list/src/main/java/alg/word/WordCount.java

package alg.word;

import java.util.ArrayList;
import java.util.List;
import java.util.stream.Collectors;

import alg.Common;
import data.CalculateFor;
import data.Sentence;
import data.Statistics;
import data.Word;

class WordCount {
	private static void calculateNoFilter(List<Sentence> corpus, Statistics stats) {
		for (Sentence s : corpus) {
			List<String> sentence = new ArrayList<>(s.getWords().size());

			if (stats.getCf() == CalculateFor.LEMMA) {
				sentence.addAll(s.getWords()
						.stream()
						.map(Word::getLemma)
						.collect(Collectors.toList()));
			} else if (stats.getCf() == CalculateFor.WORD) {
				sentence.addAll(s.getWords()
						.stream()
						.map(Word::getWord)
						.collect(Collectors.toList()));
			}

			for (String word : sentence) {
				Common.updateMap(stats.result, word);
			}
		}
	}

	private static void calculateVCC(List<Sentence> corpus, Statistics stats) {
		for (Sentence s : corpus) {
			List<String> sentence = new ArrayList<>(s.getWords().size());

			if (stats.getCf() == CalculateFor.LEMMA) {
				sentence.addAll(s.getWords()
						.stream()
						.map(Word::getCVVLemma)
						.collect(Collectors.toList()));
			} else if (stats.getCf() == CalculateFor.WORD) {
				sentence.addAll(s.getWords()
						.stream()
						.map(Word::getCVVWord)
						.collect(Collectors.toList()));
			}

			for (String word : sentence) {
				if (word.length() > stats.getSubstringLength()) {
					for (int i = 0; i <= word.length() - stats.getSubstringLength(); i++) {
						String substring = word.substring(i, i + stats.getSubstringLength());
						Common.updateMap(stats.result, substring);
					}
				}
			}
		}
	}

	private static void calculateForJosType(List<Sentence> corpus, Statistics stats) {
		for (Sentence s : corpus) {
			List<String> sentence = new ArrayList<>(s.getWords().size());
			List<Word> filteredWords = new ArrayList<>();

			for (Word word : s.getWords()) {
				if (word.getMsd() != null && word.getMsd().charAt(0) == stats.getDistributionJosWordType()) {
					filteredWords.add(word);
				}
			}

			if (stats.getCf() == CalculateFor.LEMMA) {
				sentence.addAll(filteredWords
						.stream()
						.map(Word::getLemma)
						.collect(Collectors.toList()));
			} else if (stats.getCf() == CalculateFor.WORD) {
				sentence.addAll(filteredWords
						.stream()
						.map(Word::getWord)
						.collect(Collectors.toList()));
			}

			for (String word : sentence) {
				Common.updateMap(stats.result, word);
			}
		}
	}

	private static void calculateForTaxonomyAndJosType(List<Sentence> corpus, Statistics stats) {
		for (Sentence s : corpus) {
			if (s.getTaxonomy().equalsIgnoreCase(stats.getDistributionTaxonomy())) {
				List<String> sentence = new ArrayList<>(s.getWords().size());
				List<Word> filteredWords = new ArrayList<>();

				for (Word word : s.getWords()) {
					if (word.getMsd().charAt(0) == stats.getDistributionJosWordType()) {
						filteredWords.add(word);
					}
				}

				if (stats.getCf() == CalculateFor.LEMMA) {
					sentence.addAll(filteredWords
							.stream()
							.map(Word::getLemma)
							.collect(Collectors.toList()));
				} else if (stats.getCf() == CalculateFor.WORD) {
					sentence.addAll(filteredWords
							.stream()
							.map(Word::getWord)
							.collect(Collectors.toList()));
				}

				for (String word : sentence) {
					Common.updateMap(stats.result, word);
				}
			}
		}
	}

	private static void calculateForTaxonomy(List<Sentence> corpus, Statistics stats) {
		for (Sentence s : corpus) {
			if (s.getTaxonomy().equalsIgnoreCase(stats.getDistributionTaxonomy())) {
				List<String> sentence = new ArrayList<>(s.getWords().size());

				if (stats.getCf() == CalculateFor.LEMMA) {
					sentence.addAll(s.getWords()
							.stream()
							.map(Word::getLemma)
							.collect(Collectors.toList()));
				} else if (stats.getCf() == CalculateFor.WORD) {
					sentence.addAll(s.getWords()
							.stream()
							.map(Word::getWord)
							.collect(Collectors.toList()));
				}

				for (String word : sentence) {
					Common.updateMap(stats.result, word);
				}
			}
		}
	}

	static void calculateForAll(List<Sentence> corpus, Statistics stats) {
		boolean taxonomyIsSet = stats.isTaxonomySet();
		boolean JosTypeIsSet = stats.isJOSTypeSet();

		// branching because even though the only difference is an if or two &&
		// O(if) = 1, the amount of ifs adds up and this saves some time
		if (taxonomyIsSet && JosTypeIsSet) {
			calculateForTaxonomyAndJosType(corpus, stats);
		} else if (taxonomyIsSet && !JosTypeIsSet) {
			calculateForTaxonomy(corpus, stats);
		} else if (!taxonomyIsSet && JosTypeIsSet) {
			calculateForJosType(corpus, stats);
		} else {
			if (stats.isVcc()) {
				calculateVCC(corpus, stats);
			} else {
				calculateNoFilter(corpus, stats);
			}
		}
	}
}