xref: /trunk/test/testuno/source/api/i18n/XCharacterClassificationTest.java (revision 31bbceb0f9d64c0c2c3b22a794a1666c1f33396e)
1 /**************************************************************
2  *
3  * Licensed to the Apache Software Foundation (ASF) under one
4  * or more contributor license agreements.  See the NOTICE file
5  * distributed with this work for additional information
6  * regarding copyright ownership.  The ASF licenses this file
7  * to you under the Apache License, Version 2.0 (the
8  * "License"); you may not use this file except in compliance
9  * with the License.  You may obtain a copy of the License at
10  *
11  *   http://www.apache.org/licenses/LICENSE-2.0
12  *
13  * Unless required by applicable law or agreed to in writing,
14  * software distributed under the License is distributed on an
15  * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
16  * KIND, either express or implied.  See the License for the
17  * specific language governing permissions and limitations
18  * under the License.
19  *
20  *************************************************************/
21 
22 
23 
24 package api.i18n;
25 
26 import com.sun.star.i18n.KParseTokens;
27 import com.sun.star.i18n.KParseType;
28 import com.sun.star.i18n.ParseResult;
29 import com.sun.star.i18n.XCharacterClassification;
30 import com.sun.star.lang.Locale;
31 import com.sun.star.uno.UnoRuntime;
32 import com.sun.star.uno.XComponentContext;
33 import org.junit.After;
34 import org.junit.AfterClass;
35 import org.junit.Before;
36 import org.junit.BeforeClass;
37 import org.junit.Assert;
38 import org.junit.Ignore;
39 import org.junit.Test;
40 import org.openoffice.test.uno.UnoApp;
41 
42 /**
43  * Testing <code>com.sun.star.i18n.XCharacterClassification</code>
44  * interface methods:
45  * <ul>
46  *  <li><code> toUpper() </code></li>
47  *  <li><code> toLower() </code></li>
48  *  <li><code> toTitle() </code></li>
49  *  <li><code> getType() </code></li>
50  *  <li><code> getCharacterType() </code></li>
51  *  <li><code> getStringType() </code></li>
52  *  <li><code> getCharacterDirection() </code></li>
53  *  <li><code> getScript() </code></li>
54  *  <li><code> parseAnyToken() </code></li>
55  *  <li><code> parsePredefinedToken() </code></li>
56  * </ul><p>
57  * Test is <b> NOT </b> multithread compliant. <p>
58  * @see com.sun.star.i18n.XCharacterClassification
59  */
60 public class XCharacterClassificationTest {
61     private static final UnoApp app = new UnoApp();
62 
63     private XComponentContext xContext = null;
64     public XCharacterClassification oObj = null;
65     public String[] languages = new String[]{"de","en","es","fr","ja","ko","zh"};
66     public String[] countries = new String[]{"DE","US","ES","FR","JP","KR","CN"};
67 
68     public String[] charstyles_java = new String[] {"UNASSIGNED","UPPERCASE_LETTER",
69         "LOWERCASE_LETTER","TITLECASE_LETTER","MODIFIER_LETTER","OTHER_LETTER",
70         "NON_SPACING_MARK","ENCLOSING_MARK","COMBINING_SPACING_MARK",
71         "DECIMAL_DIGIT_NUMBER","LETTER_NUMBER","OTHER_NUMBER","SPACE_SEPARATOR",
72         "LINE_SEPARATOR","PARAGRAPH_SEPARATOR","CONTROL","FORMAT","none17",
73         "PRIVATE_USE","none19","DASH_PUNCTUATION","START_PUNCTUATION","END_PUNCTUATION",
74         "CONNECTOR_PUNCTUATION","OTHER_PUNCTUATION","MATH_SYMBOL","CURRENCY_SYMBOL",
