Java Code Examples for com.networknt.utility.StringUtils#substringBefore()

/**
 * <p>Removes empty parameters.</p>
 * <code>http://www.example.com/display?a=b&amp;a=&amp;c=d&amp;e=&amp;f=g
 * &rarr; http://www.example.com/display?a=b&amp;c=d&amp;f=g</code>
 * @return this instance
 */
public URLNormalizer removeEmptyParameters() {
    // Does it have query parameters?
    if (!url.contains("?")) {
        return this;
    }
    // It does, so proceed
    List<String> keyValues = new ArrayList<>();
    String queryString = StringUtils.substringAfter(url, "?");
    String[] params = StringUtils.split(queryString, '&');
    for (String param : params) {
        if (param.contains("=")
                && StringUtils.isNotBlank(
                StringUtils.substringAfter(param, "="))
                && StringUtils.isNotBlank(
                StringUtils.substringBefore(param, "="))) {
            keyValues.add(param);
        }
    }
    String cleanQueryString = StringUtils.join(keyValues, '&');
    if (StringUtils.isNotBlank(cleanQueryString)) {
        url = StringUtils.substringBefore(
                url, "?") + "?" + cleanQueryString;
    }
    return this;
}
 

/**
 * <p>Removes a URL-based session id.  It removes PHP (PHPSESSID),
 * ASP (ASPSESSIONID), and Java EE (jsessionid) session ids.</p>
 * <code>http://www.example.com/servlet;jsessionid=1E6FEC0D14D044541DD84D2D013D29ED?a=b
 * &rarr; http://www.example.com/servlet?a=b</code>
 * <p><b>Please Note:</b> Removing session IDs from URLs is often
 * a good way to have the URL return an error once invoked.</p>
 * @return this instance
 */
public URLNormalizer removeSessionIds() {
    if (StringUtils.containsIgnoreCase(url, ";jsessionid=")) {
        url = url.replaceFirst(
                "(;jsessionid=([A-F0-9]+)((\\.\\w+)*))", "");
    } else {
        String u = StringUtils.substringBefore(url, "?");
        String q = StringUtils.substringAfter(url, "?");
        if (StringUtils.containsIgnoreCase(url, "PHPSESSID=")) {
            q = q.replaceFirst("(&|^)(PHPSESSID=[0-9a-zA-Z]*)", "");
        } else if (StringUtils.containsIgnoreCase(url, "ASPSESSIONID")) {
            q = q.replaceFirst(
                    "(&|^)(ASPSESSIONID[a-zA-Z]{8}=[a-zA-Z]*)", "");
        }
        if (!StringUtils.isBlank(q)) {
            u += "?" + StringUtils.removeStart(q, "&");
        }
        url = u;
    }
    return this;
}
 

/**
 * Converts a relative URL to an absolute one, based on the supplied
 * base URL. The base URL is assumed to be a valid URL. Behavior
 * is unexpected when base URL is invalid.
 * @param baseURL URL to the reference is relative to
 * @param relativeURL the relative URL portion to transform to absolute
 * @return absolute URL
 * @since 1.8.0
 */
public static String toAbsolute(String baseURL, String relativeURL) {
    String relURL = relativeURL;
    // Relative to protocol
    if (relURL.startsWith("//")) {
        return StringUtils.substringBefore(baseURL, "//") + "//"
                + StringUtils.substringAfter(relURL, "//");
    }
    // Relative to domain name
    if (relURL.startsWith("/")) {
        return getRoot(baseURL) + relURL;
    }
    // Relative to full full page URL minus ? or #
    if (relURL.startsWith("?") || relURL.startsWith("#")) {
        // this is a relative url and should have the full page base
        return baseURL.replaceFirst("(.*?)([\\?\\#])(.*)", "$1") + relURL;
    }

    // Relative to last directory/segment
    if (!relURL.contains("://")) {
        String base = baseURL.replaceFirst("(.*?)([\\?\\#])(.*)", "$1");
        if (StringUtils.countMatches(base, '/') > 2) {
            base = base.replaceFirst("(.*/)(.*)", "$1");
        }
        if (base.endsWith("/")) {
            // This is a URL relative to the last URL segment
            relURL = base + relURL;
        } else {
            relURL = base + "/" + relURL;
        }
    }

