MimeUtilityTest.java revision 39121c758dcc919b5fde4c893d488916e26d3140
1/*
2 * Copyright (C) 2008 The Android Open Source Project
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 *      http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
15 */
16
17package com.android.emailcommon.internet;
18
19import com.android.emailcommon.internet.MimeBodyPart;
20import com.android.emailcommon.internet.MimeHeader;
21import com.android.emailcommon.internet.MimeUtility;
22import com.android.emailcommon.internet.TextBody;
23import com.android.emailcommon.mail.BodyPart;
24import com.android.emailcommon.mail.Message;
25import com.android.emailcommon.mail.MessageTestUtils;
26import com.android.emailcommon.mail.MessagingException;
27import com.android.emailcommon.mail.Part;
28import com.android.emailcommon.mail.MessageTestUtils.MessageBuilder;
29import com.android.emailcommon.mail.MessageTestUtils.MultipartBuilder;
30
31import android.test.suitebuilder.annotation.SmallTest;
32
33import java.util.ArrayList;
34
35import junit.framework.TestCase;
36
37/**
38 * This is a series of unit tests for the MimeUtility class.  These tests must be locally
39 * complete - no server(s) required.
40 */
41@SmallTest
42public class MimeUtilityTest extends TestCase {
43
44    /** up arrow, down arrow, left arrow, right arrow */
45    private final String SHORT_UNICODE = "\u2191\u2193\u2190\u2192";
46    private final String SHORT_UNICODE_ENCODED = "=?UTF-8?B?4oaR4oaT4oaQ4oaS?=";
47
48    /** dollar and euro sign */
49    private final String PADDED2_UNICODE = "$\u20AC";
50    private final String PADDED2_UNICODE_ENCODED = "=?UTF-8?B?JOKCrA==?=";
51    private final String PADDED1_UNICODE = "$$\u20AC";
52    private final String PADDED1_UNICODE_ENCODED = "=?UTF-8?B?JCTigqw=?=";
53    private final String PADDED0_UNICODE = "$$$\u20AC";
54    private final String PADDED0_UNICODE_ENCODED = "=?UTF-8?B?JCQk4oKs?=";
55
56    /** a string without any unicode */
57    private final String SHORT_PLAIN = "abcd";
58
59    /** long subject which will be split into two MIME/Base64 chunks */
60    private final String LONG_UNICODE_SPLIT =
61        "$" +
62        "\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC" +
63        "\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC\u20AC";
64    private final String LONG_UNICODE_SPLIT_ENCODED =
65        "=?UTF-8?B?JOKCrOKCrOKCrOKCrOKCrOKCrOKCrOKCrA==?=" + "\r\n " +
66        "=?UTF-8?B?4oKs4oKs4oKs4oKs4oKs4oKs4oKs4oKs4oKs4oKs4oKs4oKs?=";
67
68    /** strings that use supplemental characters and really stress encode/decode */
69    // actually it's U+10400
70    private final String SHORT_SUPPLEMENTAL = "\uD801\uDC00";
71    private final String SHORT_SUPPLEMENTAL_ENCODED = "=?UTF-8?B?8JCQgA==?=";
72    private final String LONG_SUPPLEMENTAL = SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL +
73        SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL +
74        SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL;
75    private final String LONG_SUPPLEMENTAL_ENCODED =
76        "=?UTF-8?B?8JCQgPCQkIDwkJCA8JCQgA==?=" + "\r\n " +
77        "=?UTF-8?B?8JCQgPCQkIDwkJCA8JCQgPCQkIDwkJCA?=";
78    private final String LONG_SUPPLEMENTAL_2 = "a" + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL +
79        SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL +
80        SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL + SHORT_SUPPLEMENTAL;
81    private final String LONG_SUPPLEMENTAL_ENCODED_2 =
82        "=?UTF-8?B?YfCQkIDwkJCA8JCQgPCQkIA=?=" + "\r\n " +
83        "=?UTF-8?B?8JCQgPCQkIDwkJCA8JCQgPCQkIDwkJCA?=";
84    // Earth is U+1D300.