75         "MODIFIER_SYMBOL","OTHER_SYMBOL"};
76 
77     public String[] charstyles_office = new String[] {"UNASSIGNED","UPPERCASE_LETTER",
78         "LOWERCASE_LETTER","TITLECASE_LETTER","MODIFIER_LETTER","OTHER_LETTER",
79         "NON_SPACING_MARK","ENCLOSING_MARK","COMBINING_SPACING_MARK",
80         "DECIMAL_DIGIT_NUMBER","LETTER_NUMBER","OTHER_NUMBER","SPACE_SEPARATOR",
81         "LINE_SEPARATOR","PARAGRAPH_SEPARATOR","CONTROL","FORMAT","PRIVATE_USE",
82         "OTHER_PUNCTUATION","DASH_PUNCTUATION","START_PUNCTUATION","END_PUNCTUATION",
83         "CONNECTOR_PUNCTUATION",
84         "OTHER_PUNCTUATION","MATH_SYMBOL","CURRENCY_SYMBOL","MODIFIER_SYMBOL",
85         "OTHER_SYMBOL","INITIAL_PUNCTUATION","FINAL_PUNCTUATION","GENERAL_TYPES_COUNT"};
86 
87     public String[] unicode_script = new String[] {"U_BASIC_LATIN","U_LATIN_1_SUPPLEMENT",
88         "U_LATIN_EXTENDED_A","U_LATIN_EXTENDED_B","U_IPA_EXTENSIONS","U_SPACING_MODIFIER_LETTERS",
89         "U_COMBINING_DIACRITICAL_MARKS","U_GREEK","U_CYRILLIC","U_ARMENIAN","U_HEBREW",
90         "U_ARABIC","U_SYRIAC","U_THAANA","U_DEVANAGARI","U_BENGALI","U_GURMUKHI",
91         "U_GUJARATI","U_ORIYA","U_TAMIL","U_TELUGU","U_KANNADA","U_MALAYALAM",
92         "U_SINHALA","U_THAI","U_LAO","U_TIBETAN","U_MYANMAR","U_GEORGIAN",
93         "U_HANGUL_JAMO","U_ETHIOPIC","U_CHEROKEE","U_UNIFIED_CANADIAN_ABORIGINAL_SYLLABICS",
94         "U_OGHAM","U_RUNIC","U_KHMER","U_MONGOLIAN","U_LATIN_EXTENDED_ADDITIONAL",
95         "U_GREEK_EXTENDED","U_GENERAL_PUNCTUATION","U_SUPERSCRIPTS_AND_SUBSCRIPTS",
96         "U_CURRENCY_SYMBOLS","U_COMBINING_MARKS_FOR_SYMBOLS","U_LETTERLIKE_SYMBOLS",
97         "U_NUMBER_FORMS","U_ARROWS","U_MATHEMATICAL_OPERATORS","U_MISCELLANEOUS_TECHNICAL",
98         "U_CONTROL_PICTURES","U_OPTICAL_CHARACTER_RECOGNITION","U_ENCLOSED_ALPHANUMERICS",
99         "U_BOX_DRAWING","U_BLOCK_ELEMENTS","U_GEOMETRIC_SHAPES","U_MISCELLANEOUS_SYMBOLS",
100         "U_DINGBATS","U_BRAILLE_PATTERNS","U_CJK_RADICALS_SUPPLEMENT","U_KANGXI_RADICALS",
101         "U_IDEOGRAPHIC_DESCRIPTION_CHARACTERS","U_CJK_SYMBOLS_AND_PUNCTUATION",
102         "U_HIRAGANA","U_KATAKANA","U_BOPOMOFO","U_HANGUL_COMPATIBILITY_JAMO","U_KANBUN",
103         "U_BOPOMOFO_EXTENDED","U_ENCLOSED_CJK_LETTERS_AND_MONTHS","U_CJK_COMPATIBILITY",
104         "U_CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A","U_CJK_UNIFIED_IDEOGRAPHS","U_YI_SYLLABLES",
105         "U_YI_RADICALS","U_HANGUL_SYLLABLES","U_HIGH_SURROGATES","U_HIGH_PRIVATE_USE_SURROGATES",
106         "U_LOW_SURROGATES","U_PRIVATE_USE_AREA","U_CJK_COMPATIBILITY_IDEOGRAPHS",
107         "U_ALPHABETIC_PRESENTATION_FORMS","U_ARABIC_PRESENTATION_FORMS_A","U_COMBINING_HALF_MARKS",
108         "U_CJK_COMPATIBILITY_FORMS","U_SMALL_FORM_VARIANTS","U_ARABIC_PRESENTATION_FORMS_B",
109         "U_SPECIALS","U_HALFWIDTH_AND_FULLWIDTH_FORMS","U_CHAR_SCRIPT_COUNT","U_NO_SCRIPT"};
110 
111     // setup and close connections
112     @BeforeClass
113     public static void setUpConnection() throws Exception
114     {
115         app.start();
116     }
117 
118     @AfterClass
119     public static void tearDownConnection() throws InterruptedException, com.sun.star.uno.Exception
120     {
121         app.close();
122     }
123 
124     @Before
125     public void before() throws Exception {
126         xContext = app.getComponentContext();
