001/*
002 * Copyright (C) 2011 The Guava Authors
003 *
004 * Licensed under the Apache License, Version 2.0 (the "License");
005 * you may not use this file except in compliance with the License.
006 * You may obtain a copy of the License at
007 *
008 * http://www.apache.org/licenses/LICENSE-2.0
009 *
010 * Unless required by applicable law or agreed to in writing, software
011 * distributed under the License is distributed on an "AS IS" BASIS,
012 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
013 * See the License for the specific language governing permissions and
014 * limitations under the License.
015 */
016
017package com.google.common.net;
018
019import static com.google.common.base.CharMatcher.ASCII;
020import static com.google.common.base.CharMatcher.JAVA_ISO_CONTROL;
021import static com.google.common.base.Charsets.UTF_8;
022import static com.google.common.base.Preconditions.checkArgument;
023import static com.google.common.base.Preconditions.checkNotNull;
024import static com.google.common.base.Preconditions.checkState;
025
026import com.google.common.annotations.Beta;
027import com.google.common.annotations.GwtCompatible;
028import com.google.common.base.Ascii;
029import com.google.common.base.CharMatcher;
030import com.google.common.base.Function;
031import com.google.common.base.Joiner;
032import com.google.common.base.Joiner.MapJoiner;
033import com.google.common.base.MoreObjects;
034import com.google.common.base.Objects;
035import com.google.common.base.Optional;
036import com.google.common.collect.ImmutableListMultimap;
037import com.google.common.collect.ImmutableMultiset;
038import com.google.common.collect.ImmutableSet;
039import com.google.common.collect.Iterables;
040import com.google.common.collect.Maps;
041import com.google.common.collect.Multimap;
042import com.google.common.collect.Multimaps;
043
044import java.nio.charset.Charset;
045import java.nio.charset.IllegalCharsetNameException;
046import java.nio.charset.UnsupportedCharsetException;
047import java.util.Collection;
048import java.util.Map;
049import java.util.Map.Entry;
050
051import javax.annotation.Nullable;
052import javax.annotation.concurrent.Immutable;
053
054/**
055 * Represents an <a href="http://en.wikipedia.org/wiki/Internet_media_type">Internet Media Type</a>
056 * (also known as a MIME Type or Content Type). This class also supports the concept of media ranges
057 * <a href="http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.1">defined by HTTP/1.1</a>.
058 * As such, the {@code *} character is treated as a wildcard and is used to represent any acceptable
059 * type or subtype value. A media type may not have wildcard type with a declared subtype. The
060 * {@code *} character has no special meaning as part of a parameter. All values for type, subtype,
061 * parameter attributes or parameter values must be valid according to RFCs
062 * <a href="http://www.ietf.org/rfc/rfc2045.txt">2045</a> and
063 * <a href="http://www.ietf.org/rfc/rfc2046.txt">2046</a>.
064 *
065 * <p>All portions of the media type that are case-insensitive (type, subtype, parameter attributes)
066 * are normalized to lowercase. The value of the {@code charset} parameter is normalized to
067 * lowercase, but all others are left as-is.
068 *
069 * <p>Note that this specifically does <strong>not</strong> represent the value of the MIME
070 * {@code Content-Type} header and as such has no support for header-specific considerations such as
071 * line folding and comments.
072 *
073 * <p>For media types that take a charset the predefined constants default to UTF-8 and have a
074 * "_UTF_8" suffix. To get a version without a character set, use {@link #withoutParameters}.
075 *
076 * @since 12.0
077 *
078 * @author Gregory Kick
079 */
080@Beta
081@GwtCompatible
082@Immutable
083public final class MediaType {
084  private static final String CHARSET_ATTRIBUTE = "charset";
085  private static final ImmutableListMultimap<String, String> UTF_8_CONSTANT_PARAMETERS =
086      ImmutableListMultimap.of(CHARSET_ATTRIBUTE, Ascii.toLowerCase(UTF_8.name()));
087
088  /** Matcher for type, subtype and attributes. */
089  private static final CharMatcher TOKEN_MATCHER = ASCII.and(JAVA_ISO_CONTROL.negate())
090      .and(CharMatcher.isNot(' '))
091      .and(CharMatcher.noneOf("()<>@,;:\\\"/[]?="));
092  private static final CharMatcher QUOTED_TEXT_MATCHER = ASCII
093      .and(CharMatcher.noneOf("\"\\\r"));
094  /*
095   * This matches the same characters as linear-white-space from RFC 822, but we make no effort to
096   * enforce any particular rules with regards to line folding as stated in the class docs.
