001/*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements.  See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License.  You may obtain a copy of the License at
008 *
009 *      http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017package org.apache.commons.validator;
018
019import java.io.Serializable;
020import java.util.Arrays;
021import java.util.HashSet;
022import java.util.Set;
023import java.util.regex.Matcher;
024import java.util.regex.Pattern;
025
026import org.apache.commons.validator.routines.InetAddressValidator;
027import org.apache.commons.validator.util.Flags;
028
029/**
030 * <p>Validates URLs.</p>
031 * Behavour of validation is modified by passing in options:
032 * <ul>
033 * <li>ALLOW_2_SLASHES - [FALSE]  Allows double '/' characters in the path
034 * component.</li>
035 * <li>NO_FRAGMENT- [FALSE]  By default fragments are allowed, if this option is
036 * included then fragments are flagged as illegal.</li>
037 * <li>ALLOW_ALL_SCHEMES - [FALSE] By default only http, https, and ftp are
038 * considered valid schemes.  Enabling this option will let any scheme pass validation.</li>
039 * </ul>
040 *
041 * <p>Originally based in on php script by Debbie Dyer, validation.php v1.2b, Date: 03/07/02,
042 * http://javascript.internet.com. However, this validation now bears little resemblance
043 * to the php original.</p>
044 * <pre>
045 *   Example of usage:
046 *   Construct a UrlValidator with valid schemes of "http", and "https".
047 *
048 *    String[] schemes = {"http","https"}.
049 *    UrlValidator urlValidator = new UrlValidator(schemes);
050 *    if (urlValidator.isValid("ftp://foo.bar.com/")) {
051 *       System.out.println("url is valid");
052 *    } else {
053 *       System.out.println("url is invalid");
054 *    }
055 *
056 *    prints "url is invalid"
057 *   If instead the default constructor is used.
058 *
059 *    UrlValidator urlValidator = new UrlValidator();
060 *    if (urlValidator.isValid("ftp://foo.bar.com/")) {
061 *       System.out.println("url is valid");
062 *    } else {
063 *       System.out.println("url is invalid");
064 *    }
065 *
066 *   prints out "url is valid"
067 *  </pre>
068 *
069 * @see
070 * <a href="http://www.ietf.org/rfc/rfc2396.txt">
071 *  Uniform Resource Identifiers (URI): Generic Syntax
072 * </a>
073 *
074 * @version $Revision: 1739358 $
075 * @since Validator 1.1
076 * @deprecated Use the new UrlValidator in the routines package. This class
077 * will be removed in a future release.
078 */
079@Deprecated
080public class UrlValidator implements Serializable {
081
082    private static final long serialVersionUID = 24137157400029593L;
083
084    /**
085     * Allows all validly formatted schemes to pass validation instead of
086     * supplying a set of valid schemes.
087     */
088    public static final int ALLOW_ALL_SCHEMES = 1 << 0;
089
090    /**
091     * Allow two slashes in the path component of the URL.
092     */
093    public static final int ALLOW_2_SLASHES = 1 << 1;
094
095    /**
096     * Enabling this options disallows any URL fragments.
097     */
098    public static final int NO_FRAGMENTS = 1 << 2;
099
100    private static final String ALPHA_CHARS = "a-zA-Z";
101
102// NOT USED   private static final String ALPHA_NUMERIC_CHARS = ALPHA_CHARS + "\\d";
103
104    private static final String SPECIAL_CHARS = ";/@&=,.?:+$";
105
106    private static final String VALID_CHARS = "[^\\s" + SPECIAL_CHARS + "]";
107
108    // Drop numeric, and  "+-." for now
109    private static final String AUTHORITY_CHARS_REGEX = "\\p{Alnum}\\-\\.";
110
111    private static final String ATOM = VALID_CHARS + '+';
112
113    /**
114     * This expression derived/taken from the BNF for URI (RFC2396).