127         oObj = UnoRuntime.queryInterface(
128             XCharacterClassification.class,
129             xContext.getServiceManager().createInstanceWithContext("com.sun.star.i18n.CharacterClassification", xContext)
130         );
131     }
132 
133     /**
134     * Test calls the method for different locales. Then each result is compared
135     * with a string, converted to a upper case using
136     * <code>java.lang.String</code> method <code>toUpperCase()</code>.<p>
137     * Has <b> OK </b> status if string, returned by the method is equal to
138     * a string that is returned by String.toUpperCase() for all locales.
139     */
140     @Test
141     public void _toUpper() {
142         boolean res = true;
143         char[] characters = new char[]{586,65,97,498,721,4588,772,8413,3404};
144         String toCheck = new String(characters);
145         String get = "";
146         String exp = "";
147 
148         for (int i=0;i<7;i++) {
149             get = oObj.toUpper(toCheck, 0, toCheck.length(), getLocale(i));
150             exp = toCheck.toUpperCase(
151                 new java.util.Locale(languages[i], countries[i]));
152             res &= get.equals(exp);
153             if (!res) {
154                 System.out.println("FAILED for: language=" + languages[i] +
155                     " ; country=" + countries[i]);
156                 System.out.println("Expected: " + exp);
157                 System.out.println("Gained : " + get);
158             }
159         }
160         Assert.assertTrue("toUpper()", res);
161     }
162 
163     /**
164     * Test calls the method for different locales. Then each result is compared
165     * with a string, converted to a lower case using
166     * <code>java.lang.String</code> method <code>toLowerCase()</code>.<p>
167     * Has <b> OK </b> status if string, returned by the method is equal to
168     * a string that is returned by String.toLowerCase() for all locales.
169     */
170     @Test
171     @Ignore("Bug #87590 Incorrect casing for U+026A, U+0268, U+0197...")
172     public void _toLower() {
173         boolean res = true;
174         char[] characters = new char[]{586,65,97,498,721,4588,772,8413,3404};
175         String toCheck = new String(characters);
176         String get = "";
177         String exp = "";
178 
179         for (int i=0;i<7;i++) {
180             get = oObj.toLower(toCheck,0,toCheck.length(),getLocale(i));
181             exp = toCheck.toLowerCase(
182                 new java.util.Locale(languages[i],countries[i]));
183             res &= get.equals(exp);
184             if (!res) {
185                 System.out.println("FAILED for: language=" + languages[i]
186                     + " ; country=" + countries[i]);
187                 System.out.println("Expected: " + exp);
188                 System.out.println("Gained : " + get);
189             }
190         }
191         Assert.assertTrue("toLower()", res);
192     }
193 
194     /**
195     * Test calls the method for different locales. Then each result is compared
196     * with a string, converted to a title case using
197     * <code>java.lang.Character</code> method <code>toTitleCase()</code>.<p>
198     * Has <b> OK </b> status if string, returned by the method is equal to
199     * a string that was converted using Character.toTitleCase() for all locales.