85    private final String LONG_SUPPLEMENTAL_QP =
86        "*Monogram for Earth \uD834\uDF00. Monogram for Human \u268b.";
87    private final String LONG_SUPPLEMENTAL_QP_ENCODED =
88        "=?UTF-8?Q?*Monogram_for_Earth_?=" + "\r\n " +
89        "=?UTF-8?Q?=F0=9D=8C=80._Monogram_for_Human_=E2=9A=8B.?=";
90
91    /** a typical no-param header */
92    private final String HEADER_NO_PARAMETER =
93            "header";
94    /** a typical multi-param header */
95    private final String HEADER_MULTI_PARAMETER =
96            "header; Param1Name=Param1Value; Param2Name=Param2Value";
97    /** a multi-param header with quoting */
98    private final String HEADER_QUOTED_MULTI_PARAMETER =
99            "header; Param1Name=\"Param1Value\"; Param2Name=\"Param2Value\"";
100    /** a malformed header we're seeing in production servers */
101    private final String HEADER_MALFORMED_PARAMETER =
102            "header; Param1Name=Param1Value; filename";
103
104    /**
105     * a string generated by google calendar that contains two interesting gotchas:
106     * 1.  Uses windows-1252 encoding, and en-dash recoded appropriately (\u2013 / =96)
107     * 2.  Because the first encoded char requires '=XX' encoding, we create an "internal"
108     *     "?=" that the decoder must correctly skip over.
109     **/
110    private final String CALENDAR_SUBJECT_UNICODE =
111        "=?windows-1252?Q?=5BReminder=5D_test_=40_Fri_Mar_20_10=3A30am_=96_11am_=28andro?=" +
112        "\r\n\t" +
113        "=?windows-1252?Q?id=2Etr=40gmail=2Ecom=29?=";
114    private final String CALENDAR_SUBJECT_PLAIN =
115        "[Reminder] test @ Fri Mar 20 10:30am \u2013 11am (android.tr@gmail.com)";
116
117    /**
118     * Some basic degenerate strings designed to exercise error handling in the decoder
119     */
120    private final String CALENDAR_DEGENERATE_UNICODE_1 =
121        "=?windows-1252?Q=5B?=";
122    private final String CALENDAR_DEGENERATE_UNICODE_2 =
123        "=?windows-1252Q?=5B?=";
124    private final String CALENDAR_DEGENERATE_UNICODE_3 =
125        "=?windows-1252?=";
126    private final String CALENDAR_DEGENERATE_UNICODE_4 =
127        "=?windows-1252";
128
129    /**
130     * Test that decode/unfold is efficient when it can be
131     */
132    public void testEfficientUnfoldAndDecode() {
133        String result1 = MimeUtility.unfold(SHORT_PLAIN);
134        String result2 = MimeUtility.decode(SHORT_PLAIN);
135        String result3 = MimeUtility.unfoldAndDecode(SHORT_PLAIN);
136
137        assertSame(SHORT_PLAIN, result1);
138        assertSame(SHORT_PLAIN, result2);
139        assertSame(SHORT_PLAIN, result3);
140    }
141
142    // TODO:  more tests for unfold(String s)
143
144    /**
145     * Test that decode is working for simple strings
146     */
147    public void testDecodeSimple() {
148        String result1 = MimeUtility.decode(SHORT_UNICODE_ENCODED);
149        assertEquals(SHORT_UNICODE, result1);
150    }
151
152    // TODO:  tests for decode(String s)
153
154    /**
155     * Test that unfoldAndDecode is working for simple strings
156     */
157    public void testUnfoldAndDecodeSimple() {
158        String result1 = MimeUtility.unfoldAndDecode(SHORT_UNICODE_ENCODED);
159        assertEquals(SHORT_UNICODE, result1);
160    }
161
162    /**
163     * test decoding complex string from google calendar that has two gotchas for the decoder.
164     * also tests a couple of degenerate cases that should "fail" decoding and pass through.