    // Not detected as relative, so return as is
    return relURL;
}
 

/**
 * Constructor.
 * It is possible to only supply a query string as opposed to an
 * entire URL.
 * Key and values making up a query string are assumed to be URL-encoded.
 * Will throw a {@link RuntimeException} if the supplied encoding is
 * unsupported or invalid.
 * @param urlWithQueryString a URL from which to extract a query string.
 * @param encoding character encoding
 */
public QueryString(String urlWithQueryString, String encoding) {
    if (StringUtils.isBlank(encoding)) {
        this.encoding = StandardCharsets.UTF_8.toString();
    } else {
        this.encoding = encoding;
    }
    String paramString = urlWithQueryString;
    if (paramString.contains("?")) {
        paramString = StringUtils.substringBefore(paramString, "#");
        paramString = paramString.replaceAll("(.*?)(\\?)(.*)", "$3");
    }
    String[] paramParts = paramString.split("\\&");
    for (int i = 0; i < paramParts.length; i++) {
        String paramPart = paramParts[i];
        if (StringUtils.isBlank(paramPart)) {
            continue;
        }
        String key;
        String value;
        if (paramPart.contains("=")) {
            key = StringUtils.substringBefore(paramPart, "=");
            value = StringUtils.substringAfter(paramPart, "=");
        } else {
            key = paramPart;
            value = StringUtils.EMPTY;
        }
        try {
            addString(URLDecoder.decode(key, this.encoding),
                    URLDecoder.decode(value, this.encoding));
        } catch (UnsupportedEncodingException e) {
            throw new RuntimeException(
                    "Cannot URL-decode query string (key="
                            + key + "; value=" + value + ").", e);
        }
    }
}
 

/**
 * Apply this url QueryString on the given URL. If a query string already
 * exists, it is replaced by this one.
 * @param url the URL to apply this query string.
 * @return url with query string added
 */
public String applyOnURL(String url) {
    if (StringUtils.isBlank(url)) {
        return url;
    }
    return StringUtils.substringBefore(url, "?") + toString();
}
 

/**
 * <p>Sorts query parameters.</p>
 * <code>http://www.example.com/?z=bb&amp;y=cc&amp;z=aa &rarr;
 *       http://www.example.com/?y=cc&amp;z=bb&amp;z=aa</code>
 * @return this instance
 */
public URLNormalizer sortQueryParameters() {
    // Does it have query parameters?
    if (!url.contains("?")) {
        return this;
    }
    // It does, so proceed
    List<String> keyValues = new ArrayList<>();
    String queryString = StringUtils.substringAfter(url, "?");

    // extract and remove any fragments
    String fragment = StringUtils.substringAfter(queryString, "#");
    if (StringUtils.isNotBlank(fragment)) {
        fragment = "#" + fragment;
    }
    queryString = StringUtils.substringBefore(queryString, "#");

    String[] params = StringUtils.split(queryString, '&');
    for (String param : params) {
        keyValues.add(param);
    }
    // sort it so that query string are in order
    Collections.sort(keyValues);

    String sortedQueryString = StringUtils.join(keyValues, '&');
    if (StringUtils.isNotBlank(sortedQueryString)) {
        url = StringUtils.substringBefore(
                url, "?") + "?" + sortedQueryString + fragment;
    }

    return this;
}
 

/**
 * <p>
 * Encodes space characters into plus signs (+) if they are part of the
 * query string. Spaces part of the URL path are percent-encoded to %20.
 * </p>
 * <p>
 * To encode all non-ASCII characters (including spaces), use
 * {@link #encodeNonURICharacters()} instead.
 * </p>
 * <code>http://www.example.com/a b c &rarr;
 *       http://www.example.com/a+b+c</code>
 * @return this instance
 * @since 1.8.0
 */
public URLNormalizer encodeSpaces() {
    String path = StringUtils.substringBefore(url, "?");
    path = StringUtils.replace(path, " ", "%20", -1);
    String qs = StringUtils.substringAfter(url, "?");
    if (StringUtils.isNotBlank(qs)) {
        qs = StringUtils.replace(qs, " ", "+", -1);
        url = path + "?" + qs;
    } else {
        url = path;
    }
    return this;
}