200     */
201     @Test
202     public void _toTitle() {
203         boolean res = true;
204         String toCheck = new String(new char[]{8112});
205         String get = "";
206         String exp = "";
207 
208         for (int i=0;i<7;i++) {
209             get = oObj.toTitle(toCheck, 0, 1, getLocale(i));
210             exp = new String(
211                 new char[]{Character.toTitleCase(toCheck.toCharArray()[0])});
212             res &= get.equals(exp);
213             if (!res) {
214                 System.out.println("FAILED for: language=" + languages[i]
215                     + " ; country=" + countries[i]);
216                 System.out.println("Expected: " + exp);
217                 System.out.println("Gained : " + get);
218             }
219         }
220         Assert.assertTrue("toTitle()", res);
221     }
222 
223     /**
224     * At first we define <code>int[]</code> and <code>char[]</code> arrays of
225     * unicode symbol numbers, arranged as sequences, where symbols are sorted
226     * by type, so the character of <code>i<sup><small>th</small></sup></code>
227     * type is located on <code>i<sup><small>th</small></sup></code> position.<p>
228     * Has <b> OK </b> status if for all 30 types the method returns value, that
229     * is equal to an element number.<p>
230     * @see com.sun.star.i18n.CharType
231     */
232     @Test
233     public void _getType() {
234         boolean res = true;
235         // If or when this list gets out of date again,
236         // find new characters of the required type with:
237         // for (int i = 30; i <= 0xffff; i++) {
238         //     String s = new String("" + (char)i);
239         //     int type = oObj.getType(s, 0);
240         //     System.out.println("character " + i + " has type " + charstyles_office[type]);
241         // }
242         char[] characters = new char[]{888,65,97,498,721,4588,772,8413,3404,
243             48,8544,179,32,8232,8233,144,8204,57344,56320,45,40,41,95,3852,247,
244             3647,901,3896,171,187};
245         String toCheck = new String(characters);
246 
247         for (int i=0;i<characters.length;i++) {
248             int get = oObj.getType(toCheck, i);
249             res &= (charstyles_office[get] == charstyles_office[i]);
250             if (!res) {
251                 System.out.println("Code: " + Integer.toHexString((int)characters[i]));
252                 System.out.println("Gained: " + charstyles_office[get]);
253                 System.out.println("Expected: " + charstyles_office[i]);
254             }
255         }
256         Assert.assertTrue("getType()", res);
257     }
258 
259     /**
260     * After defining string to be checked and array of expected types, test
261     * calls the method for each character of a string and for all locales.<p>
262     * Has <b> OK </b> status if the method returns type, expected for a given
263     * character and locale.
264     */
265     @Test
266     public void _getCharacterType() {
267         boolean res = true;
268         // If or when this list gets out of date again,
269         // find new characters of the required type with:
270         //for (int i = 32; i <= 0xffff; i++) {
271         //    int charType = oObj.getCharacterType("" + (char)i, 0, getLocale(1));
272         //    if (charType == 32) {
273         //        System.out.println("character " + i + " has type 32");
274         //    }
275         //}
276         String toCheck = "Ab0*";
277         int[] expected = new int[]{226,228,97,32};
278 
279         for (int i=0;i<toCheck.length();i++) {
280             for (int j=1;j<7;j++) {
281                 int get = oObj.getCharacterType(toCheck, i, getLocale(j));
282                 res &= (get == expected[i]);
283                 if (!res) {
284                     System.out.println("FAILED for: language=" + languages[j] +
285                         " ; country=" + countries[j]);
286                     System.out.println("Symbol: " + toCheck.toCharArray()[i]);
287                     System.out.println("Gained: " + get);
288                     System.out.println("Expected: " + expected[i]);
289                 }
290             }
291         }
292         Assert.assertTrue("getCharacterType()", res);
293     }
294 
295     /**
296     * After defining array of strings to be checked and array of expected types,
297     * test calls the method for each string of an array and for all locales.<p>
298     * Has <b> OK </b> status if the method returns type, expected for a given
299     * string and locale.