097   */
098  private static final CharMatcher LINEAR_WHITE_SPACE = CharMatcher.anyOf(" \t\r\n");
099
100  // TODO(gak): make these public?
101  private static final String APPLICATION_TYPE = "application";
102  private static final String AUDIO_TYPE = "audio";
103  private static final String IMAGE_TYPE = "image";
104  private static final String TEXT_TYPE = "text";
105  private static final String VIDEO_TYPE = "video";
106
107  private static final String WILDCARD = "*";
108
109  private static final Map<MediaType, MediaType> KNOWN_TYPES = Maps.newHashMap();
110
111  private static MediaType createConstant(String type, String subtype) {
112    return addKnownType(new MediaType(type, subtype, ImmutableListMultimap.<String, String>of()));
113  }
114
115  private static MediaType createConstantUtf8(String type, String subtype) {
116    return addKnownType(new MediaType(type, subtype, UTF_8_CONSTANT_PARAMETERS));
117  }
118
119  private static MediaType addKnownType(MediaType mediaType) {
120    KNOWN_TYPES.put(mediaType, mediaType);
121    return mediaType;
122  }
123
124  /*
125   * The following constants are grouped by their type and ordered alphabetically by the constant
126   * name within that type. The constant name should be a sensible identifier that is closest to the
127   * "common name" of the media.  This is often, but not necessarily the same as the subtype.
128   *
129   * Be sure to declare all constants with the type and subtype in all lowercase. For types that
130   * take a charset (e.g. all text/* types), default to UTF-8 and suffix the constant name with
131   * "_UTF_8".
132   */
133
134  public static final MediaType ANY_TYPE = createConstant(WILDCARD, WILDCARD);
135  public static final MediaType ANY_TEXT_TYPE = createConstant(TEXT_TYPE, WILDCARD);
136  public static final MediaType ANY_IMAGE_TYPE = createConstant(IMAGE_TYPE, WILDCARD);
137  public static final MediaType ANY_AUDIO_TYPE = createConstant(AUDIO_TYPE, WILDCARD);
138  public static final MediaType ANY_VIDEO_TYPE = createConstant(VIDEO_TYPE, WILDCARD);
139  public static final MediaType ANY_APPLICATION_TYPE = createConstant(APPLICATION_TYPE, WILDCARD);
140
141  /* text types */
142  public static final MediaType CACHE_MANIFEST_UTF_8 =
143      createConstantUtf8(TEXT_TYPE, "cache-manifest");
144  public static final MediaType CSS_UTF_8 = createConstantUtf8(TEXT_TYPE, "css");
145  public static final MediaType CSV_UTF_8 = createConstantUtf8(TEXT_TYPE, "csv");
146  public static final MediaType HTML_UTF_8 = createConstantUtf8(TEXT_TYPE, "html");
147  public static final MediaType I_CALENDAR_UTF_8 = createConstantUtf8(TEXT_TYPE, "calendar");
148  public static final MediaType PLAIN_TEXT_UTF_8 = createConstantUtf8(TEXT_TYPE, "plain");
149  /**
150   * <a href="http://www.rfc-editor.org/rfc/rfc4329.txt">RFC 4329</a> declares
151   * {@link #JAVASCRIPT_UTF_8 application/javascript} to be the correct media type for JavaScript,
152   * but this may be necessary in certain situations for compatibility.
153   */
154  public static final MediaType TEXT_JAVASCRIPT_UTF_8 = createConstantUtf8(TEXT_TYPE, "javascript");
155  /**
156   * <a href="http://www.iana.org/assignments/media-types/text/tab-separated-values">
157   * Tab separated values</a>.
158   *
159   * @since 15.0
160   */
161  public static final MediaType TSV_UTF_8 = createConstantUtf8(TEXT_TYPE, "tab-separated-values");
162  public static final MediaType VCARD_UTF_8 = createConstantUtf8(TEXT_TYPE, "vcard");
163  public static final MediaType WML_UTF_8 = createConstantUtf8(TEXT_TYPE, "vnd.wap.wml");
164  /**
165   * As described in <a href="http://www.ietf.org/rfc/rfc3023.txt">RFC 3023</a>, this constant
166   * ({@code text/xml}) is used for XML documents that are "readable by casual users."
167   * {@link #APPLICATION_XML_UTF_8} is provided for documents that are intended for applications.
