{"id":1507,"date":"2023-02-13T09:54:31","date_gmt":"2023-02-13T01:54:31","guid":{"rendered":"http:\/\/www.algmain.com\/?page_id=1507"},"modified":"2023-02-13T13:52:15","modified_gmt":"2023-02-13T05:52:15","slug":"mad-max","status":"publish","type":"page","link":"http:\/\/www.algmain.com\/index.php\/nlp-history\/basic-work\/mad-max\/","title":{"rendered":"\u75af\u72c2\u7684\u9ea6\u514b\u65af"},"content":{"rendered":"\n

\u672c\u7ae0\u5c06\u4ecb\u7ecdC#\u591a\u4efb\u52a1\u5728\u6570\u636e\u7edf\u8ba1\u5206\u6790\u4e0a\u7684\u5e94\u7528\u3002\u53e6\u5916\uff0c\u4f5c\u8005\u4ece\u4e00\u4e2a\u4fa7\u9762\u8868\u8fbe\u4e00\u4e0b\uff1aNLP\u4e2d\u7684\u8ba1\u7b97\u91cf\u662f\u76f8\u5f53\u201c\u6050\u6016\u201d\u7684\u3002<\/p>\n\n\n\n

\u4e00\u3001\u57fa\u672c\u60c5\u51b5<\/h2>\n\n\n\n

\u539f\u59cb\u8bed\u6599\u5206\u4e3a\u4e24\u4e2a\u90e8\u5206\uff1a\uff081\uff09\u8bed\u6599\u5e93\uff1b\uff082\uff09\u5b57\u5178\u5e93\u3002
\u5176\u4e2d\u8bed\u6599\u6570\u636e\u603b\u8ba110,101,283\u6761\uff0c\u603b\u8ba1977,485,287\u4e2a\u5b57\u7b26\u3002\u5b57\u5178\u4e0d\u91cd\u590d\u8bb0\u5f5512,926,204\u6761\u3002<\/p>\n\n\n\n

\u901a\u8fc7\u5bf9\u8bed\u6599\u5e93\u7684\u7edf\u8ba1\u5206\u6790\uff0c\u83b7\u5f97\u5b57\u7b26\u5171\u8ba112692\u4e2aUnicode\uff08\u5305\u62ec\u7279\u6b8a\u5b57\u7b26\uff09\u3002
\u901a\u8fc7\u5bf9\u8bed\u6599\u5e93\u7684\u7edf\u8ba1\u5206\u6790\uff0c\u5171\u83b7\u5f9748,218,815\u4e0d\u540c\u7684\u201c\u77ed\u53e5\u201d\u3002\u6240\u8c13\u77ed\u53e5\u5c31\u662f\u5b8c\u6574\u53e5\u5b50\u4e2d\u76f8\u5bf9\u5b8c\u6574\u7684\u4e00\u90e8\u5206\u3002<\/p>\n\n\n\n

\u4f8b\u5982\uff1a<\/p>\n\n\n

\n\u201c\u4f60\u6839\u672c\u4e0d\u77e5\u9053\u4f60\u81ea\u5df1\u7684\u4ef7\u503c\uff0c\u201d\u67ef\u6797\u65af\u592a\u592a\u7b11\u7740\u8bf4\uff0e\u201c\u50cf\u4f60\u8fd9\u6837\u76f4\u7387\u7684\u4eba\u62b5\u5f97\u4e0a\u4e00\u767e\u4e2a\u8c08\u5410\u6e29\u6587\u5c14\u96c5\u7684\u98ce\u6d41\u7537\u5b50\u3002\u201d\n<\/pre><\/div>\n\n\n
\u90a3\u4e48\u53ef\u4ee5\u4ece\u8fd9\u4e2a\u53e5\u5b50\u4e2d\u83b7\u5f97\u4e09\u4e2a\u77ed\u53e5\uff1a\u201c\u4f60\u6839\u672c\u4e0d\u77e5\u9053\u4f60\u81ea\u5df1\u7684\u4ef7\u503c\u201d\uff0c\u201c\u67ef\u6797\u65af\u592a\u592a\u7b11\u7740\u8bf4\u201d\uff0c\u201c\u50cf\u4f60\u8fd9\u6837\u76f4\u7387\u7684\u4eba\u62b5\u5f97\u4e0a\u4e00\u767e\u4e2a\u8c08\u5410\u6e29\u6587\u5c14\u96c5\u7684\u98ce\u6d41\u7537\u5b50\u201d\u3002<\/p>\n\n\n\n
\u8fd9\u4e9b\u77ed\u53e5\u7684\u957f\u5ea6\u5206\u5e03\u60c5\u51b5\u5982\u4e0b\uff1a<\/p>\n\n\n\n
\u5b57\u7b26\u4e32\u957f\u5ea6\uff08\u6309\u7167Unicode\u8ba1\u7b97\uff09<\/th>\u603b\u6570<\/th><\/tr>
1<\/td>5563<\/td><\/tr>
2<\/td>277200<\/td><\/tr>
3<\/td>1082830<\/td><\/tr>
4<\/td>3156576<\/td><\/tr>
5<\/td>3727422<\/td><\/tr>
6<\/td>5092530<\/td><\/tr>
7<\/td>5006767<\/td><\/tr>
8<\/td>4829769<\/td><\/tr>
9<\/td>4301055<\/td><\/tr>
10<\/td>3766264<\/td><\/tr>
11<\/td>3196160<\/td><\/tr>
\u2026\u2026<\/td>\u2026\u2026<\/td><\/tr>
674<\/td>1<\/td><\/tr>
904<\/td>1<\/td><\/tr>
1419<\/td>1<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n
\u4ece\u8868\u4e2d\u53ef\u4ee5\u770b\u51fa\uff0c\u521a\u5f00\u59cb\u968f\u5b57\u7b26\u4e32\u957f\u5ea6\u589e\u52a0\uff0c\u603b\u6570\u9010\u6e10\u589e\u52a0\u5230\u6700\u5927\u503c\u3002\u968f\u540e\u968f\u957f\u5ea6\u589e\u52a0\u800c\u9010\u6b65\u9012\u51cf\u3002<\/p>\n\n\n\n
\u4e8c\u3001\u7edf\u8ba1\u5904\u7406<\/h2>\n\n\n\n
\u8fd9\u91cc\u6240\u9700\u8981\u8fdb\u884c\u7684\u7edf\u8ba1\u5206\u6210\u597d\u51e0\u4e2a\u90e8\u5206\u3002\u5148\u6765\u4ecb\u7ecd\u4e0b\u4e00\u4e9b\u6700\u57fa\u672c\u7684\u5206\u6790\u3002<\/p>\n\n\n\n
\uff081\uff09\u5b57\u7b26\u7edf\u8ba1<\/p>\n\n\n\n
\u8fd9\u4e2a\u90e8\u5206\u7684\u7edf\u8ba1\u6700\u4e3a\u7b80\u5355\uff1a\u4ece\u539f\u59cb\u8bed\u6599\u5e93\u4e2d\u63d0\u53d6\u4e00\u884c\u6570\u636e\uff0c\u9010\u4e2a\u5b57\u7b26\u8fdb\u884c\u5206\u6790\u3002
\u5982\u679c\u5b57\u7b26\u5728Dictionary<string,int>\u5bf9\u8c61\u4e2d\uff0c\u53ef\u4ee5\u67e5\u5230\uff0c\u5219\u8ba1\u6570\u5668\u52a01\uff1b\u5426\u5219\uff0c\u5219\u589e\u52a0\u4e00\u6761\u65b0\u7684\u8bb0\u5f55\u3002<\/p>\n\n\n\n
\u8fd9\u4e2a\u90e8\u5206\u7684\u7edf\u8ba1\u5bf9\u5185\u5b58\u548cCPU\u7684\u6d88\u8017\u5e76\u4e0d\u9ad8\uff0c\u5f88\u5feb\u5c31\u53ef\u4ee5\u5b8c\u6210\u3002<\/p>\n\n\n\n
\uff082\uff09\u62c6\u5206\u77ed\u53e5<\/p>\n\n\n\n
