koron / ugmatcha

Library to matches many words simultaneously.

UG Matcha

UG Matcha matches against many words simultaneously, with less memory at runtime.

Links

• API Reference (javadoc)
• Benchmark and result

Usage

Please use UGMatcher#newMatcher() and UGMatcher#match() for maximum performance.

import net.kaoriya.ugmatcha.UGMatcher;
import net.kaoriya.ugmatcha.MatchHandler;

// Prepare a matcher.
UGMatcher m = UGMatcher.newMatcher("foo", "bar", "baz");

m.match("foobarbazqux", new MatchHandler() {
  // This method is called when found each words.
  public boolean matched(UGMatcher u, int wordId, String text, int index) {
    // You can retrieve matched word by UGMatcher#getWord()
    String word = u.getWord(wordId);
    // Do something with matched "word".

    // Return true to continue matching, otherwise false to terminate.
    return true;
  }
});

For convenient usecase, you can use UGMatcher#find(). But this method expends more memory than UGMatcher#match() at runtime.

import net.kaoriya.ugmatcha.UGMatcher;
import net.kaoriya.ugmatcha.Match;

import java.util.List;

// Prepare a matcher.
UGMatcher m = UGMatcher.newMatcher("foo", "bar", "baz");

List<Match> list = m.find("foobarbazqux");
for (Match m : list) {
  // Access matched "word" by Match#text.
  String word = m.text;
  // Access poistion of matched "word" by Match.index.
  int index = m.index;
}

UGMatcher#newMatcher() accepts List<String> too. And it accepts another UGMatcher to duplicate it. A UGMatcher instance is not support multi-threads use/access. So making duplication is used for multi-threads.

import net.kaoriya.ugmatcha.UGMatcher;
import java.util.ArrayList;

ArrayList<String> list = new ArrayList<>();
list.add("foo");
list.add("bar");
list.add("baz");

UGMatcher original = UGMatcher.newMatcher(list);
UGMatcher copied = UGMatcher.newMatcher(original);