!7

p {
  maxConsoleChars(10000);
  new LinkedHashMap<U, S> sentences; // sentence => title
  
  int n = 0, max = 1000;
  for (WikiPage page : indexedSimpleWikipedia_allPages()) {
    for (S s : sentencesAfterLines(page.text))
      putIfNotThere(sentences, toU(s), page.title);
    if ((++n % 1000) == 0) print(n + " / " + l(sentences));
    if (n >= max) break;
  }
  
  Pair<L<U>> p = splitAccordingToPredicate(f mechList_isEnglishSentence_u, keys(sentences));
  
  printAsciiHeading("Non-sentences");
  pnl(keysToLinkedHashMap(sentences, p.a));
  
  printAsciiHeading("Sentences");
  pnl(keysToLinkedHashMap(sentences, p.b));
}

Began life as a copy of #1014173