Ejemplo n.º 1
0
        public void test2()
        {
            string s    = "中国|国人|zg人";
            string test = "我是中国人";

            WordsSearchEx wordsSearch = new WordsSearchEx();

            wordsSearch.SetKeywords(s.Split('|').ToList());

            var b = wordsSearch.ContainsAny(test);

            Assert.AreEqual(true, b);


            var f = wordsSearch.FindFirst(test);

            Assert.AreEqual("中国", f.Keyword);

            var alls = wordsSearch.FindAll(test);

            Assert.AreEqual("中国", alls[0].Keyword);
            Assert.AreEqual(2, alls[0].Start);
            Assert.AreEqual(3, alls[0].End);
            Assert.AreEqual(0, alls[0].Index);//返回索引Index,默认从0开始
            Assert.AreEqual("国人", alls[1].Keyword);
            Assert.AreEqual(2, alls.Count);

            var t = wordsSearch.Replace(test, '*');

            Assert.AreEqual("我是***", t);
        }
Ejemplo n.º 2
0
        private static void InitPyWords()
        {
            if (_search == null)
            {
                lock (lockObj) {
                    if (_search == null)
                    {
                        var ass = typeof(WordsHelper).Assembly;
#if NETSTANDARD2_1
                        const string resourceName = "ToolGood.Words.dict.pyWords.txt.br";
#else
                        const string resourceName = "ToolGood.Words.dict.pyWords.txt.z";
#endif
                        Stream sm = ass.GetManifestResourceStream(resourceName);
                        byte[] bs = new byte[sm.Length];
                        sm.Read(bs, 0, (int)sm.Length);
                        sm.Close();
                        var bytes = Decompress(bs);
                        var tStr  = Encoding.UTF8.GetString(bytes);
                        bytes = null;

                        var lines = tStr.Split(new char[] { '\n' }, StringSplitOptions.RemoveEmptyEntries);
                        tStr = null;

                        var           wordPy      = new List <ushort>();
                        List <string> keywords    = new List <string>();
                        List <int>    wordPyIndex = new List <int>();
                        wordPyIndex.Add(0);

                        foreach (var line in lines)
                        {
                            var sp = line.Split(',');
                            keywords.Add(sp[0]);
                            for (int i = 1; i < sp.Length; i++)
                            {
                                var idx = sp[i];
                                wordPy.Add(ushort.Parse(idx, System.Globalization.NumberStyles.HexNumber));
                            }
                            wordPyIndex.Add(wordPy.Count);
                        }
                        var search = new WordsSearchEx();
                        search.SetKeywords(keywords);
                        _wordPyIndex = wordPyIndex.ToArray();
                        _wordPy      = wordPy.ToArray();
                        _search      = search;

                        wordPy      = null;
                        keywords    = null;
                        wordPyIndex = null;
                    }
                    GC.Collect();
                }
            }
        }
Ejemplo n.º 3
0
        private static void WordsSearchExSearch(List <string> list, string txt)
        {
            WordsSearchEx wordsSearch = new WordsSearchEx();

            wordsSearch.SetKeywords(list);
            wordsSearch.Save("WordsSearchEx.dat");

            Stopwatch watch = new Stopwatch();

            watch.Start();
            for (int i = 0; i < 100000; i++)
            {
                wordsSearch.FindAll(txt);
            }
            watch.Stop();
            Console.WriteLine("WordsSearchEx: " + watch.ElapsedMilliseconds.ToString("N0") + "ms");
        }