\u8fd9\u4e2a\u90e8\u5206\u7684\u5904\u7406\u5176\u5b9e\u4e5f\u4e0d\u590d\u6742\uff1a\u4ece\u539f\u59cb\u8bed\u6599\u5e93\u4e2d\u63d0\u53d6\u4e00\u884c\u6570\u636e\uff0c\u6309\u7167\u6807\u70b9\u7b26\u53f7\u8fdb\u884c\u5206\u6bb5\u3002\u5404\u4e2a\u5206\u6bb5\u90e8\u5206\u7684\u6570\u636e\uff0c\u53ef\u89c6\u4e3a\u201c\u77ed\u53e5\u201d\u3002\u6216\u8005\u53ea\u63d0\u53d6\u4e2d\u6587\u90e8\u5206\u5373\u53ef\uff0c\u5982\u4e0b\u9762\u4ee3\u7801\u6240\u793a\uff1a<\/p>\n\n\n
\nMatch match = Regex.Match(item, @"[\\u4E00-\\u9FA5]+");\n<\/pre><\/div>\n\n\n
\u4e3a\u4e86\u66f4\u52a0\u7cbe\u786e\u5730\u5206\u6790\uff0c\u4f5c\u8005\u5728\u6574\u4e2a\u8fc7\u7a0b\u4e2d\uff0c\u589e\u52a0\u4e86\u6570\u91cf\u8bcd\u63d0\u53d6\u8fc7\u7a0b\u3002\u6bd5\u7adf\u6709\u4e0d\u5c11\u8bed\u6599\u4e2d\u5b58\u5728\u6570\u91cf\u8bcd\uff08\u5305\u62ec\u65e5\u671f\uff09\uff0c\u8fd9\u4e9b\u8bcd\u4f1a\u5e72\u6270\u5230\u77ed\u53e5\u7684\u5207\u5206\u8fc7\u7a0b\u3002<\/p>\n\n\n
\n2008\u5e749\u670820\u65e5\u672c\u62a5\u8baf\u2026\u2026\n<\/pre><\/div>\n\n\n
\u5728\u4e0a\u9762\u7684\u4f8b\u5b50\u4e2d\uff0c\u5982\u679c\u53ea\u63d0\u53d6\u4e2d\u6587\uff0c\u5c31\u4f1a\u8df3\u8fc7\u6570\u5b57\u628a\u201c\u65e5\u201d\u5b57\u548c\u201c\u672c\u201d\u5b57\u7ed1\u5230\u4e00\u8d77\uff0c\u4e0e\u539f\u672c\u7684\u610f\u601d\u76f8\u8fdd\u80cc\u3002\u8fd9\u79cd\u60c5\u51b5\u8fd8\u771f\u6709\u4e0d\u5c11\u3002<\/p>\n\n\n\n
\u4e3a\u4e86\u80fd\u51c6\u786e\u63d0\u53d6\u6570\u91cf\u8bcd\uff0c\u9700\u8981\u5efa\u7acb\u5927\u91cf\u7684\u6b63\u5219\u89c4\u5219\u6a21\u677f\u3002\u901a\u8fc7\u591a\u6b21\u5339\u914d\u548c\u878d\u5408\uff0c\u624d\u80fd\u8f83\u4e3a\u51c6\u786e\u5730\u63d0\u53d6\u6570\u91cf\u8bcd\u3002<\/p>\n\n\n
\n        private static readonly string[] QUANTITY_RULES =\n        {\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u6b63\u5219\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            \/\/ \u6b63\u5219\u8868\u8fbe\u5f0f\n            \/\/ \u7f16\u53f7\n            "0\\\\d*", \/\/\u7f16\u53f7\n            "No.\\\\d*", \/\/\u7f16\u53f7\n            \/\/ \u6574\u6570\n            "-?[1-9]\\\\d*", \/\/\u6574\u6570\n            "[1-9]\\\\d*[\\\\s]?-[\\\\s]?[1-9]\\\\d*", \/\/\u6574\u6570\n            "[1-9]\\\\d{0,2}(,\\\\d{3})+", \/\/\u6574\u6570\n            "[1-9]\\\\d*[\\\\s]?[\u5341|\u767e|\u5343|\u4e07|\u4ebf]", \/\/\u6574\u6570\n            "[1-9]\\\\d*[\\\\s]?[\u5341|\u767e|\u5343|\u4e07|\u4ebf|\u5146]\u4ebf", \/\/\u6574\u6570\n            \/\/ \u767e\u5206\u6570\n            "\\\\d+(\\\\.\\\\d+)?%", \/\/\u767e\u5206\u6570\n            "\\\\d+(\\\\.\\\\d+)?%-\\\\d+(\\\\.\\\\d+)?%", \/\/\u767e\u5206\u6570\n            \/\/ \u5206\u6570\n            "[1-9]\\\\d*[\\\\s]?\/[\\\\s]?[1-9]\\\\d*", \/\/\u5206\u6570\n            "[1-9]\\\\d*[\\\\s]?\/[\\\\s]?[1-9]\\\\d*-[1-9]\\\\d*\/[1-9]\\\\d*", \/\/\u5206\u6570\n            \/\/ \u82f1\u6587\u5355\u8bcd\n            "[A-Za-z][\\'A-Za-z]*", \/\/\u82f1\u6587\n            \/\/ \u6d6e\u70b9\u6570\n            "-?([1-9]\\\\d*\\\\.\\\\d+|0\\\\.\\\\d*[1-9]\\\\d*|0?\\\\.0+)", \/\/\u6d6e\u70b9\u6570\n            "([1-9]\\\\d*\\\\.\\\\d+|0\\\\.\\\\d*[1-9]\\\\d*|0?\\\\.0+)[\\\\s]?[\u5341|\u767e|\u5343|\u4e07|\u4ebf]", \/\/\u6d6e\u70b9\u6570\n            "([1-9]\\\\d*\\\\.\\\\d+|0\\\\.\\\\d*[1-9]\\\\d*|0?\\\\.0+)[\\\\s]?[\u5341|\u767e|\u5343|\u4e07|\u4ebf|\u5146]\u4ebf", \/\/\u6d6e\u70b9\u6570\n\n            \/\/ \u65e5\u671f\n            "(\\\\d{4}|\\\\d{2})[\\\\s]?\u5e74", \/\/\u65e5\u671f\n            "(0?[1-9]|1[0-2])[\\\\s]?\u6708", \/\/\u65e5\u671f\n            "((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u53f7", \/\/\u65e5\u671f\n            "((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u65e5", \/\/\u65e5\u671f\n            "(\\\\d{4}|\\\\d{2})[\\\\s]?\u5e74[\\\\s]?((1[0-2])|(0?[1-9]))[\\\\s]?\u6708", \/\/\u65e5\u671f\n            "(0?[1-9]|1[0-2])[\\\\s]?\u6708((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u65e5", \/\/\u65e5\u671f\n            "(\\\\d{4}|\\\\d{2})-((1[0-2])|(0?[1-9]))[\\\\s]?-[\\\\s]?(([12][0-9])|(3[01])|(0?[1-9]))", \/\/\u65e5\u671f\n            "(\\\\d{4}|\\\\d{2})[\\\\s]?\u5e74((1[0-2])|(0?[1-9]))[\\\\s]?\u6708(([12][0-9])|(3[01])|(0?[1-9]))[\\\\s]?\u65e5", \/\/\u65e5\u671f\n            \/\/ \u65f6\u95f4\n            "((1|0?)[0-9]|2[0-3])[\\\\s]?\u65f6", \/\/\u65f6\u95f4\n            "((1|0?)[0-9]|2[0-3])[\\\\s]?:[\\\\s]?([0-5][0-9])", \/\/\u65f6\u95f4\n            "((1|0?)[0-9]|2[0-3])[\\\\s]?\u65f6([0-5][0-9])[\\\\s]?\u5206", \/\/\u65f6\u95f4\n            "((1|0?)