168   */
169  public static final MediaType XML_UTF_8 = createConstantUtf8(TEXT_TYPE, "xml");
170
171  /* image types */
172  public static final MediaType BMP = createConstant(IMAGE_TYPE, "bmp");
173  /**
174   * The media type for the <a href="http://en.wikipedia.org/wiki/Camera_Image_File_Format">Canon
175   * Image File Format</a> ({@code crw} files), a widely-used "raw image" format for cameras. It is
176   * found in {@code /etc/mime.types}, e.g. in <href=
177   * "http://anonscm.debian.org/gitweb/?p=collab-maint/mime-support.git;a=blob;f=mime.types;hb=HEAD"
178   * >Debian 3.48-1</a>.
179   *
180   * @since 15.0
181   */
182  public static final MediaType CRW = createConstant(IMAGE_TYPE, "x-canon-crw");
183  public static final MediaType GIF = createConstant(IMAGE_TYPE, "gif");
184  public static final MediaType ICO = createConstant(IMAGE_TYPE, "vnd.microsoft.icon");
185  public static final MediaType JPEG = createConstant(IMAGE_TYPE, "jpeg");
186  public static final MediaType PNG = createConstant(IMAGE_TYPE, "png");
187  /**
188   * The media type for the Photoshop File Format ({@code psd} files) as defined by <a href=
189   * "http://www.iana.org/assignments/media-types/image/vnd.adobe.photoshop">IANA</a>, and found in
190   * {@code /etc/mime.types}, e.g. <a href=
191   * "http://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x/conf/mime.types"></a> of the Apache
192   * <a href="http://httpd.apache.org/">HTTPD project</a>; for the specification, see
193   * <href="http://www.adobe.com/devnet-apps/photoshop/fileformatashtml/PhotoshopFileFormats.htm">
194   * Adobe Photoshop Document Format</a> and <a href=
195   * "http://en.wikipedia.org/wiki/Adobe_Photoshop#File_format">Wikipedia</a>; this is the regular
196   * output/input of Photoshop (which can also export to various image formats; note that files with
197   * extension "PSB" are in a distinct but related format).
198   * <p>This is a more recent replacement for the older, experimental type
199   * {@code x-photoshop}: <a href="http://tools.ietf.org/html/rfc2046#section-6">RFC-2046.6</a>.
200   *
201   * @since 15.0
202   */
203  public static final MediaType PSD = createConstant(IMAGE_TYPE, "vnd.adobe.photoshop");
204  public static final MediaType SVG_UTF_8 = createConstantUtf8(IMAGE_TYPE, "svg+xml");
205  public static final MediaType TIFF = createConstant(IMAGE_TYPE, "tiff");
206  public static final MediaType WEBP = createConstant(IMAGE_TYPE, "webp");
207
208  /* audio types */
209  public static final MediaType MP4_AUDIO = createConstant(AUDIO_TYPE, "mp4");
210  public static final MediaType MPEG_AUDIO = createConstant(AUDIO_TYPE, "mpeg");
211  public static final MediaType OGG_AUDIO = createConstant(AUDIO_TYPE, "ogg");
212  public static final MediaType WEBM_AUDIO = createConstant(AUDIO_TYPE, "webm");
213
214  /* video types */
215  public static final MediaType MP4_VIDEO = createConstant(VIDEO_TYPE, "mp4");
216  public static final MediaType MPEG_VIDEO = createConstant(VIDEO_TYPE, "mpeg");
217  public static final MediaType OGG_VIDEO = createConstant(VIDEO_TYPE, "ogg");
218  public static final MediaType QUICKTIME = createConstant(VIDEO_TYPE, "quicktime");
219  public static final MediaType WEBM_VIDEO = createConstant(VIDEO_TYPE, "webm");
220  public static final MediaType WMV = createConstant(VIDEO_TYPE, "x-ms-wmv");
221
222  /* application types */
223  /**
224   * As described in <a href="http://www.ietf.org/rfc/rfc3023.txt">RFC 3023</a>, this constant
225   * ({@code application/xml}) is used for XML documents that are "unreadable by casual users."
226   * {@link #XML_UTF_8} is provided for documents that may be read by users.
227   */
228  public static final MediaType APPLICATION_XML_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "xml");
229  public static final MediaType ATOM_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "atom+xml");
230  public static final MediaType BZIP2 = createConstant(APPLICATION_TYPE, "x-bzip2");
231
232  /**
233   * Media type for <a href="http://en.wikipedia.org/wiki/Embedded_OpenType">Embedded OpenType</a>
234   * fonts. This is
235   * <a href="http://www.iana.org/assignments/media-types/application/vnd.ms-fontobject">registered
236   * </a> with the IANA.