165     */
166    public void testComplexDecode() {
167        String result1 = MimeUtility.unfoldAndDecode(CALENDAR_SUBJECT_UNICODE);
168        assertEquals(CALENDAR_SUBJECT_PLAIN, result1);
169
170        // These degenerate cases should "fail" and return the same string
171        String degenerate1 = MimeUtility.unfoldAndDecode(CALENDAR_DEGENERATE_UNICODE_1);
172        assertEquals("degenerate case 1", CALENDAR_DEGENERATE_UNICODE_1, degenerate1);
173        String degenerate2 = MimeUtility.unfoldAndDecode(CALENDAR_DEGENERATE_UNICODE_2);
174        assertEquals("degenerate case 2", CALENDAR_DEGENERATE_UNICODE_2, degenerate2);
175        String degenerate3 = MimeUtility.unfoldAndDecode(CALENDAR_DEGENERATE_UNICODE_3);
176        assertEquals("degenerate case 3", CALENDAR_DEGENERATE_UNICODE_3, degenerate3);
177        String degenerate4 = MimeUtility.unfoldAndDecode(CALENDAR_DEGENERATE_UNICODE_4);
178        assertEquals("degenerate case 4", CALENDAR_DEGENERATE_UNICODE_4, degenerate4);
179    }
180
181    // TODO:  more tests for unfoldAndDecode(String s)
182
183    /**
184     * Test that fold/encode is efficient when it can be
185     */
186    public void testEfficientFoldAndEncode() {
187        String result1 = MimeUtility.foldAndEncode(SHORT_PLAIN);
188        String result2 = MimeUtility.foldAndEncode2(SHORT_PLAIN, 10);
189        String result3 = MimeUtility.fold(SHORT_PLAIN, 10);
190
191        assertSame(SHORT_PLAIN, result1);
192        assertSame(SHORT_PLAIN, result2);
193        assertSame(SHORT_PLAIN, result3);
194    }
195
196    /**
197     * Test about base64 padding variety.
198     */
199    public void testPaddingOfFoldAndEncode2() {
200        String result1 = MimeUtility.foldAndEncode2(PADDED2_UNICODE, 0);
201        String result2 = MimeUtility.foldAndEncode2(PADDED1_UNICODE, 0);
202        String result3 = MimeUtility.foldAndEncode2(PADDED0_UNICODE, 0);
203
204        assertEquals("padding 2", PADDED2_UNICODE_ENCODED, result1);
205        assertEquals("padding 1", PADDED1_UNICODE_ENCODED, result2);
206        assertEquals("padding 0", PADDED0_UNICODE_ENCODED, result3);
207    }
208
209    // TODO:  more tests for foldAndEncode(String s)
210
211    /**
212     * Test that foldAndEncode2 is working for simple strings
213     */
214    public void testFoldAndEncode2() {
215        String result1 = MimeUtility.foldAndEncode2(SHORT_UNICODE, 10);
216        assertEquals(SHORT_UNICODE_ENCODED, result1);
217    }
218
219    /**
220     * Test that foldAndEncode2 is working for long strings which needs splitting.
221     */
222    public void testFoldAndEncode2WithLongSplit() {
223        String result = MimeUtility.foldAndEncode2(LONG_UNICODE_SPLIT, "Subject: ".length());
224
225        assertEquals("long string", LONG_UNICODE_SPLIT_ENCODED, result);
226    }
227
228    /**
229     * Tests of foldAndEncode2 that involve supplemental characters (UTF-32)
230     *
231     * Note that the difference between LONG_SUPPLEMENTAL and LONG_SUPPLEMENTAL_2 is the
232     * insertion of a single character at the head of the string. This is intended to disrupt
233     * the code that splits the long string into multiple encoded words, and confirm that it
234     * properly applies the breaks between UTF-32 code points.
235     */
236    public void testFoldAndEncode2Supplemental() {
237        String result1 = MimeUtility.foldAndEncode2(SHORT_SUPPLEMENTAL, "Subject: ".length());
238        String result2 = MimeUtility.foldAndEncode2(LONG_SUPPLEMENTAL, "Subject: ".length());
239        String result3 = MimeUtility.foldAndEncode2(LONG_SUPPLEMENTAL_2, "Subject: ".length());
240        assertEquals("short supplemental", SHORT_SUPPLEMENTAL_ENCODED, result1);
241        assertEquals("long supplemental", LONG_SUPPLEMENTAL_ENCODED, result2);
242        assertEquals("long supplemental 2", LONG_SUPPLEMENTAL_ENCODED_2, result3);
243    }
244
245    /**
246     * Tests of foldAndEncode2 that involve supplemental characters (UTF-32)
247     *
248     * Note that the difference between LONG_SUPPLEMENTAL and LONG_SUPPLEMENTAL_QP is that
249     * the former will be encoded as base64 but the latter will be encoded as quoted printable.