115     */
116    private static final String URL_REGEX =
117            "^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\\?([^#]*))?(#(.*))?";
118    //                                                                      12            3  4          5       6   7        8 9
119    private static final Pattern URL_PATTERN = Pattern.compile(URL_REGEX);
120
121    /**
122     * Schema/Protocol (ie. http:, ftp:, file:, etc).
123     */
124    private static final int PARSE_URL_SCHEME = 2;
125
126    /**
127     * Includes hostname/ip and port number.
128     */
129    private static final int PARSE_URL_AUTHORITY = 4;
130
131    private static final int PARSE_URL_PATH = 5;
132
133    private static final int PARSE_URL_QUERY = 7;
134
135    private static final int PARSE_URL_FRAGMENT = 9;
136
137    /**
138     * Protocol (ie. http:, ftp:,https:).
139     */
140    private static final Pattern SCHEME_PATTERN = Pattern.compile("^\\p{Alpha}[\\p{Alnum}\\+\\-\\.]*");
141
142    private static final String AUTHORITY_REGEX =
143       "^([" + AUTHORITY_CHARS_REGEX + "]*)(:\\d*)?(.*)?";
144    //                                                                            1                          2  3       4
145    private static final Pattern AUTHORITY_PATTERN = Pattern.compile(AUTHORITY_REGEX);
146
147    private static final int PARSE_AUTHORITY_HOST_IP = 1;
148
149    private static final int PARSE_AUTHORITY_PORT = 2;
150
151    /**
152     * Should always be empty.
153     */
154    private static final int PARSE_AUTHORITY_EXTRA = 3;
155
156    private static final Pattern PATH_PATTERN = Pattern.compile("^(/[-\\w:@&?=+,.!/~*'%$_;]*)?$");
157
158    private static final Pattern QUERY_PATTERN = Pattern.compile("^(.*)$");
159
160    private static final Pattern LEGAL_ASCII_PATTERN = Pattern.compile("^\\p{ASCII}+$");
161
162    private static final Pattern DOMAIN_PATTERN =
163            Pattern.compile("^" + ATOM + "(\\." + ATOM + ")*$");
164
165    private static final Pattern PORT_PATTERN = Pattern.compile("^:(\\d{1,5})$");
166
167    private static final Pattern ATOM_PATTERN = Pattern.compile("^(" + ATOM + ").*?$");
168
169    private static final Pattern ALPHA_PATTERN = Pattern.compile("^[" + ALPHA_CHARS + "]");
170
171    /**
172     * Holds the set of current validation options.
173     */
174    private final Flags options;
175
176    /**
177     * The set of schemes that are allowed to be in a URL.
178     */
179    private final Set<String> allowedSchemes = new HashSet<String>();
180
181    /**
182     * If no schemes are provided, default to this set.
183     */
184    protected String[] defaultSchemes = {"http", "https", "ftp"};
185
186    /**
187     * Create a UrlValidator with default properties.
188     */
189    public UrlValidator() {
190        this(null);
191    }
192
193    /**
194     * Behavior of validation is modified by passing in several strings options:
195     * @param schemes Pass in one or more url schemes to consider valid, passing in
196     *        a null will default to "http,https,ftp" being valid.
197     *        If a non-null schemes is specified then all valid schemes must
198     *        be specified. Setting the ALLOW_ALL_SCHEMES option will
199     *        ignore the contents of schemes.
200     */
201    public UrlValidator(String[] schemes) {
202        this(schemes, 0);
203    }
204
205    /**
206     * Initialize a UrlValidator with the given validation options.
207     * @param options The options should be set using the public constants declared in
208     * this class.  To set multiple options you simply add them together.  For example,
209     * ALLOW_2_SLASHES + NO_FRAGMENTS enables both of those options.
210     */
211    public UrlValidator(int options) {
212        this(null, options);
213    }
214
215    /**
216     * Behavour of validation is modified by passing in options:
217     * @param schemes The set of valid schemes.