237   *
238   * @since 17.0
239   */
240  public static final MediaType EOT = createConstant(APPLICATION_TYPE, "vnd.ms-fontobject");
241  /**
242   * As described in the <a href="http://idpf.org/epub">International Digital Publishing Forum</a>
243   * EPUB is the distribution and interchange format standard for digital publications and
244   * documents. This media type is defined in the
245   * <a href="http://www.idpf.org/epub/30/spec/epub30-ocf.html">EPUB Open Container Format</a>
246   * specification.
247   *
248   * @since 15.0
249   */
250  public static final MediaType EPUB = createConstant(APPLICATION_TYPE, "epub+zip");
251  public static final MediaType FORM_DATA = createConstant(APPLICATION_TYPE,
252      "x-www-form-urlencoded");
253  /**
254   * As described in <a href="https://www.rsa.com/rsalabs/node.asp?id=2138">PKCS #12: Personal
255   * Information Exchange Syntax Standard</a>, PKCS #12 defines an archive file format for storing
256   * many cryptography objects as a single file.
257   *
258   * @since 15.0
259   */
260  public static final MediaType KEY_ARCHIVE = createConstant(APPLICATION_TYPE, "pkcs12");
261  /**
262   * This is a non-standard media type, but is commonly used in serving hosted binary files as it is
263   * <a href="http://code.google.com/p/browsersec/wiki/Part2#Survey_of_content_sniffing_behaviors">
264   * known not to trigger content sniffing in current browsers</a>. It <i>should not</i> be used in
265   * other situations as it is not specified by any RFC and does not appear in the <a href=
266   * "http://www.iana.org/assignments/media-types">/IANA MIME Media Types</a> list. Consider
267   * {@link #OCTET_STREAM} for binary data that is not being served to a browser.
268   *
269   *
270   * @since 14.0
271   */
272  public static final MediaType APPLICATION_BINARY = createConstant(APPLICATION_TYPE, "binary");
273  public static final MediaType GZIP = createConstant(APPLICATION_TYPE, "x-gzip");
274   /**
275    * <a href="http://www.rfc-editor.org/rfc/rfc4329.txt">RFC 4329</a> declares this to be the
276    * correct media type for JavaScript, but {@link #TEXT_JAVASCRIPT_UTF_8 text/javascript} may be
277    * necessary in certain situations for compatibility.
278    */
279  public static final MediaType JAVASCRIPT_UTF_8 =
280      createConstantUtf8(APPLICATION_TYPE, "javascript");
281  public static final MediaType JSON_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "json");
282  public static final MediaType KML = createConstant(APPLICATION_TYPE, "vnd.google-earth.kml+xml");
283  public static final MediaType KMZ = createConstant(APPLICATION_TYPE, "vnd.google-earth.kmz");
284  public static final MediaType MBOX = createConstant(APPLICATION_TYPE, "mbox");
285
286  /**
287   * Media type for
288   * <a href="http://goo.gl/1pGBFm">Apple over-the-air mobile configuration profiles</a>.
289   *
290   * @since 18.0
291   */
292  public static final MediaType APPLE_MOBILE_CONFIG =
293      createConstant(APPLICATION_TYPE, "x-apple-aspen-config");
294  public static final MediaType MICROSOFT_EXCEL = createConstant(APPLICATION_TYPE, "vnd.ms-excel");
295  public static final MediaType MICROSOFT_POWERPOINT =
296      createConstant(APPLICATION_TYPE, "vnd.ms-powerpoint");
297  public static final MediaType MICROSOFT_WORD = createConstant(APPLICATION_TYPE, "msword");
298  public static final MediaType OCTET_STREAM = createConstant(APPLICATION_TYPE, "octet-stream");
299  public static final MediaType OGG_CONTAINER = createConstant(APPLICATION_TYPE, "ogg");
300  public static final MediaType OOXML_DOCUMENT = createConstant(APPLICATION_TYPE,
301      "vnd.openxmlformats-officedocument.wordprocessingml.document");
302  public static final MediaType OOXML_PRESENTATION = createConstant(APPLICATION_TYPE,
303      "vnd.openxmlformats-officedocument.presentationml.presentation");
304  public static final MediaType OOXML_SHEET =
305      createConstant(APPLICATION_TYPE, "vnd.openxmlformats-officedocument.spreadsheetml.sheet");
306  public static final MediaType OPENDOCUMENT_GRAPHICS =
307      createConstant(APPLICATION_TYPE, "vnd.oasis.opendocument.graphics");
308  public static final MediaType OPENDOCUMENT_PRESENTATION =
