Java Code Examples for org.ansj.domain.Result#size()

The following examples show how to use org.ansj.domain.Result#size() . You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may check out the related API usage on the sidebar.

Example 1

Source File: WordSegmenter.java From SnowGraph with Apache License 2.0

6 votes

private static void tokenizeDocxFile(String filePath) {
    File file = new File(filePath);
    DocumentInfo doc = DocumentParser.parseFileToDocumentInfo(file);
    if(doc instanceof WordDocumentInfo) {
        String content = ((WordDocumentInfo) doc).getDocStr();
        Result terms = ToAnalysis.parse(content);
        for (int i = 0; i < terms.size(); i++) {
            String words = terms.get(i).getName();
            boolean filtered = false;
            for(String stopToken : stopTokens)
                if(words.equals(stopToken)) { filtered = true; break; }
            char firstLetter = words.charAt(0);
            if((firstLetter >= 'A' && firstLetter <= 'Z') ||
                    (firstLetter >= 'a' && firstLetter <= 'z') ||
                    (firstLetter >= '0' && firstLetter <= '9'))
                filtered = true;
            if(filtered) continue;
            wordsCN.add(words);
        }
    }
    else System.out.println("Not a docx file");
}

Example 2

Source File: WordSegmenter.java From SnowGraph with Apache License 2.0

5 votes

public static ArrayList<String> demo(String strToParse) {
    String str = strToParse;
            //"我年纪还轻，阅历不深的时候，我父亲教导过我一句话，我至今还念念不忘。 \n" +
            //"“每逢你想要批评任何人的时候，”他对我说，“你就记住，这个世界上所有的人，并不是个个都有过你拥有的那些优越的条件。”";
    ArrayList<String> ret = new ArrayList<>();
    Result terms = ToAnalysis.parse(str);
    for (int i = 0; i < terms.size(); i++) {
        String words = terms.get(i).getName();// 获取单词
        String nominal = terms.get(i).getNatureStr();// 获取词性
        ret.add(words);
        //System.out.print(words + "\t" + nominal + "\n");
    }
    return ret;
}