218     * @param options The options should be set using the public constants declared in
219     * this class.  To set multiple options you simply add them together.  For example,
220     * ALLOW_2_SLASHES + NO_FRAGMENTS enables both of those options.
221     */
222    public UrlValidator(String[] schemes, int options) {
223        this.options = new Flags(options);
224
225        if (this.options.isOn(ALLOW_ALL_SCHEMES)) {
226            return;
227        }
228
229        if (schemes == null) {
230            schemes = this.defaultSchemes;
231        }
232
233        this.allowedSchemes.addAll(Arrays.asList(schemes));
234    }
235
236    /**
237     * <p>Checks if a field has a valid url address.</p>
238     *
239     * @param value The value validation is being performed on.  A <code>null</code>
240     * value is considered invalid.
241     * @return true if the url is valid.
242     */
243    public boolean isValid(String value) {
244        if (value == null) {
245            return false;
246        }
247        if (!LEGAL_ASCII_PATTERN.matcher(value).matches()) {
248           return false;
249        }
250
251        // Check the whole url address structure
252        Matcher urlMatcher = URL_PATTERN.matcher(value);
253        if (!urlMatcher.matches()) {
254            return false;
255        }
256
257        if (!isValidScheme(urlMatcher.group(PARSE_URL_SCHEME))) {
258            return false;
259        }
260
261        if (!isValidAuthority(urlMatcher.group(PARSE_URL_AUTHORITY))) {
262            return false;
263        }
264
265        if (!isValidPath(urlMatcher.group(PARSE_URL_PATH))) {
266            return false;
267        }
268
269        if (!isValidQuery(urlMatcher.group(PARSE_URL_QUERY))) {
270            return false;
271        }
272
273        if (!isValidFragment(urlMatcher.group(PARSE_URL_FRAGMENT))) {
274            return false;
275        }
276
277        return true;
278    }
279
280    /**
281     * Validate scheme. If schemes[] was initialized to a non null,
282     * then only those scheme's are allowed.  Note this is slightly different
283     * than for the constructor.
284     * @param scheme The scheme to validate.  A <code>null</code> value is considered
285     * invalid.
286     * @return true if valid.
287     */
288    protected boolean isValidScheme(String scheme) {
289        if (scheme == null) {
290            return false;
291        }
292
293        if (!SCHEME_PATTERN.matcher(scheme).matches()) {
294            return false;
295        }
296
297        if (options.isOff(ALLOW_ALL_SCHEMES) && !allowedSchemes.contains(scheme)) {
298            return false;
299        }
300
301        return true;
302    }
303
304    /**
305     * Returns true if the authority is properly formatted.  An authority is the combination
306     * of hostname and port.  A <code>null</code> authority value is considered invalid.
307     * @param authority Authority value to validate.
308     * @return true if authority (hostname and port) is valid.
309     */
310    protected boolean isValidAuthority(String authority) {
311        if (authority == null) {
312            return false;
313        }
314
315        InetAddressValidator inetAddressValidator =
316                InetAddressValidator.getInstance();
317
318        Matcher authorityMatcher = AUTHORITY_PATTERN.matcher(authority);
319        if (!authorityMatcher.matches()) {
320            return false;
321        }
322
323        boolean hostname = false;
324        // check if authority is IP address or hostname
325        String hostIP = authorityMatcher.group(PARSE_AUTHORITY_HOST_IP);
326        boolean ipV4Address = inetAddressValidator.isValid(hostIP);
327
328        if (!ipV4Address) {
329            // Domain is hostname name
330            hostname = DOMAIN_PATTERN.matcher(hostIP).matches();
331        }
332
333        //rightmost hostname will never start with a digit.