[0-9]|2[0-3])[\\\\s]?\u70b9([0-5][0-9])[\\\\s]?\u5206", \/\/\u65f6\u95f4\n            "((1|0?)[0-9]|2[0-3])[\\\\s]?:[\\\\s]?([0-5][0-9])[\\\\s]?:[\\\\s]?([0-5][0-9])", \/\/\u65f6\u95f4\n            "((1|0?)[0-9]|2[0-3])[\\\\s]?\u65f6[\\\\s]?([0-5][0-9])\u5206([0-5][0-9])[\\\\s]?\u79d2", \/\/\u65f6\u95f4\n            "((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u65e5[\\\\s]?((1|0?)[0-9]|2[0-3])[\\\\s]?\u65f6[\\\\s]?([0-5][0-9])[\\\\s]?\u5206", \/\/\u65f6\u95f4\n            \/\/ \u65f6\u95f4\u6bb5\n            "(\\\\d{4}|\\\\d{2})[\\\\s]?\u5e74[\\\\s]?-[\\\\s]?(\\\\d{4}|\\\\d{2})[\\\\s]?\u5e74", \/\/\u65f6\u95f4\u6bb5\n            "(\\\\d{4}|\\\\d{2})[\\\\s]?\/[\\\\s]?(\\\\d{4}|\\\\d{2})[\\\\s]?\u8d22\u5e74", \/\/\u65f6\u95f4\u6bb5\n            "((1|0?)[0-9]|2[0-3])[\\\\s]?[\u70b9|\u65f6][\\\\s]?-[\\\\s]?((1|0?)[0-9]|2[0-3])[\\\\s]?[\u70b9|\u65f6]", \/\/\u65f6\u95f4\u6bb5\n            "((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?[\u65e5|\u53f7][\\\\s]?-[\\\\s]?((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?[\u65e5|\u53f7]", \/\/\u65f6\u95f4\u6bb5\n            "(\\\\d{4}|\\\\d{2})\u5e74((1[0-2])|(0?[1-9]))[\\\\s]?\u6708[\\\\s]?-[\\\\s]?(\\\\d{4}|\\\\d{2})\u5e74((1[0-2])|(0?[1-9]))[\\\\s]?\u6708", \/\/\u65f6\u95f4\u6bb5\n            "((1[0-2])|(0?[1-9]))[\\\\s]?\u6708[\\\\s]?(([12][0-9])|(3[01])|(0?[1-9]))[\\\\s]?\u65e5[\\\\s]?-[\\\\s]?((1[0-2])|(0?[1-9]))[\\\\s]?\u6708[\\\\s]?(([12][0-9])|(3[01])|(0?[1-9]))[\\\\s]?\u65e5", \/\/\u65f6\u95f4\u6bb5\n            "(\\\\d{4}|\\\\d{2})[\\\\s]?\u5e74[\\\\s]?((1[0-2])|(0?[1-9]))[\\\\s]?\u6708[\\\\s]?(([12][0-9])|(3[01])|(0?[1-9]))[\\\\s]?\u65e5[\\\\s]?-[\\\\s]?(\\\\d{4}|\\\\d{2})[\\\\s]?\u5e74[\\\\s]?((1[0-2])|(0?[1-9]))[\\\\s]?\u6708[\\\\s]?(([12][0-9])|(3[01])|(0?[1-9]))[\\\\s]?\u65e5", \/\/\u65f6\u95f4\u6bb5\n\n            \/\/ \u7279\u6b8a\u6807\u8bc6\n            \/\/ \u8eab\u4efd\u8bc1\u53f7\u7801\n            "\\\\d{15}(\\\\d\\\\d[0-9xX])?", \/\/\u8eab\u4efd\u8bc1\n            \/\/ \u56fd\u5185\u7535\u8bdd\u53f7\u7801\uff08\u6613\u6df7\u6dc6\uff09\n            "(\\\\d{4}|\\\\d{3})?[\\\\s]?-[\\\\s]?(\\\\d{8}|\\\\d{7})", \/\/\u7535\u8bdd\n            "(\\\\d{4}|\\\\d{3})?[\\\\s]?-[\\\\s]?(\\\\d{4}|\\\\d{3})?-[\\\\s]?(\\\\d{4}|\\\\d{3})", \/\/\u7535\u8bdd\n            \/\/ HTML\u6807\u8bb0\uff08\u6613\u6df7\u6dc6\u3001\u6709\u6b7b\u5faa\u73af\u53ef\u80fd\uff09\n            \/\/ "<(.*)(.*)>.*<\\\\\/\\\\1>|<(.*) \\\\\/>", \/\/HTML\u6807\u8bb0\n            \/\/ EMail\u5730\u5740\n            "\\\\w+([-+.]\\\\w+)*@\\\\w+([-.]\\\\w+)*\\\\.\\\\w+([-.]\\\\w+)*", \/\/\u7535\u5b50\u90ae\u7bb1\n            \/\/ IP\u5730\u5740\n            "((2[0-4]\\\\d|25[0-5]|[01]?\\\\d\\\\d?)\\\\.){3}(2[0-4]\\\\d|25[0-5]|[01]?\\\\d\\\\d?)", \/\/IP\u5730\u5740\n            \/\/ URL\u5730\u5740\uff08\u6613\u6df7\u6dc6\uff09\n            "\\\\b(([\\\\w-]+:\/\/?|www[.])[^\\\\s()<>]+(?:\\\\([\\\\w\\\\d]+\\\\)|([^[:punct:]\\\\s]|\/)))", \/\/URL\u5730\u5740\n\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u6570\u8bcd\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            \/\/ \u6570\u8bcd\n            "$n[\\\\.|\\\\)]", \/\/\u7f16\u53f7\n            "\u201c[\\\\s]?$n[\\\\s]?\u201d", \/\/\u7f16\u53f7\n            "[0-9][\\\\dA-Za-z]*", \/\/\u7f16\u53f7\n            "$e+$n", \/\/\u7f16\u53f7\n            "$e+$n$e*", \/\/\u7f16\u53f7\n            "\u201c[\\\\s]?$n[\\\\s]?\u00b7$n[\\\\s]?\u201d", \/\/\u7f16\u53f7\n            "\\\\([\\\\s]?$n[\\\\s]?\\\\)|\\\\<[\\\\s]?$n[\\\\s]?\\\\>|\\\\{[\\\\s]?$n[\\\\s]?\\\\}", \/\/\u7f16\u53f7\n\n            "$d[\u591a|\u4f59](\u4e07|\u4ebf)", \/\/\u6570\u91cf\n            "$d[\\\\s]?:[\\\\s]?$d", \/\/\u6bd4\u4f8b\n            "$d[\\\\s]?(~|\u81f3)[\\\\s]?$d", \/\/\u533a\u95f4\n\n            "(\u96f6|$c){2,}", \/\/\u5e8f\u53f7\n            "\u5341$c", \/\/\u5341\u4f4d\u6570\n            "$c\u5341$c?", \/\/\u5341\u4f4d\u6570\n            "$c\u767e(\u96f6|$c\u5341)?$c?", \/\/\u767e\u4f4d\u6570\n            "$c\u5343((\u96f6|$c\u767e)?(\u96f6|$c\u5341)?$c?)", \/\/\u5343\u4f4d\u6570\n            "$c\u5343((\u96f6|$c\u767e)?(\u96f6|$c\u5341)?$c?)\u4e07", \/\/\u4e07\u4f4d\u6570\n            "$c\u4e07((\u96f6|$c\u5343)?((\u96f6|$c\u767e)?(\u96f6|$c\u5341)?$c?))", \/\/\u4e07\u4f4d\u6570\n\n            "$f\u4e2a\u767e\u5206\u70b9", \/\/\u767e\u5206\u70b9\n            "\u767e\u5206\u4e4b\u96f6\u70b9(\u51e0|(\u96f6|$c)+)", \/\/\u767e\u5206\u6570\n            "\u767e\u5206\u4e4b(($c?