HTML text extraction in j2me without any HTML tags

问题:

I have a string from an RSS file after parsing.

String htmlString=

<p><img border="1" align="left" width="200" vspace="2" hspace="2" height="133" alt="Prime Minister Manmohan Singh will leave for Iran on August   28, 2012 to attend the Non-Aligned Movement summit, which will   be preceded by crucial bilateral talks with Iran&rsquo;s supreme   leader Ayotollah Ali Khamenei and Iranian President Mahmoud   Ahmadinejad." src="/tmdbuserfiles/manmohan ahmadi(3).jpg" />Prime Minister summit, which will be preceded by crucial bilateral talks with Iran&rsquo;s supreme leader place at a time when the U.S. is pushing India to reduce engagement with Iran and implement sanctions imposed by some countries over its controversial nuclear programme.<br />
    <br />
    &nbsp;</p>

I have a requirement to display the text without any HTML tags and without HTML special characters from above htmlString on my LWUIT Form like:

Prime Minister ManmohanSingh will leave for Iran on August 28, 2012 to attend the Non-Aligned Movement summit, which will  
be preceded by crucial bilateral talks with Iran supreme leader Ayotollah Ali Khamenei and Iranian etc...........?

回答1:

It also helps to open the HttpConnection input stream with UTF-8 encoding like this :

String encoding = "UTF-8";
Reader reader = new InputStreamReader(in, encoding);

Use this suite of String Utils to get clean and well formatted text. :

 /**
 * Method removes HTML tags from given string.
 *
 * @param text  Input parameter containing HTML tags (eg. <b>cat</b>)
 * @return      String without HTML tags (eg. cat)
 */
public static String removeHtml(String text) {
    try {
        int idx = text.indexOf("<");
        if (idx == -1) {
            text = decodeEntities(text);
            return text;
        }

        String plainText = "";
        String htmlText = text;
        int htmlStartIndex = htmlText.indexOf("<", 0);
        if (htmlStartIndex == -1) {
            return text;
        }
        htmlText = StringUtils.replace(htmlText, "</p>", "\r\n");
        htmlText = StringUtils.replace(htmlText, "<br/>", "\r\n");
        htmlText = StringUtils.replace(htmlText, "<br>", "\r\n");
        while (htmlStartIndex >= 0) {
            plainText += htmlText.substring(0, htmlStartIndex);
            int htmlEndIndex = htmlText.indexOf(">", htmlStartIndex);
            htmlText = htmlText.substring(htmlEndIndex + 1);
            htmlStartIndex = htmlText.indexOf("<", 0);
        }
        plainText = plainText.trim();
        plainText = decodeEntities(plainText);
        return plainText;
    } catch (Exception e) {
        System.err.println("Error while removing HTML: " + e.toString());
        return text;
    }
}

public static String decodeEntities(String html) {
    String result = StringUtils.replace(html, "&lt;", "<");
    result = StringUtils.replace(result, "&gt;", ">");
    result = StringUtils.replace(result, "&nbsp;", " ");
    result = StringUtils.replace(result, "&amp;", "&");
    result = StringUtils.replace(result, "&auml;", "ä");
    result = StringUtils.replace(result, "&ouml;", "ö");
    result = StringUtils.replace(result, "&quot;", "'");
    result = StringUtils.replace(result, "&lquot;", "'");
    result = StringUtils.replace(result, "&rquot;", "'");
    result = StringUtils.replace(result, "&#xd;", "\r");
    return result;
}

/* Replace all instances of a String in a String.
 *   @param  s  String to alter.
 *   @param  f  String to look for.
 *   @param  r  String to replace it with, or null to just remove it.
 */
public static String replace(String s, String f, String r) {
    if (s == null) {
        return s;
    }
    if (f == null) {
        return s;
    }
    if (r == null) {
        r = "";
    }
    int index01 = s.indexOf(f);
    while (index01 != -1) {
        s = s.substring(0, index01) + r + s.substring(index01 + f.length());
        index01 += r.length();
        index01 = s.indexOf(f, index01);
    }
    return s;
}

public static String cleanEncodedString(String str) {
    String resultStr = str;
    String encoding = "UTF-8";

    InputStream in = new ByteArrayInputStream(str.getBytes());
    InputStreamReader isr;
    try {
        isr = new InputStreamReader(in, encoding);

        ByteArrayOutputStream buf = new ByteArrayOutputStream();
        int result = isr.read();
        while (result != -1) {
            byte b = (byte) result;
            buf.write(b);
            result = isr.read();
        }
        resultStr = buf.toString();

        return resultStr;
    } catch (Exception uee) {
        uee.printStackTrace();
    }
    return resultStr;
}

回答2:

    int startIndex = htmlString.indexOf("alt=\"");
    int endIndex = htmlString.indexOf("\"", startIndex+5);
    String resultantString = htmlString.substring(startIndex+5, endIndex);
    resultantString = replaceAll(resultantString ,""","\"");
    resultantString = replaceAll(resultantString ,"&","&");
    resultantString = replaceAll(resultantString ,"’","’");




private String replaceAll(String source, String pattern, String replacement) {
        if (source == null) {
            return "";
        }
        StringBuffer sb = new StringBuffer();
        int index;
        int patIndex = 0;
        while ((index = source.indexOf(pattern, patIndex)) != -1) {
            sb.append(source.substring(patIndex, index));
            sb.append(replacement);
            patIndex = index + pattern.length();
        }
        sb.append(source.substring(patIndex));
        return sb.toString();
    }