300     */
301     @Test
302     public void _getStringType() {
303         boolean res = true;
304         String[] toCheck = new String[]{"01234","AAAAA","bbbbb","AA()bb"};
305         int[] exp = new int[]{97,226,228,230};
306 
307         for (int j=0;j<toCheck.length;j++) {
308             for (int i=0;i<7;i++) {
309                 int get = oObj.getStringType(toCheck[j], 0,
310                     toCheck[j].length(), getLocale(i));
311                 res &= (get == exp[j]);
312                 if (!res) {
313                     System.out.println("FAILED for: language=" + languages[i] +
314                         " ; country=" + countries[i]);
315                     System.out.println("Expected: " + exp[j]);
316                     System.out.println("Gained : " + get);
317                 }
318             }
319         }
320         Assert.assertTrue("getStringType()", res);
321     }
322 
323     /**
324     * After string to be checked is initialized (all symbols are sorted
325     * by direction, so the character of <code>i<sup><small>th</small></sup></code>
326     * direction is located on <code>i<sup><small>th</small></sup></code>
327     * position), test calls the method for every character of that string. <p>
328     * Has <b> OK </b> status if the method returns direction, that's equal to
329     * a symbol position in the string.
330     */
331     @Test
332     public void _getCharacterDirection() {
333         boolean res = true;
334         String toCheck = new String(new char[]{65,1470,48,43,35,1632,44,10,
335                                 9,12,33,8234,8237,1563,8235,8238,8236,768,1});
336         for (short i=0;i<19;i++) {
337             short get = oObj.getCharacterDirection(toCheck, i);
338             res &= (get == i);
339             if (!res) {
340                 System.out.println("Code :" + toCheck.toCharArray()[i]);
341                 System.out.println("Gained: " + get);
342                 System.out.println("Expected: " + i);
343             }
344         }
345         Assert.assertTrue("getCharacterDirection()", res);
346     }
347 
348     /**
349     * At first we define <code>int[]</code> and <code>char[]</code> arrays of
350     * unicode symbol numbers, arranged as sequences, where symbols are sorted
351     * by type, so the character of <code>i<sup><small>th</small></sup></code>
352     * type is located on <code>i<sup><small>th</small></sup></code> position.<p>
353     * Has <b> OK </b> status if for each character method returns value, that
354     * is equal to a number where element is located in array. Also method has
355     * <b> OK </b> status for symbol with code 55296, because it doesn't work
356     * since it hasn't the right neighborhood.<p>
357     * @see "http://ppewww.ph.gla.ac.uk/~flavell/unicode/unidata.html"
358     */
359     @Test
360     public void _getScript() {
361         boolean res = true;
362         char[] characters = new char[]{65,128,256,384,592,750,773,924,1030,1331,1448,
363             1569,1792,1936,2313,2465,2570,2707,2822,2972,3079,3240,3337,3464,3590,
364             3745,3906,4097,4274,4357,4621,5040,5200,5776,5806,6030,6155,7683,7943,
365             8202,8319,8352,8413,8452,8545,8616,8715,8965,9217,9281,9336,9474,9608,9719,
366             9734,9999,10247,11911,12034,12274,12294,12358,12456,12552,12605,12688,12727,
367             12806,13065,13312,19968,40964,42152,44032,55296,56192,56320,57344,63744,
368             64257,64370,65056,65073,65131,65146,65532,65288};
369         String toCheck = new String(characters);
370 
371         for (int i=0;i<characters.length;i++) {
372             int get = oObj.getScript(toCheck, i);
373             //The HIGH_SURROGATE 55296 doesn't work since it hasn't the right
374             //neighborhood. Neither does 56192 - it combines with 55296 into another code point.