\u767e)|((($c\u5341)|\u5341)?$c?))", \/\/\u767e\u5206\u6570\n            "\u767e\u5206\u4e4b(($c?\u767e)|((($c\u5341)|\u5341)?$c?))($c|\u51e0|\u5de6\u53f3)?", \/\/\u767e\u5206\u6570\n\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u5e8f\u53f7\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            "\u7b2c$d", \/\/\u5e8f\u6570\n            "\u7b2c[\\\\s]?$d[\\\\s]?$q", \/\/\u5e8f\u6570\n\n            "\u7b2c$c", \/\/\u5e8f\u6570\n            "\u7b2c[\\\\s]?$c[\\\\s]?$q", \/\/\u5e8f\u6570\n\n            "\u7b2c\u5341$c?", \/\/\u5e8f\u6570\n            "\u7b2c[\\\\s]?\u5341$c?[\\\\s]?$q", \/\/\u5e8f\u6570\n\n            "\u7b2c$c\u5341$c?", \/\/\u5e8f\u6570\n            "\u7b2c[\\\\s]?$c\u5341$c?[\\\\s]?$q", \/\/\u5e8f\u6570\n\n            "\u7b2c$c\u767e(\u96f6|$c\u5341)?$c?", \/\/\u5e8f\u6570\n            "\u7b2c[\\\\s]?$c\u767e(\u96f6|$c\u5341)?$c?[\\\\s]?$q", \/\/\u5e8f\u6570\n\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u6570\u91cf\u8bcd\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            "$d($q|$u)", \/\/\u6570\u91cf\n\n            "$d[\\\\s]?[\u591a|\u4f59]($q|$u)", \/\/\u7ea6\u6570\n\n            "\u51e0(\u4ebf|\u4e07|\u5343|\u767e|\u5341)[\\\\s]?($q|$u)", \/\/\u7ea6\u6570\n\n            "$c[\\\\s]?($q|$u)", \/\/\u6570\u91cf\n\n            "\u5341$c?[\\\\s]?($q|$u)", \/\/\u6570\u91cf\n\n            "$c\u5341$c?[\\\\s]?($q|$u)", \/\/\u6570\u91cf\n\n            "$c\u767e(\u96f6|$c\u5341)?$c?[\\\\s]?($q|$u)", \/\/\u6570\u91cf\n\n            "$c\u5343(\u96f6|$c\u767e)?(\u96f6|$c\u5341)?$c?[\\\\s]?($q|$u)", \/\/\u6570\u91cf\n\n            "(\u5341|$c)\u4e07(\u96f6|$c\u5343)?(\u96f6|$c\u767e)?(\u96f6|$c\u5341)?$c?[\\\\s]?($q|$u)", \/\/\u6570\u91cf\n\n            "$f[\\\\s]?($v)", \/\/\u6570\u91cf\n\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u7ea6\u6570\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            "$f[\\\\s]?(\u4ebf|\u4e07|\u5343|\u767e)?(\u4f59|\u591a)?($q|$u)", \/\/\u7ea6\u6570\n\n            "(\u5341|$c)[\\\\s]?(\u4ebf|\u4e07|\u5343|\u767e|\u5341)(\u4f59|\u591a)?($q|$u)", \/\/\u7ea6\u6570\n\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u8d27\u5e01\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            "$f[\\\\s]?(\u767e|\u5343|\u4e07)?(\u4ebf|\u4e07)?(\u4f59|\u591a)?$y", \/\/\u8d27\u5e01\n\n            "$f[\\\\s]?[\u591a|\u4f59]?(\u767e|\u5343|\u4e07)?(\u4ebf|\u4e07)?(\u4f59|\u591a)?$y", \/\/\u8d27\u5e01\n\n            "[1-9]\\\\d{0,2}(,\\\\d{3})+(\u767e|\u5343|\u4e07)?(\u4ebf|\u4e07)?(\u4f59|\u591a)?$y", \/\/\u8d27\u5e01\n\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u65f6\u95f4\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            \/\/ \u65f6\u95f4\n            "[1-9]\\\\d*[\\\\s]?\u4e2a\u6708", \/\/\u6708\n            "[\u8fc7]?\u5e74[\u524d|\u4e2d|\u5e95|\u540e]", \/\/\u5e74\n            "(\u7b2c)?(\u4e00|\u4e8c|\u4e09|\u56db)\u5b63[\u5ea6]?", \/\/\u5b63\u5ea6\n            "\u5468[\u4e00|\u4e8c|\u4e09|\u56db|\u4e94|\u516d|\u65e5]", \/\/\u5468\n            "\u661f\u671f[\u4e00|\u4e8c|\u4e09|\u56db|\u4e94|\u516d|\u65e5]", \/\/\u661f\u671f\n            "((1[0-2])|(0?[1-9]))[\\\\s]?\u6708\u4efd", \/\/\u6708\u4efd\n            "([1-9]\\\\d*[\\\\s]?\u3001)+([1-9]\\\\d*[\\\\s]?)(\u4e00|\u4e8c|\u4e09|\u56db|\u4e94|\u516d|\u4e03|\u516b|\u4e5d|\u5341|\u5341\u4e00|\u5341\u4e8c)\u4e2a\u6708", \/\/\u6708\n            "(((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u3001)*((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u65e5", \/\/\u65e5\u671f\n            "((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u81f3[\\\\s]?((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u65e5", \/\/\u65e5\u671f\n            "((1[0-2])|(0?[1-9]))[\\\\s]?\u6708[\\\\s]?(((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u65e5)(\u3001((0?[1-9])|((1|2)[0-9])|30|31)[\\\\s]?\u65e5)*", \/\/\u65e5\u671f\n\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n            \/\/\n            \/\/ \u52a0\u8f7d\u5176\u4ed6\u89c4\u5219\u3002\n            \/\/\n            \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\n\n            \/\/ \u7279\u6b8a\u8bcd\u6c47\n            "[\u53cc|\u4e24|\u53c1|\u591a][\\\\s]?(\u65b9|\u70b9|\u4e2a|\u9879|\u6b21|\u6839|\u9897|\u6761|\u540d|\u5957|\u4efd)", \/\/\u6570\u91cf\n\n            \/\/ \u822a\u73ed\n            "$e+$n\u6b21[\\\\s]?\u822a\u73ed", \/\/\u822a\u73ed\n            \/\/ \u80a1\u7968\n            "(A|B|H)\u80a1", \/\/\u80a1\u7968\n            \/\/ \u6cb9\u6599\n            "$n#", \/\/\u6cb9\u6599\n            "$f\u4e2a\u6cb9", \/\/\u6cb9\u8017\n            \/\/ \u7ef4\u751f\u7d20\n            "\u7ef4\u751f\u7d20(A|B|C|D|E|K|H|P|M|T|U)\\\\d{0,2}", \/\/\u7ef4\u751f\u7d20\n        };\n<\/pre><\/div>\n\n\n