334        if (hostname) {
335            // LOW-TECH FIX FOR VALIDATOR-202
336            // TODO: Rewrite to use ArrayList and .add semantics: see VALIDATOR-203
337            char[] chars = hostIP.toCharArray();
338            int size = 1;
339            for(int i=0; i<chars.length; i++) {
340                if(chars[i] == '.') {
341                    size++;
342                }
343            }
344            String[] domainSegment = new String[size];
345            boolean match = true;
346            int segmentCount = 0;
347            int segmentLength = 0;
348
349            while (match) {
350                Matcher atomMatcher = ATOM_PATTERN.matcher(hostIP);
351                match = atomMatcher.matches();
352                if (match) {
353                    domainSegment[segmentCount] = atomMatcher.group(1);
354                    segmentLength = domainSegment[segmentCount].length() + 1;
355                    hostIP =
356                            (segmentLength >= hostIP.length())
357                            ? ""
358                            : hostIP.substring(segmentLength);
359
360                    segmentCount++;
361                }
362            }
363            String topLevel = domainSegment[segmentCount - 1];
364            if (topLevel.length() < 2 || topLevel.length() > 4) { // CHECKSTYLE IGNORE MagicNumber (deprecated code)
365                return false;
366            }
367
368            // First letter of top level must be a alpha
369            if (!ALPHA_PATTERN.matcher(topLevel.substring(0, 1)).matches()) {
370                return false;
371            }
372
373            // Make sure there's a host name preceding the authority.
374            if (segmentCount < 2) {
375                return false;
376            }
377        }
378
379        if (!hostname && !ipV4Address) {
380            return false;
381        }
382
383        String port = authorityMatcher.group(PARSE_AUTHORITY_PORT);
384        if (port != null && !PORT_PATTERN.matcher(port).matches()) {
385            return false;
386        }
387
388        String extra = authorityMatcher.group(PARSE_AUTHORITY_EXTRA);
389        if (!GenericValidator.isBlankOrNull(extra)) {
390            return false;
391        }
392
393        return true;
394    }
395
396    /**
397     * Returns true if the path is valid.  A <code>null</code> value is considered invalid.
398     * @param path Path value to validate.
399     * @return true if path is valid.
400     */
401    protected boolean isValidPath(String path) {
402        if (path == null) {
403            return false;
404        }
405
406        if (!PATH_PATTERN.matcher(path).matches()) {
407            return false;
408        }
409
410        int slash2Count = countToken("//", path);
411        if (options.isOff(ALLOW_2_SLASHES) && (slash2Count > 0)) {
412            return false;
413        }
414
415        int slashCount = countToken("/", path);
416        int dot2Count = countToken("..", path);
417        if (dot2Count > 0 && (slashCount - slash2Count - 1) <= dot2Count){
418            return false;
419        }
420
421        return true;
422    }
423
424    /**
425     * Returns true if the query is null or it's a properly formatted query string.
426     * @param query Query value to validate.
427     * @return true if query is valid.
428     */
429    protected boolean isValidQuery(String query) {
430        if (query == null) {
431            return true;
432        }
433
434        return QUERY_PATTERN.matcher(query).matches();
435    }
436
437    /**
438     * Returns true if the given fragment is null or fragments are allowed.
439     * @param fragment Fragment value to validate.
440     * @return true if fragment is valid.
441     */
442    protected boolean isValidFragment(String fragment) {
443        if (fragment == null) {
444            return true;
445        }
446
447        return options.isOff(NO_FRAGMENTS);
448    }
449
450    /**
451     * Returns the number of times the token appears in the target.
452     * @param token Token value to be counted.
453     * @param target Target value to count tokens in.
454     * @return the number of tokens.
455     */
456    protected int countToken(String token, String target) {
457        int tokenIndex = 0;
458        int count = 0;
459        while (tokenIndex != -1) {
460            tokenIndex = target.indexOf(token, tokenIndex);
461            if (tokenIndex > -1) {
462                tokenIndex++;
463                count++;
464            }
465        }
466        return count;
467    }
468}