0

(I am aware that regex is not the recommended way to deal with html, but this is my assignment)

I need a regex in Java that will capture html tags and their attributes. I am trying to achieve this with one regex using groups. I expected this regex to work:

<(?!!)(?!/)\s*(\w+)(?:\s*(\S+)=['"]{1}[^>]*?['"]{1})*\s*>
<                                                            the tag starts with <
 (?!!)                                                       I dont't want comments
      (?!/)                                                  I dont't want closing tags
           \s*                                               any number of white spaces 
              (\w+)                                          the tag
                   (?:                                       do not capture the following group
                      \s*                                    any number of white spaces before the first attribute
                         (\S+)                               capture the attributes name
                              =['"]{1}[^>]*?['"]{1}          the ="bottm" or ='bottm' etc.
                                                   )*        close the not-capturing group, it can occure multiple times or zero times
                                                     \s*     any white spaces before the closing of the tag
                                                        >    close the tag

I expected the result for a tag like:

<div id="qwerty" class='someClass' >
group(1) = "div"
group(2) = "id"
group(3) = "class"

but the result is:

group(1) = "div"
group(2) = "class"

Is seems that it is not possible to capture a group multiple times (...)*, is this correct?

As for now I use a reg ex like: 

<(?!!)(?!/)\s*(\w+) (?:\s*(\S+)=['"]{1}[^>]*?['"]{1}){0,1} (?:\s*(\S+)=['"]{1}[^>]*?['"]{1}){0,1} (...){0,1} (...){0,1} ... \s*>

I repeat the capturing group for the attribute multiple times and get results like:

<div id="qwerty" class='someClass' >
group(1) = "div"
group(2) = "id"
group(3) = "class" 
group(4) = null 
group(5) = null 
group(6) = null 
...

What other approaches can I use? (I can use multiple regexes, but it is preferred to do it with just one)

4

1 回答 1

2

似乎不可能多次使用匹配一个匹配组。所以使用的结果

(..regex for group...)*

仍将只是一个匹配组。

在第一步中捕获整个标签的代码，然后捕获所有属性：

URL url = new URL("http://stackoverflow.com/");
URLConnection connection = url.openConnection();
BufferedReader reader = new BufferedReader(new InputStreamReader(connection.getInputStream()));
StringBuilder stringBuilder = new StringBuilder();
String inputLine;
while ((inputLine = reader.readLine()) != null) {
    stringBuilder.append(inputLine);
}
String pageContent = stringBuilder.toString();
Pattern pattern = Pattern.compile("<(?!!)(?!/)\\s*([a-zA-Z0-9]+)(.*?)>");
Matcher matcher = pattern.matcher(pageContent);
while (matcher.find()) {
    String tagName = matcher.group(1);
    String attributes = matcher.group(2);
    System.out.println("tag name: " + tagName);
    System.out.println("     rest of the tag: " + attributes);
    Pattern attributePattern = Pattern.compile("(\\S+)=['\"]{1}([^>]*?)['\"]{1}");
    Matcher attributeMatcher = attributePattern.matcher(attributes);
    while(attributeMatcher.find()) {
        String attributeName = attributeMatcher.group(1);
        String attributeValue = attributeMatcher.group(2);
        System.out.println("         attribute name: " + attributeName + "    value: " + attributeValue);
    }
}
于 2013-11-02T00:56:32.283 回答