\u5728\u968f\u540e\u7684\u5b9e\u9645\u8fd0\u884c\u4e2d\uff0c\u5c31\u53d1\u73b0\u5904\u7406\u901f\u5ea6\u5341\u5206\u7f13\u6162\u3002\u521d\u6b65\u4f30\u7b97\u4e86\u4e00\u4e0b\uff0c\u8981\u5c06\u8bed\u6599\uff0810,101,283\u6761\uff09\u5168\u90e8\u5904\u7406\u5b8c\u6210\uff0c\u81f3\u5c11\u9700\u8981\u4e09\u5929\u7684\u65f6\u95f4\u3002<\/p>\n\n\n\n
\u901a\u8fc7Visual Studio Profiler\u63a2\u6d4b\u5de5\u5177\u5bf9CPU\u7684\u8fd0\u884c\u60c5\u51b5\u505a\u4e86\u68c0\u6d4b\uff0c\u6700\u540e\u53d1\u73b085%\u7684\u8fd0\u884c\u65f6\u95f4\u90fd\u88ab\u5206\u914d\u5230\u4e86\u8be5\u51fd\u6570\uff1a<\/p>\n\n\n
\n        private static Quantity[] GetMatches(string content)\n        {\n            \/\/ \u521b\u5efa\u5bf9\u8c61\n            List<Quantity> quantities = new List<Quantity>();\n            \/\/ \u5faa\u73af\u5904\u7406\n            foreach(string rule in QUANTITY_RULES)\n            {\n                \/\/ \u6062\u590d\u89c4\u5219\n                string newRule = Quantity.RecoverRule(rule);\n                \/\/ \u68c0\u67e5\u5339\u914d\n                \/\/ \u6b64\u51fd\u6570\u975e\u5e38\u6d88\u8017\u65f6\u95f4\uff01\uff01\uff01\n                Match match = Regex.Match(content, newRule);\n                \/\/ \u68c0\u67e5\u7ed3\u679c\n                while(match.Success)\n                {\n                    \/\/ \u521b\u5efa\u5bf9\u8c61\n                    Quantity quantity =\n                        new Quantity();\n                    \/\/ \u8bbe\u7f6e\u6570\u503c\n                    quantity.Index = match.Index;\n                    quantity.Value = match.Value;\n                    quantity.Length = match.Length;\n                    \/\/ \u52a0\u5165\u5217\u8868\n                    quantities.Add(quantity);\n                    \/\/ \u4e0b\u4e00\u4e2a\n                    match = match.NextMatch();\n                }\n            }\n            \/\/ \u6309\u7167\u7d22\u5f15\u6570\u503c\u6392\u5e8f\n            quantities = quantities.OrderBy(q => q.Index).ToList();\n            \/\/ \u8fd4\u56de\u7ed3\u679c\n            return quantities.ToArray();\n        }\n<\/pre><\/div>\n\n\n
\u5176\u4e2d\u4ee5Regex.Match\u5360\u7528\u65f6\u95f4\u6700\u591a\u3002\u8fd9\u4e2a\u53ef\u662f\u5185\u7f6e\u7cfb\u7edf\u51fd\u6570\uff0c\u4f5c\u8005\u8868\u793a\u65e0\u80fd\u4e3a\u529b\u3002<\/p>\n\n\n\n
\u4e0d\u8fc7\u968f\u540e\u53d1\u73b0\uff1a\u7a0b\u5e8f\u8fd0\u884c\u7684CPU\u5360\u6bd4\u5e76\u4e0d\u9ad8\uff0c\u5927\u7ea6\u53ea\u670912%\u5de6\u53f3\u3002\u4e8e\u662f\u5c31\u60f3\u5c1d\u8bd5\u4e0b\u591a\u7ebf\u7a0b\u5e76\u884c\u5904\u7406\u3002\u6700\u540e\u7ec8\u4e8e\u5728\u7f51\u7edc\u4e0a\u627e\u5230\u4e86\u5408\u9002\u7684\u89e3\u51b3\u65b9\u6848\u3002\u8fd9\u91cc\u9700\u8981\u7528\u5230\u4e00\u4e2a\u7c7bLimitedConcurrencyLevelTaskScheduler\uff1a<\/p>\n\n\n
\nusing System;\nusing System.Text;\nusing System.Data;\nusing System.Threading;\nusing System.Data.SqlClient;\nusing System.Threading.Tasks;\nusing System.Collections.Generic;\nusing System.Text.RegularExpressions;\n\nnamespace Misc\n{\n    \/\/ Provides a task scheduler that ensures a maximum concurrency level while\n    \/\/ running on top of the thread pool.\n    public class LimitedConcurrencyLevelTaskScheduler : TaskScheduler\n    {\n        \/\/ Indicates whether the current thread is processing work items.\n        [ThreadStatic]\n        private static bool _currentThreadIsProcessingItems;\n\n        \/\/ The list of tasks to be executed\n        private readonly LinkedList<Task> _tasks = new LinkedList<Task>(); \/\/ protected by lock(_tasks)\n\n        \/\/ The maximum concurrency level allowed by this scheduler.\n        private readonly int _maxDegreeOfParallelism;\n\n        \/\/ Indicates whether the scheduler is currently processing work items.\n        private int _delegatesQueuedOrRunning = 0;\n\n        \/\/ Creates a new instance with the specified degree of parallelism.