250     */
251    public void testFoldAndEncode2SupplementalQuotedPrintable() {
252        String result = MimeUtility.foldAndEncode2(LONG_SUPPLEMENTAL_QP, "Subject: ".length());
253        assertEquals("long supplement quoted printable",
254                     LONG_SUPPLEMENTAL_QP_ENCODED, result);
255    }
256
257    // TODO:  more tests for foldAndEncode2(String s)
258    // TODO:  more tests for fold(String s, int usedCharacters)
259
260    /**
261     * Basic tests of getHeaderParameter()
262     *
263     * Typical header value:  multipart/mixed; boundary="----E5UGTXUQQJV80DR8SJ88F79BRA4S8K"
264     *
265     * Function spec says:
266     *  if header is null:  return null
267     *  if name is null:    if params, return first param.  else return full field
268     *  else:               if param is found (case insensitive) return it
269     *                        else return null
270     */
271    public void testGetHeaderParameter() {
272        // if header is null, return null
273        assertNull("null header check", MimeUtility.getHeaderParameter(null, "name"));
274
275        // if name is null, return first param or full header
276        // NOTE:  The docs are wrong - it returns the header (no params) in that case
277//      assertEquals("null name first param per docs", "Param1Value",
278//              MimeUtility.getHeaderParameter(HEADER_MULTI_PARAMETER, null));
279        assertEquals("null name first param per code", "header",
280                MimeUtility.getHeaderParameter(HEADER_MULTI_PARAMETER, null));
281        assertEquals("null name full header", HEADER_NO_PARAMETER,
282                MimeUtility.getHeaderParameter(HEADER_NO_PARAMETER, null));
283
284        // find name
285        assertEquals("get 1st param", "Param1Value",
286                MimeUtility.getHeaderParameter(HEADER_MULTI_PARAMETER, "Param1Name"));
287        assertEquals("get 2nd param", "Param2Value",
288                MimeUtility.getHeaderParameter(HEADER_MULTI_PARAMETER, "Param2Name"));
289        assertEquals("get missing param", null,
290                MimeUtility.getHeaderParameter(HEADER_MULTI_PARAMETER, "Param3Name"));
291
292        // case insensitivity
293        assertEquals("get 2nd param all LC", "Param2Value",
294                MimeUtility.getHeaderParameter(HEADER_MULTI_PARAMETER, "param2name"));
295        assertEquals("get 2nd param all UC", "Param2Value",
296                MimeUtility.getHeaderParameter(HEADER_MULTI_PARAMETER, "PARAM2NAME"));
297
298        // quoting
299        assertEquals("get 1st param", "Param1Value",
300                MimeUtility.getHeaderParameter(HEADER_QUOTED_MULTI_PARAMETER, "Param1Name"));
301        assertEquals("get 2nd param", "Param2Value",
302                MimeUtility.getHeaderParameter(HEADER_QUOTED_MULTI_PARAMETER, "Param2Name"));
303
304        // Don't fail when malformed
305        assertEquals("malformed filename param", null,
306                MimeUtility.getHeaderParameter(HEADER_MALFORMED_PARAMETER, "filename"));
307    }
308
309    // TODO:  tests for findFirstPartByMimeType(Part part, String mimeType)
310
311    /** Tests for findPartByContentId(Part part, String contentId) */
312    public void testFindPartByContentIdTestCase() throws MessagingException, Exception {
313        final String cid1 = "cid.1@android.com";
314        final Part cid1bp = MessageTestUtils.bodyPart("image/gif", cid1);
315        final String cid2 = "cid.2@android.com";
316        final Part cid2bp = MessageTestUtils.bodyPart("image/gif", "<" + cid2 + ">");
317
318        final Message msg1 = new MessageBuilder()
319            .setBody(new MultipartBuilder("multipart/related")
320                 .addBodyPart(MessageTestUtils.bodyPart("text/html", null))
321                 .addBodyPart((BodyPart)cid1bp)
322                 .build())
323            .build();
324        // found cid1 part
325        final Part actual1_1 = MimeUtility.findPartByContentId(msg1, cid1);
326        assertEquals("could not found expected content-id part", cid1bp, actual1_1);
327