309      createConstant(APPLICATION_TYPE, "vnd.oasis.opendocument.presentation");
310  public static final MediaType OPENDOCUMENT_SPREADSHEET =
311      createConstant(APPLICATION_TYPE, "vnd.oasis.opendocument.spreadsheet");
312  public static final MediaType OPENDOCUMENT_TEXT =
313      createConstant(APPLICATION_TYPE, "vnd.oasis.opendocument.text");
314  public static final MediaType PDF = createConstant(APPLICATION_TYPE, "pdf");
315  public static final MediaType POSTSCRIPT = createConstant(APPLICATION_TYPE, "postscript");
316  /**
317   * <a href="http://tools.ietf.org/html/draft-rfernando-protocol-buffers-00">Protocol buffers</a>
318   *
319   * @since 15.0
320   */
321  public static final MediaType PROTOBUF = createConstant(APPLICATION_TYPE, "protobuf");
322  public static final MediaType RDF_XML_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "rdf+xml");
323  public static final MediaType RTF_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "rtf");
324  /**
325   * Media type for SFNT fonts (which includes
326   * <a href="http://en.wikipedia.org/wiki/TrueType/">TrueType</a> and
327   * <a href="http://en.wikipedia.org/wiki/OpenType/">OpenType</a> fonts). This is
328   * <a href="http://www.iana.org/assignments/media-types/application/font-sfnt">registered</a>
329   * with the IANA.
330   *
331   * @since 17.0
332   */
333  public static final MediaType SFNT = createConstant(APPLICATION_TYPE, "font-sfnt");
334  public static final MediaType SHOCKWAVE_FLASH = createConstant(APPLICATION_TYPE,
335      "x-shockwave-flash");
336  public static final MediaType SKETCHUP = createConstant(APPLICATION_TYPE, "vnd.sketchup.skp");
337  public static final MediaType TAR = createConstant(APPLICATION_TYPE, "x-tar");
338  /**
339   * Media type for the
340   * <a href="http://en.wikipedia.org/wiki/Web_Open_Font_Format">Web Open Font Format</a> (WOFF)
341   * <a href="http://www.w3.org/TR/WOFF/">defined</a> by the W3C. This is
342   * <a href="http://www.iana.org/assignments/media-types/application/font-woff">registered</a>
343   * with the IANA.
344   *
345   * @since 17.0
346   */
347  public static final MediaType WOFF = createConstant(APPLICATION_TYPE, "font-woff");
348  public static final MediaType XHTML_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "xhtml+xml");
349  /**
350   * Media type for Extensible Resource Descriptors. This is not yet registered with the IANA, but
351   * it is specified by OASIS in the
352   * <a href="http://docs.oasis-open.org/xri/xrd/v1.0/cd02/xrd-1.0-cd02.html"> XRD definition</a>
353   * and implemented in projects such as
354   * <a href="http://code.google.com/p/webfinger/">WebFinger</a>.
355   */
356  public static final MediaType XRD_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "xrd+xml");
357  public static final MediaType ZIP = createConstant(APPLICATION_TYPE, "zip");
358
359  private final String type;
360  private final String subtype;
361  private final ImmutableListMultimap<String, String> parameters;
362
363  private MediaType(String type, String subtype,
364      ImmutableListMultimap<String, String> parameters) {
365    this.type = type;
366    this.subtype = subtype;
367    this.parameters = parameters;
368  }
369
370  /** Returns the top-level media type.  For example, {@code "text"} in {@code "text/plain"}. */
371  public String type() {
372    return type;
373  }
374
375  /** Returns the media subtype.  For example, {@code "plain"} in {@code "text/plain"}. */
376  public String subtype() {
377    return subtype;
378  }
379
380  /** Returns a multimap containing the parameters of this media type. */
381  public ImmutableListMultimap<String, String> parameters() {
382    return parameters;
383  }
384
385  private Map<String, ImmutableMultiset<String>> parametersAsMap() {
386    return Maps.transformValues(parameters.asMap(),
387        new Function<Collection<String>, ImmutableMultiset<String>>() {
388          @Override public ImmutableMultiset<String> apply(Collection<String> input) {
389            return ImmutableMultiset.copyOf(input);
390          }
391        });
392  }
393
394  /**
395   * Returns an optional charset for the value of the charset parameter if it is specified.