\n        public LimitedConcurrencyLevelTaskScheduler(int maxDegreeOfParallelism)\n        {\n            if (maxDegreeOfParallelism < 1) throw new ArgumentOutOfRangeException("maxDegreeOfParallelism");\n            _maxDegreeOfParallelism = maxDegreeOfParallelism;\n        }\n\n        \/\/ Queues a task to the scheduler.\n        protected sealed override void QueueTask(Task task)\n        {\n            \/\/ Add the task to the list of tasks to be processed.  If there aren't enough\n            \/\/ delegates currently queued or running to process tasks, schedule another.\n            lock (_tasks)\n            {\n                _tasks.AddLast(task);\n                if (_delegatesQueuedOrRunning < _maxDegreeOfParallelism)\n                {\n                    ++_delegatesQueuedOrRunning;\n                    NotifyThreadPoolOfPendingWork();\n                }\n            }\n        }\n\n        \/\/ Inform the ThreadPool that there's work to be executed for this scheduler.\n        private void NotifyThreadPoolOfPendingWork()\n        {\n            ThreadPool.UnsafeQueueUserWorkItem(_ =>\n            {\n                \/\/ Note that the current thread is now processing work items.\n                \/\/ This is necessary to enable inlining of tasks into this thread.\n                _currentThreadIsProcessingItems = true;\n                try\n                {\n                    \/\/ Process all available items in the queue.\n                    while (true)\n                    {\n                        Task item;\n                        lock (_tasks)\n                        {\n                            \/\/ When there are no more items to be processed,\n                            \/\/ note that we're done processing, and get out.\n                            if (_tasks.Count == 0)\n                            {\n                                --_delegatesQueuedOrRunning;\n                                break;\n                            }\n\n                            \/\/ Get the next item from the queue\n                            item = _tasks.First.Value;\n                            _tasks.RemoveFirst();\n                        }\n\n                        \/\/ Execute the task we pulled out of the queue\n                        base.TryExecuteTask(item);\n                    }\n                }\n                \/\/ We're done processing items on the current thread\n                finally { _currentThreadIsProcessingItems = false; }\n            }, null);\n        }\n\n        \/\/ Attempts to execute the specified task on the current thread.\n        protected sealed override bool TryExecuteTaskInline(Task task, bool taskWasPreviouslyQueued)\n        {\n            \/\/ If this thread isn't already processing a task, we don't support inlining\n            if (!_currentThreadIsProcessingItems) return false;\n\n            \/\/ If the task was previously queued, remove it from the queue\n            if (taskWasPreviouslyQueued)\n                \/\/ Try to run the task.\n                if (TryDequeue(task))\n                    return base.TryExecuteTask(task);\n                else\n                    return false;\n            else\n                return base.TryExecuteTask(task);\n        }\n\n        \/\/ Attempt to remove a previously scheduled task from the scheduler.\n        protected sealed override bool TryDequeue(Task task)\n        {\n            lock (_tasks) return _tasks.Remove(task);\n        }\n\n        \/\/ Gets the maximum concurrency level supported by this scheduler.\n        public sealed override int MaximumConcurrencyLevel { get { return _maxDegreeOfParallelism; } }\n\n        \/\/ Gets an enumerable of the tasks currently scheduled on this scheduler.\n        protected sealed override IEnumerable<Task> GetScheduledTasks()\n        {\n            bool lockTaken = false;\n            try\n            {\n                Monitor.TryEnter(_tasks, ref lockTaken);\n                if (lockTaken) return _tasks;\n                else throw new NotSupportedException();\n            }\n            finally\n            {\n                if (lockTaken) Monitor.Exit(_tasks);\n            }\n        }\n    }\n}\n<\/pre><\/div>\n\n\n
\u901a\u8fc7\u8fd9\u4e2a\u7c7b\u53ef\u4ee5\u5b9e\u73b0\u63a7\u5236\u7ebf\u7a0b\u6c60\u7684\u5927\u5c0f\u3002<\/p>\n\n\n