328        final Message msg2 = new MessageBuilder()
329            .setBody(new MultipartBuilder("multipart/mixed")
330                .addBodyPart(MessageTestUtils.bodyPart("image/tiff", "cid.4@android.com"))
331                .addBodyPart(new MultipartBuilder("multipart/related")
332                    .addBodyPart(new MultipartBuilder("multipart/alternative")
333                        .addBodyPart(MessageTestUtils.bodyPart("text/plain", null))
334                        .addBodyPart(MessageTestUtils.bodyPart("text/html", null))
335                        .buildBodyPart())
336                    .addBodyPart((BodyPart)cid1bp)
337                    .buildBodyPart())
338                .addBodyPart(MessageTestUtils.bodyPart("image/gif", "cid.3@android.com"))
339                .addBodyPart((BodyPart)cid2bp)
340                .build())
341            .build();
342        // found cid1 part
343        final Part actual2_1 = MimeUtility.findPartByContentId(msg2, cid1);
344        assertEquals("found part from related multipart", cid1bp, actual2_1);
345
346        // found cid2 part
347        final Part actual2_2 = MimeUtility.findPartByContentId(msg2, cid2);
348        assertEquals("found part from mixed multipart", cid2bp, actual2_2);
349    }
350
351    /** Tests for findPartByContentId(Part part, String contentId) */
352    public void testCollectParts() throws MessagingException, Exception {
353        // golden cases; these will marked as attachments
354        final String cid1 = "<i_12e8248b4f0874cb>";
355        final Part cid1bp = MessageTestUtils.bodyPart("image/gif; name=\"im1.gif\"", cid1);
356        final String cid2 = "<ii_12e8248b4f0874cb>";
357        final Part cid2bp = MessageTestUtils.bodyPart("image/gif", cid2);
358        cid2bp.addHeader(MimeHeader.HEADER_CONTENT_DISPOSITION, "inline; filename=\"im2.gif\"");
359        final String cid3 = "<iii_12e8248b4f0874cb>";
360        final Part cid3bp = MessageTestUtils.bodyPart("image/gif", cid3);
361        cid3bp.addHeader(MimeHeader.HEADER_CONTENT_DISPOSITION, "attachment; filename=\"im3.gif\"");
362        // error cases; these will NOT be marked as attachments
363        final String cid4 = "<iv_12e8248b4f0874cb>";
364        final Part cid4bp = MessageTestUtils.bodyPart("image/gif", cid4);  // no name attr
365        final String cid5 = "<v_12e8248b4f0874cb>";
366        final Part cid5bp = MessageTestUtils.bodyPart("image/gif", cid5);
367        cid5bp.addHeader(MimeHeader.HEADER_CONTENT_DISPOSITION, "inline"); // no filename attr
368
369        // Default content disposition
370        final ArrayList<Part> view1 = new ArrayList<Part>();
371        final ArrayList<Part> attach1 = new ArrayList<Part>();
372        MimeUtility.collectParts(cid1bp, view1, attach1);
373        assertEquals(1, attach1.size());
374        assertEquals(attach1.get(0), cid1bp);
375
376        // Explicit content disposition of "inline"
377        final ArrayList<Part> view2 = new ArrayList<Part>();
378        final ArrayList<Part> attach2 = new ArrayList<Part>();
379        MimeUtility.collectParts(cid2bp, view2, attach2);
380        assertEquals(1, attach2.size());
381        assertEquals(attach2.get(0), cid2bp);
382
383        // Explicit content disposition of "attachment"
384        final ArrayList<Part> view3 = new ArrayList<Part>();
385        final ArrayList<Part> attach3 = new ArrayList<Part>();
386        MimeUtility.collectParts(cid3bp, view3, attach3);
387        assertEquals(1, attach3.size());
388        assertEquals(attach3.get(0), cid3bp);
389
390        // Default content disposition; missing name attribute on content-type
391        final ArrayList<Part> view4 = new ArrayList<Part>();
392        final ArrayList<Part> attach4 = new ArrayList<Part>();
393        MimeUtility.collectParts(cid4bp, view4, attach4);
394        assertEquals(0, attach4.size());
395
396        // Content disposition of "inline"; missing filename attribute
397        final ArrayList<Part> view5 = new ArrayList<Part>();
398        final ArrayList<Part> attach5 = new ArrayList<Part>();