396   *
397   * @throws IllegalStateException if multiple charset values have been set for this media type
398   * @throws IllegalCharsetNameException if a charset value is present, but illegal
399   * @throws UnsupportedCharsetException if a charset value is present, but no support is available
400   *     in this instance of the Java virtual machine
401   */
402  public Optional<Charset> charset() {
403    ImmutableSet<String> charsetValues = ImmutableSet.copyOf(parameters.get(CHARSET_ATTRIBUTE));
404    switch (charsetValues.size()) {
405      case 0:
406        return Optional.absent();
407      case 1:
408        return Optional.of(Charset.forName(Iterables.getOnlyElement(charsetValues)));
409      default:
410        throw new IllegalStateException("Multiple charset values defined: " + charsetValues);
411    }
412  }
413
414  /**
415   * Returns a new instance with the same type and subtype as this instance, but without any
416   * parameters.
417   */
418  public MediaType withoutParameters() {
419    return parameters.isEmpty() ? this : create(type, subtype);
420  }
421
422  /**
423   * <em>Replaces</em> all parameters with the given parameters.
424   *
425   * @throws IllegalArgumentException if any parameter or value is invalid
426   */
427  public MediaType withParameters(Multimap<String, String> parameters) {
428    return create(type, subtype, parameters);
429  }
430
431  /**
432   * <em>Replaces</em> all parameters with the given attribute with a single parameter with the
433   * given value. If multiple parameters with the same attributes are necessary use
434   * {@link #withParameters}. Prefer {@link #withCharset} for setting the {@code charset} parameter
435   * when using a {@link Charset} object.
436   *
437   * @throws IllegalArgumentException if either {@code attribute} or {@code value} is invalid
438   */
439  public MediaType withParameter(String attribute, String value) {
440    checkNotNull(attribute);
441    checkNotNull(value);
442    String normalizedAttribute = normalizeToken(attribute);
443    ImmutableListMultimap.Builder<String, String> builder = ImmutableListMultimap.builder();
444    for (Entry<String, String> entry : parameters.entries()) {
445      String key = entry.getKey();
446      if (!normalizedAttribute.equals(key)) {
447        builder.put(key, entry.getValue());
448      }
449    }
450    builder.put(normalizedAttribute, normalizeParameterValue(normalizedAttribute, value));
451    MediaType mediaType = new MediaType(type, subtype, builder.build());
452    // Return one of the constants if the media type is a known type.
453    return MoreObjects.firstNonNull(KNOWN_TYPES.get(mediaType), mediaType);
454  }
455
456  /**
457   * Returns a new instance with the same type and subtype as this instance, with the
458   * {@code charset} parameter set to the {@link Charset#name name} of the given charset. Only one
459   * {@code charset} parameter will be present on the new instance regardless of the number set on
460   * this one.
461   *
462   * <p>If a charset must be specified that is not supported on this JVM (and thus is not
463   * representable as a {@link Charset} instance, use {@link #withParameter}.
464   */
465  public MediaType withCharset(Charset charset) {
466    checkNotNull(charset);
467    return withParameter(CHARSET_ATTRIBUTE, charset.name());
468  }
469
470  /** Returns true if either the type or subtype is the wildcard. */
471  public boolean hasWildcard() {
472    return WILDCARD.equals(type) || WILDCARD.equals(subtype);
473  }
474
475  /**
476   * Returns {@code true} if this instance falls within the range (as defined by
477   * <a href="http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html">the HTTP Accept header</a>)
478   * given by the argument according to three criteria:
479   *
480   * <ol>
481   * <li>The type of the argument is the wildcard or equal to the type of this instance.
482   * <li>The subtype of the argument is the wildcard or equal to the subtype of this instance.
483   * <li>All of the parameters present in the argument are present in this instance.
484   * </ol>
485   *
486   * <p>For example: <pre>   {@code
487   *   PLAIN_TEXT_UTF_8.is(PLAIN_TEXT_UTF_8) // true
488   *   PLAIN_TEXT_UTF_8.is(HTML_UTF_8) // false
489   *   PLAIN_TEXT_UTF_8.is(ANY_TYPE) // true
490   *   PLAIN_TEXT_UTF_8.is(ANY_TEXT_TYPE) // true
491   *   PLAIN_TEXT_UTF_8.is(ANY_IMAGE_TYPE) // false
492   *   PLAIN_TEXT_UTF_8.is(ANY_TEXT_TYPE.withCharset(UTF_8)) // true
493   *   PLAIN_TEXT_UTF_8.withoutParameters().is(ANY_TEXT_TYPE.withCharset(UTF_8)) // false
494   *   PLAIN_TEXT_UTF_8.is(ANY_TEXT_TYPE.withCharset(UTF_16)) // false}</pre>
495   *
496   * <p>Note that while it is possible to have the same parameter declared multiple times within a
497   * media type this method does not consider the number of occurrences of a parameter.  For
498   * example, {@code "text/plain; charset=UTF-8"} satisfies
499   * {@code "text/plain; charset=UTF-8; charset=UTF-8"}.
