'.
    thisTest progressesUpTo: limit.
    var consolidated is analyser
      spellcheck: allTargetTexts
      withDictionaryTree: tree
      pre: {:i thisTest progress: i + 1. }
      post: { thisTest show. }.
    #{
      outfile write: (
        #analyser prettyPrint: consolidated tokens: allTokens toArray
        consolidated fmap: (\:x '%s -- %L' % x), join: '\n'
      ).
    #} should output: testfile read to: outfile.
  }.

}.