399        MimeUtility.collectParts(cid5bp, view5, attach5);
400        assertEquals(0, attach5.size());
401    }
402
403    /** Tests for getTextFromPart(Part part) */
404    public void testGetTextFromPartContentTypeCase() throws MessagingException {
405        final String theText = "This is the text of the part";
406        TextBody tb = new TextBody(theText);
407        MimeBodyPart p = new MimeBodyPart();
408
409        // 1. test basic text/plain mode
410        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE, "text/plain");
411        p.setBody(tb);
412        String gotText = MimeUtility.getTextFromPart(p);
413        assertEquals(theText, gotText);
414
415        // 2. mixed case is OK
416        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE, "TEXT/PLAIN");
417        p.setBody(tb);
418        gotText = MimeUtility.getTextFromPart(p);
419        assertEquals(theText, gotText);
420
421        // 3. wildcards OK
422        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE, "text/other");
423        p.setBody(tb);
424        gotText = MimeUtility.getTextFromPart(p);
425        assertEquals(theText, gotText);
426    }
427
428    /** Test for usage of Content-Type in getTextFromPart(Part part).
429     *
430     * For example 'Content-Type: text/html; charset=utf-8'
431     *
432     *  If the body part has no mime-type, refuses to parse content as text.
433     *  If the mime-type does not match text/*, it will not get parsed.
434     *  Then, the charset parameter is used, with a default of ASCII.
435     *
436     *  This test works by using a string that is valid Unicode, and is also
437     *  valid when decoded from UTF-8 bytes into Windows-1252 (so that
438     *  auto-detection is not possible), and checks that the correct conversion
439     *  was made, based on the Content-Type header.
440     *
441     */
442    public void testContentTypeCharset() throws MessagingException {
443        final String UNICODE_EXPECT = "This is some happy unicode text \u263a";
444        // What you get if you encode to UTF-8 (\xe2\x98\xba) and reencode with Windows-1252
445        final String WINDOWS1252_EXPECT = "This is some happy unicode text \u00e2\u02dc\u00ba";
446        TextBody tb = new TextBody(UNICODE_EXPECT);
447        MimeBodyPart p = new MimeBodyPart();
448
449        String gotText, mimeType, charset;
450        // TEST 0: Standard Content-Type header; no extraneous spaces or fields
451        p.setBody(tb);
452        // We call setHeader after setBody, since setBody overwrites Content-Type
453        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE, "text/html; charset=utf-8");
454        gotText = MimeUtility.getTextFromPart(p);
455        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
456        assertEquals(UNICODE_EXPECT, gotText);
457
458        p.setBody(tb);
459        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE, "text/html; charset=windows-1252");
460        gotText = MimeUtility.getTextFromPart(p);
461        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
462        assertEquals(WINDOWS1252_EXPECT, gotText);
463
464        // TEST 1: Extra fields and quotes in Content-Type (from RFC 2045)
465        p.setBody(tb);
466        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE,
467                    "text/html; prop1 = \"test\"; charset = \"utf-8\"; prop2 = \"test\"");
468        gotText = MimeUtility.getTextFromPart(p);
469        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
470        assertEquals(UNICODE_EXPECT, gotText);
471
472        p.setBody(tb);
473        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE,
474                    "text/html; prop1 = \"test\"; charset = \"windows-1252\"; prop2 = \"test\"");
475        gotText = MimeUtility.getTextFromPart(p);
476        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
477        assertEquals(WINDOWS1252_EXPECT, gotText);
478
479        // TEST 2: Mixed case in Content-Type header:
480        // RFC 2045 says that content types, subtypes and parameter names
481        // are case-insensitive.