500   */
501  public boolean is(MediaType mediaTypeRange) {
502    return (mediaTypeRange.type.equals(WILDCARD) || mediaTypeRange.type.equals(this.type))
503        && (mediaTypeRange.subtype.equals(WILDCARD) || mediaTypeRange.subtype.equals(this.subtype))
504        && this.parameters.entries().containsAll(mediaTypeRange.parameters.entries());
505  }
506
507  /**
508   * Creates a new media type with the given type and subtype.
509   *
510   * @throws IllegalArgumentException if type or subtype is invalid or if a wildcard is used for the
511   * type, but not the subtype.
512   */
513  public static MediaType create(String type, String subtype) {
514    return create(type, subtype, ImmutableListMultimap.<String, String>of());
515  }
516
517  /**
518   * Creates a media type with the "application" type and the given subtype.
519   *
520   * @throws IllegalArgumentException if subtype is invalid
521   */
522  static MediaType createApplicationType(String subtype) {
523    return create(APPLICATION_TYPE, subtype);
524  }
525
526  /**
527   * Creates a media type with the "audio" type and the given subtype.
528   *
529   * @throws IllegalArgumentException if subtype is invalid
530   */
531  static MediaType createAudioType(String subtype) {
532    return create(AUDIO_TYPE, subtype);
533  }
534
535  /**
536   * Creates a media type with the "image" type and the given subtype.
537   *
538   * @throws IllegalArgumentException if subtype is invalid
539   */
540  static MediaType createImageType(String subtype) {
541    return create(IMAGE_TYPE, subtype);
542  }
543
544  /**
545   * Creates a media type with the "text" type and the given subtype.
546   *
547   * @throws IllegalArgumentException if subtype is invalid
548   */
549  static MediaType createTextType(String subtype) {
550    return create(TEXT_TYPE, subtype);
551  }
552
553  /**
554   * Creates a media type with the "video" type and the given subtype.
555   *
556   * @throws IllegalArgumentException if subtype is invalid
557   */
558  static MediaType createVideoType(String subtype) {
559    return create(VIDEO_TYPE, subtype);
560  }
561
562  private static MediaType create(String type, String subtype,
563      Multimap<String, String> parameters) {
564    checkNotNull(type);
565    checkNotNull(subtype);
566    checkNotNull(parameters);
567    String normalizedType = normalizeToken(type);
568    String normalizedSubtype = normalizeToken(subtype);
569    checkArgument(!WILDCARD.equals(normalizedType) || WILDCARD.equals(normalizedSubtype),
570        "A wildcard type cannot be used with a non-wildcard subtype");
571    ImmutableListMultimap.Builder<String, String> builder = ImmutableListMultimap.builder();
572    for (Entry<String, String> entry : parameters.entries()) {
573      String attribute = normalizeToken(entry.getKey());
574      builder.put(attribute, normalizeParameterValue(attribute, entry.getValue()));
575    }
576    MediaType mediaType = new MediaType(normalizedType, normalizedSubtype, builder.build());
577    // Return one of the constants if the media type is a known type.
578    return MoreObjects.firstNonNull(KNOWN_TYPES.get(mediaType), mediaType);
579  }
580
581  private static String normalizeToken(String token) {
582    checkArgument(TOKEN_MATCHER.matchesAllOf(token));
583    return Ascii.toLowerCase(token);
584  }
585
586  private static String normalizeParameterValue(String attribute, String value) {
587    return CHARSET_ATTRIBUTE.equals(attribute) ? Ascii.toLowerCase(value) : value;
588  }
589
590  /**
591   * Parses a media type from its string representation.