375             if (toCheck.charAt(i) != 55296 && toCheck.charAt(i) != 56192) {
376                 res &= (get == i);
377             }
378             if (!res) {
379                 System.out.println("-- " + toCheck.substring(i, i + 1).hashCode());
380                 System.out.println("Code: " + Integer.toHexString((int)characters[i]));
381                 System.out.println("Gained: " + get + " (" + (0 <= get && get < unicode_script.length ? unicode_script[get] : "out of range") + ")");
382                 System.out.println("Expected: " + unicode_script[i]);
383             }
384         }
385         Assert.assertTrue("getScript()", res);
386     }
387 
388     /**
389     * After defining a string to be parsed and parse conditions (flags), test
390     * calls the method for different locales three times with different parameters,
391     * checking result after every call.  <p>
392     * Has <b> OK </b> status if the method returns right results all three
393     * times.
394     */
395     @Test
396     public void _parseAnyToken() {
397         int nStartFlags = KParseTokens.ANY_ALPHA | KParseTokens.ASC_UNDERSCORE;
398         int nContFlags = KParseTokens.ANY_ALNUM | KParseTokens.ASC_UNDERSCORE
399                         | KParseTokens.ASC_DOT;
400         String toCheck = " 18 i18n ^";
401         ParseResult pRes = null;
402         boolean res = true;
403 
404         for (int i=0;i<7;i++) {
405             pRes = oObj.parseAnyToken(toCheck, 1, getLocale(i),
406                 nStartFlags, "", nContFlags, "");
407             res = ( (pRes.CharLen==2)
408                  && (pRes.TokenType==32)
409                  && (pRes.Value==18.0) );
410             pRes = oObj.parseAnyToken(toCheck, 4, getLocale(i),
411                 nStartFlags, "", nContFlags, "");
412             res &= ( (pRes.CharLen==4)
413                   && (pRes.TokenType==4)
414                   && (pRes.Value==0.0) );
415             pRes = oObj.parseAnyToken(toCheck, 9, getLocale(i),
416                 nStartFlags, "", nContFlags, "");
417             res &= ( (pRes.CharLen==1)
418                   && (pRes.TokenType==1)
419                   && (pRes.Value==0.0) );
420         }
421         Assert.assertTrue("parseAnyToken()", res);
422     }
423 
424     /**
425     * After defining a string to be parsed and parse conditions (flags), test
426     * calls the method for different locales two times with different parameters,
427     * checking result after every call. <p>
428     * Has <b> OK </b> status if the method returns right results.
429     */
430     @Test
431     public void _parsePredefinedToken() {
432         int nStartFlags = KParseTokens.ANY_ALPHA | KParseTokens.ASC_UNDERSCORE;
433         int nContFlags = nStartFlags;
434         String toCheck = " 18 int";
435         ParseResult pRes = null;
436         boolean res = true;
437 
438         for (int i=0;i<7;i++) {
439             pRes = oObj.parsePredefinedToken(KParseType.IDENTNAME, toCheck,
440                 1, getLocale(i), nStartFlags, "", nContFlags, "");
441             res = (pRes.CharLen==0);
442             pRes = oObj.parsePredefinedToken(KParseType.IDENTNAME, toCheck,
443                 4, getLocale(i), nStartFlags, "", nContFlags, "");
444             res &= ( (pRes.CharLen==3)
445                   && (pRes.TokenType==4)
446                   && (pRes.Value==0.0) );
447         }
448         Assert.assertTrue("parsePredefinedToken()", res);
449     }
450 
451 
452     /**
453     * Method returns locale for a given language and country.
454     * @param k index of needed locale.
455     */
456     private Locale getLocale(int k) {
457         return new Locale(languages[k],countries[k],"");
458     }
459 
460 
461 } // end XCharacterClassification
462