482
483        p.setBody(tb);
484        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE, "TEXT/HtmL ; CHARseT=utf-8");
485        gotText = MimeUtility.getTextFromPart(p);
486        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
487        assertEquals(UNICODE_EXPECT, gotText);
488
489        p.setBody(tb);
490        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE, "TEXT/HtmL ; CHARseT=windows-1252");
491        gotText = MimeUtility.getTextFromPart(p);
492        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
493        assertEquals(WINDOWS1252_EXPECT, gotText);
494
495        // TEST 3: Comments in Content-Type header field (from RFC 2045)
496        // Thunderbird permits comments after the end of a parameter, as in this example.
497        // Not something that I have seen in the real world outside RFC 2045.
498
499        p.setBody(tb);
500        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE,
501                    "text/html; charset=utf-8 (Plain text)");
502        gotText = MimeUtility.getTextFromPart(p);
503        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
504        // Note: This test does not pass.
505        //assertEquals(UNICODE_EXPECT, gotText);
506
507        p.setBody(tb);
508        p.setHeader(MimeHeader.HEADER_CONTENT_TYPE,
509                    "text/html; charset=windows-1252 (Plain text)");
510        gotText = MimeUtility.getTextFromPart(p);
511        assertTrue(MimeUtility.mimeTypeMatches(p.getMimeType(), "text/html"));
512        // Note: These tests does not pass.
513        //assertEquals(WINDOWS1252_EXPECT, gotText);
514    }
515
516    /** Tests for various aspects of mimeTypeMatches(String mimeType, String matchAgainst) */
517    public void testMimeTypeMatches() {
518        // 1. No match
519        assertFalse(MimeUtility.mimeTypeMatches("foo/bar", "TEXT/PLAIN"));
520
521        // 2. Match
522        assertTrue(MimeUtility.mimeTypeMatches("text/plain", "text/plain"));
523
524        // 3. Match (mixed case)
525        assertTrue(MimeUtility.mimeTypeMatches("text/plain", "TEXT/PLAIN"));
526        assertTrue(MimeUtility.mimeTypeMatches("TEXT/PLAIN", "text/plain"));
527
528        // 4. Match (wildcards)
529        assertTrue(MimeUtility.mimeTypeMatches("text/plain", "*/plain"));
530        assertTrue(MimeUtility.mimeTypeMatches("text/plain", "text/*"));
531        assertTrue(MimeUtility.mimeTypeMatches("text/plain", "*/*"));
532
533        // 5. No Match (wildcards)
534        assertFalse(MimeUtility.mimeTypeMatches("foo/bar", "*/plain"));
535        assertFalse(MimeUtility.mimeTypeMatches("foo/bar", "text/*"));
536    }
537
538    /** Tests for various aspects of mimeTypeMatches(String mimeType, String[] matchAgainst) */
539    public void testMimeTypeMatchesArray() {
540        // 1. Zero-length array
541        String[] arrayZero = new String[0];
542        assertFalse(MimeUtility.mimeTypeMatches("text/plain", arrayZero));
543
544        // 2. Single entry, no match
545        String[] arrayOne = new String[] { "text/plain" };
546        assertFalse(MimeUtility.mimeTypeMatches("foo/bar", arrayOne));
547
548        // 3. Single entry, match
549        assertTrue(MimeUtility.mimeTypeMatches("text/plain", arrayOne));
550
551        // 4. Multi entry, no match
552        String[] arrayTwo = new String[] { "text/plain", "match/this" };
553        assertFalse(MimeUtility.mimeTypeMatches("foo/bar", arrayTwo));
554
555        // 5. Multi entry, match first
556        assertTrue(MimeUtility.mimeTypeMatches("text/plain", arrayTwo));
557
558        // 6. Multi entry, match not first
559        assertTrue(MimeUtility.mimeTypeMatches("match/this", arrayTwo));
560    }
561
562    // TODO:  tests for decodeBody(InputStream in, String contentTransferEncoding)
563    // TODO:  tests for collectParts(Part part, ArrayList<Part> viewables, ArrayList<Part> attachments)
564
565}
566