592   *
593   * @throws IllegalArgumentException if the input is not parsable
594   */
595  public static MediaType parse(String input) {
596    checkNotNull(input);
597    Tokenizer tokenizer = new Tokenizer(input);
598    try {
599      String type = tokenizer.consumeToken(TOKEN_MATCHER);
600      tokenizer.consumeCharacter('/');
601      String subtype = tokenizer.consumeToken(TOKEN_MATCHER);
602      ImmutableListMultimap.Builder<String, String> parameters = ImmutableListMultimap.builder();
603      while (tokenizer.hasMore()) {
604        tokenizer.consumeCharacter(';');
605        tokenizer.consumeTokenIfPresent(LINEAR_WHITE_SPACE);
606        String attribute = tokenizer.consumeToken(TOKEN_MATCHER);
607        tokenizer.consumeCharacter('=');
608        final String value;
609        if ('"' == tokenizer.previewChar()) {
610          tokenizer.consumeCharacter('"');
611          StringBuilder valueBuilder = new StringBuilder();
612          while ('"' != tokenizer.previewChar()) {
613            if ('\\' == tokenizer.previewChar()) {
614              tokenizer.consumeCharacter('\\');
615              valueBuilder.append(tokenizer.consumeCharacter(ASCII));
616            } else {
617              valueBuilder.append(tokenizer.consumeToken(QUOTED_TEXT_MATCHER));
618            }
619          }
620          value = valueBuilder.toString();
621          tokenizer.consumeCharacter('"');
622        } else {
623          value = tokenizer.consumeToken(TOKEN_MATCHER);
624        }
625        parameters.put(attribute, value);
626      }
627      return create(type, subtype, parameters.build());
628    } catch (IllegalStateException e) {
629      throw new IllegalArgumentException("Could not parse '" + input + "'", e);
630    }
631  }
632
633  private static final class Tokenizer {
634    final String input;
635    int position = 0;
636
637    Tokenizer(String input) {
638      this.input = input;
639    }
640
641    String consumeTokenIfPresent(CharMatcher matcher) {
642      checkState(hasMore());
643      int startPosition = position;
644      position = matcher.negate().indexIn(input, startPosition);
645      return hasMore() ? input.substring(startPosition, position) : input.substring(startPosition);
646    }
647
648    String consumeToken(CharMatcher matcher) {
649      int startPosition = position;
650      String token = consumeTokenIfPresent(matcher);
651      checkState(position != startPosition);
652      return token;
653    }
654
655    char consumeCharacter(CharMatcher matcher) {
656      checkState(hasMore());
657      char c = previewChar();
658      checkState(matcher.matches(c));
659      position++;
660      return c;
661    }
662
663    char consumeCharacter(char c) {
664      checkState(hasMore());
665      checkState(previewChar() == c);
666      position++;
667      return c;
668    }
669
670    char previewChar() {
671      checkState(hasMore());
672      return input.charAt(position);
673    }
674
675    boolean hasMore() {
676      return (position >= 0) && (position < input.length());
677    }
678  }
679
680  @Override public boolean equals(@Nullable Object obj) {
681    if (obj == this) {
682      return true;
683    } else if (obj instanceof MediaType) {
684      MediaType that = (MediaType) obj;
685      return this.type.equals(that.type)
686          && this.subtype.equals(that.subtype)
687          // compare parameters regardless of order
688          && this.parametersAsMap().equals(that.parametersAsMap());
689    } else {
690      return false;
691    }
692  }
693
694  @Override public int hashCode() {
695    return Objects.hashCode(type, subtype, parametersAsMap());
696  }
697
698  private static final MapJoiner PARAMETER_JOINER = Joiner.on("; ").withKeyValueSeparator("=");
699
700  /**
701   * Returns the string representation of this media type in the format described in <a
702   * href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>.
703   */
704  @Override public String toString() {
705    StringBuilder builder = new StringBuilder().append(type).append('/').append(subtype);
706    if (!parameters.isEmpty()) {
707      builder.append("; ");
708      Multimap<String, String> quotedParameters = Multimaps.transformValues(parameters,
709          new Function<String, String>() {
710            @Override public String apply(String value) {
711              return TOKEN_MATCHER.matchesAllOf(value) ? value : escapeAndQuote(value);
712            }
713          });
714      PARAMETER_JOINER.appendTo(builder, quotedParameters.entries());
715    }
716    return builder.toString();
717  }
718
719  private static String escapeAndQuote(String value) {
720    StringBuilder escaped = new StringBuilder(value.length() + 16).append('"');
721    for (char ch : value.toCharArray()) {
722      if (ch == '\r' || ch == '\\' || ch == '"') {
723        escaped.append('\\');
724      }
725      escaped.append(ch);
726    }
727    return escaped.append('"').toString();
728  }
729
730}