\n\t\tpublic static void ExtractSentences(bool flag)\n        {\n            \/\/ \u8bb0\u5f55\u65e5\u5fd7\n            Log.LogMessage("SentenceExtractor", "ExtractSentences", "\u63d0\u53d6\u53e5\u5b50\u6570\u636e\uff01");\n\n            \/\/\u6307\u4ee4\u5b57\u7b26\u4e32\n            string cmdString =\n                  "SELECT [rid], [content] FROM [dbo].[RawContent];";\n\n            \/\/ \u521b\u5efa\u6570\u636e\u5e93\u8fde\u63a5\n            SqlConnection sqlConnection = new SqlConnection(Common.CONNECT_STRING);\n\n            try\n            {\n\t\t\t\t\/\/ \u8ba1\u6570\u5668\n\t\t\t\tint taskCount = 0;\n\t\t\t\t\/\/ \u540c\u6b65\u9501\u5bf9\u8c61\n\t\t\t\tobject lockObject = new object();\n\t\t\t\t\/\/ \u4efb\u52a1\u6570\u7ec4\n\t\t\t\tList<Task> tasks = new List<Task>();\n\t\t\t\t\/\/ \u751f\u6210\u4efb\u52a1\u63a7\u5236\u5668\n\t\t\t\tTaskFactory factory = new TaskFactory(\n\t\t\t\t\tnew LimitedConcurrencyLevelTaskScheduler(Common.MAX_THREADS));\n\t\t\t\t\n\t\t\t\t\/\/ \u8ba1\u6570\u5668\n\t\t\t\tint total = 0;\n                \/\/ \u5f00\u542f\u6570\u636e\u5e93\u8fde\u63a5\n                sqlConnection.Open();\n                \/\/ \u521b\u5efa\u6307\u4ee4\n                SqlCommand sqlCommand =\n                    new SqlCommand(cmdString, sqlConnection);\n                \/\/ \u521b\u5efa\u6570\u636e\u9605\u8bfb\u5668\n                SqlDataReader reader = sqlCommand.ExecuteReader();\n                \/\/ \u5faa\u73af\u5904\u7406\n                while (reader.Read())\n                {\n                    \/\/ \u589e\u52a0\u8ba1\u6570\n                    total++;\n                    \/\/ \u68c0\u67e5\u8ba1\u6570\n                    if (total % 10000 == 0)\n                    {\n                        \/\/ \u6253\u5370\u8bb0\u5f55\n                        Log.LogMessage(string.Format("{0} items processed !", total));\n                    }\n\n                    \/\/ \u83b7\u5f97\u7f16\u53f7\n                    int rid = reader.GetInt32(0);\n                    \/\/ \u83b7\u5f97\u5185\u5bb9\n                    string content = reader.GetString(1);\n                    \/\/ \u68c0\u67e5\u5185\u5bb9\n                    if (content == null || content.Length <= 0) continue;\n\n\t\t\t\t\t\/\/ \u542f\u52a8\u8fdb\u7a0b\n\t\t\t\t\ttasks.Add(factory.StartNew\n\t\t\t\t\t(() =>\n\t\t\t\t\t{\n\t\t\t\t\t\tlock (lockObject)\n\t\t\t\t\t\t{\n\t\t\t\t\t\t\t\/\/ \u589e\u52a0\u8ba1\u6570\n\t\t\t\t\t\t\ttaskCount++;\n\t\t\t\t\t\t\t\/\/ \u68c0\u67e5\u8ba1\u6570\n\t\t\t\t\t\t\tif (taskCount % 10000 == 0)\n\t\t\t\t\t\t\t{\n\t\t\t\t\t\t\t\t\/\/ \u6253\u5370\u8bb0\u5f55\n\t\t\t\t\t\t\t\tLog.LogMessage(string.Format("{0} tasks finised !", taskCount));\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t}\n\n\t\t\t\t\t\t\/\/ \u6e05\u7406\u5185\u5bb9\n\t\t\t\t\t\tcontent = MiscTool.ClearContent(content);\n\t\t\t\t\t\t\/\/ \u68c0\u67e5\u5185\u5bb9\n\t\t\t\t\t\tif (content == null || content.Length <= 0) return;\n\n\t\t\t\t\t\t\/\/ \u5207\u5206\u5b57\u7b26\u4e32\n\t\t\t\t\t\tstring[] output = SplitContent(content);\n\t\t\t\t\t\t\/\/ \u68c0\u67e5\u7ed3\u679c\n\t\t\t\t\t\tif (output == null || output.Length <= 0) return;\n\n\t\t\t\t\t\t\/\/ \u5408\u5e76\u5b57\u7b26\u4e32\n\t\t\t\t\t\toutput = MergeContent(output);\n\t\t\t\t\t\t\/\/ \u68c0\u67e5\u7ed3\u679c\n\t\t\t\t\t\tif (output == null || output.Length <= 0) return;\n\n\t\t\t\t\t\t\/\/ \u83b7\u5f97\u62c6\u5206\u7ed3\u679c\n\t\t\t\t\t\tstring[][] sentences = GetSentences(output);\n\t\t\t\t\t\t\/\/ \u68c0\u67e5\u7ed3\u679c\n\t\t\t\t\t\tif (sentences == null || sentences.Length <= 0) return;\n\n\t\t\t\t\t\t\/\/ \u68c0\u67e5\u6807\u8bb0\u4f4d\n\t\t\t\t\t\tif (flag)\n\t\t\t\t\t\t{\n\t\t\t\t\t\t\t\/\/ \u5faa\u73af\u5904\u7406\n\t\t\t\t\t\t\tforeach (string[] sentence in sentences)\n\t\t\t\t\t\t\t{\n\t\t\t\t\t\t\t\t\/\/ \u52a0\u5165\u8bb0\u5f55\n\t\t\t\t\t\t\t\tSentenceContent.AddSentence(Concatenate(sentence), rid);\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t}\n\t\t\t\t\t}));\n                }\n                \/\/ \u5173\u95ed\u6570\u636e\u9605\u8bfb\u5668\n                reader.Close();\n                \/\/ \u8bb0\u5f55\u65e5\u5fd7\n                Log.LogMessage(string.Format("\\tcontent.count = {0}", total));\n\n\t\t\t\t\/\/ \u8bb0\u5f55\u65e5\u5fd7\n\t\t\t\tLog.LogMessage("SentenceExtractor", "ExtractSentences", "\u7b49\u5f85\u6240\u6709\u4efb\u52a1\u5b8c\u6210 \uff01");\n\t\t\t\t\/\/ \u7b49\u5f85\u6240\u6709\u4efb\u52a1\u5b8c\u6210\n\t\t\t\tTask.WaitAll(tasks.ToArray());\n\t\t\t}\n\t\t\tcatch (System.Exception ex)\n            {\n                \/\/ \u8bb0\u5f55\u65e5\u5fd7\n                Log.LogMessage("SentenceExtractor", "ExtractSentences", "unexpected exit \uff01");\n                Log.LogMessage(string.Format("\\texception.message = {0}", ex.Message));\n            }\n            finally\n            {\n                \/\/ \u68c0\u67e5\u72b6\u6001\u5e76\u5173\u95ed\u8fde\u63a5\n                if (sqlConnection.State == ConnectionState.Open) sqlConnection.Close();\n            }\n\n            \/\/ \u8bb0\u5f55\u65e5\u5fd7\n            Log.LogMessage("SentenceExtractor", "ExtractSentences", "\u53e5\u5b50\u63d0\u53d6\u5b8c\u6bd5\uff01");\n        }\n<\/pre><\/div>\n\n\n
\u7edf\u8ba1\u7a0b\u5e8f\u6700\u7ec8\u4e00\u53e3\u6c14\u542f\u52a8\u4e861000\u591a\u4e07\u4e2a\u4efb\u52a1\uff0c\u7136\u540eWaitAll\u53bb\u5b8c\u6210\u3002\u8fd9\u6b63\u662f\u4e3a\u641c\u522e\u8ba1\u7b97\u673a\u7684\u8d44\u6e90\u7ede\u5c3d\u8111\u6c41\u3002<\/p>\n\n\n\n
\u6700\u540e\u5b9e\u9645\u8fd0\u884c\u7684\u65f6\u5019\uff0cCPU\u6574\u4f53\u6d88\u8017\u7ec8\u4e8e\u722c\u523080%\u4ee5\u4e0a\uff08\u7559\u70b9\u4f59\u5730\u505a\u5176\u4ed6\u4e8b\u60c5\uff09\u3002\u5927\u7ea616\u4e2a\u5c0f\u65f6\u5de6\u53f3\u5c31\u5b8c\u6210\u4e86\u7edf\u8ba1\u4efb\u52a1\uff0c\u6bd4\u4e4b\u524d\u7684\u65b9\u5f0f\u8981\u5feb6\u500d\u3002<\/p>\n\n\n\n
\u4e00\u53e3\u6c14\u542f\u52a81000\u591a\u4e07\u4e2a\u4efb\u52a1\uff0c\u8fd8\u4e0d\u662f\u6700\u5938\u5f20\u7684\u4e8b\u60c5\uff01<\/p>\n\n\n\n
\uff083\uff09\u7edf\u8ba1\u77ed\u53e5\u9891\u6b21<\/p>\n\n\n\n
\u5230\u4e86\u8fd9\u91cc\uff0c\u624d\u662f\u771f\u6b63\u9047\u5230\u4e86\u201c\u74f6\u9888\u201d\u3002\u8981\u5206\u679048,218,815\u6761\u4e0d\u540c\u7684\u77ed\u53e5\u5728\u8bed\u6599\uff0810,101,283\u6761\uff09\u4e2d\u7684\u9891\u6b21\uff0c\u8fd9\u4e2a\u5339\u914d\u8fd0\u7b97\u6570\u91cf\u7ea7\u51e0\u4e4e\u8981\u4e0a\u5929\uff01<\/p>\n\n\n\n
\u4e3a\u4e86\u63d0\u9ad8\u8fd0\u884c\u6548\u7387\uff0c\u7167\u4e0a\u9762\u7684\u65b9\u6cd5\u4f7f\u7528\u591a\u4efb\u52a1\u3002\u4f46\u662f\u6b64\u6b21\u6ca1\u6709\u540c\u65f6\u5f00\u542f4800\u591a\u4e07\u4e2a\u4efb\u52a1\u3002\u800c\u662f\u8981\u63a7\u5236\u4efb\u52a1\u961f\u5217\u7684\u6570\u91cf\u3002\u4e3a\u6b64\u589e\u52a0\u4e86\u5982\u4e0b\u4e00\u6bb5\u4ee3\u7801\uff1a<\/p>\n\n\n
\n                            \/\/ \u68c0\u67e5\u4efb\u52a1\u961f\u5217\u957f\u5ea6\n                            while(tasks.Count > 10000000)\n                            {\n                                \/\/ \u8bb0\u5f55\u65e5\u5fd7\n                                Log.LogMessage("SentenceStatistic", "MakeStatistic", "\u4efb\u52a1\u5df2\u6ee1\uff0c\u9700\u8981\u7b49\u5f85\u5168\u90e8\u7ed3\u675f\uff01");\n                                \/\/ \u7b49\u5f85\u5168\u90e8\u4efb\u52a1\u7ed3\u675f\n                                Task.WaitAll(tasks.ToArray()); tasks.Clear();\n                                \/\/ \u8bb0\u5f55\u65e5\u5fd7\n                                Log.LogMessage("SentenceStatistic", "MakeStatistic", "\u4efb\u52a1\u5168\u90e8\u7ed3\u675f\uff0c\u7ee7\u7eed\u5206\u914d\u65b0\u4efb\u52a1\uff01");\n                            }\n\n<\/pre><\/div>\n\n\n
\u8fd8\u662f\u540c\u65f6\u5f00\u542f1000\u591a\u4e07\u4e2a\u4efb\u52a1\uff0c\u7136\u540e\u7b49\u5f85\u5b8c\u6210\u3002\u8fd9\u6b21\u7531\u4e8e\u5339\u914d\u6bd4\u8f83\u7b80\u5355\uff0c\u6ca1\u6709\u91c7\u7528Regex.Match\u51fd\u6570\uff0c\u5426\u5219\u6548\u7387\u53ef\u80fd\u8fd8\u8981\u66f4\u4f4e\u3002<\/p>\n\n\n\n
\u6839\u636e\u5b9e\u9645\u8fd0\u884c\u7684\u6548\u679c\uff0c\u4f5c\u8005\u4f30\u7b97\uff1a\u5b8c\u6210\u5168\u90e8\u7684\u7edf\u8ba1\u5206\u6790\u9700\u89813\u5e74\u7684\u65f6\u95f4\uff01\u8fd9\u6837\u6765\u770b\uff0c\u4e0d\u8981\u5f00\u542f\u8fc7\u591a\u7684\u9891\u6b21\u7edf\u8ba1\uff0c\u5426\u5219\u8ba1\u7b97\u91cf\u4e0d\u662f\u4e00\u822c\u7684\u5927\u3002\u73b0\u5728\u9700\u8981\u51b3\u5b9a\uff1a\u53ea\u7edf\u8ba1\u591a\u957f\u7684\u5b57\u7b26\u4e32\u5c31\u8db3\u591f\u4e86\u3002\u5426\u5219\uff0c\u8fd9\u53f0\u8bbe\u5907\u662f\u65e0\u6cd5\u5b8c\u6210\u4efb\u52a1\u7684\u3002<\/p>\n\n\n\n
\u4e09\u3001\u75af\u72c2\u7684\u603b\u7ed3<\/h2>\n\n\n\n
\u5728\u4e0a\u8ff0\u7684\u5185\u5bb9\u4e2d\uff0c\u505a\u4e86\u4e00\u4e9b\u6bd4\u8f83\u75af\u72c2\u7684\u4e8b\u60c5\uff0c\u4f46\u662fWindows 11\u4e00\u5207\u8fd0\u884c\u6b63\u5e38\uff0c\u5e76\u672a\u51fa\u73b0\u4efb\u4f55\u95ee\u9898\u3002<\/p>\n\n\n\n