Examples with UnicodeSet android.icu.text.UnicodeSet used on opensource projects

Example 86 with UnicodeSet

use of android.icu.text.UnicodeSet in project j2objc by google.

the class CollationMiscTest method TestImport.

@Test
public void TestImport() {
    try {
        RuleBasedCollator vicoll = (RuleBasedCollator) Collator.getInstance(new ULocale("vi"));
        RuleBasedCollator escoll = (RuleBasedCollator) Collator.getInstance(new ULocale("es"));
        RuleBasedCollator viescoll = new RuleBasedCollator(vicoll.getRules() + escoll.getRules());
        RuleBasedCollator importviescoll = new RuleBasedCollator("[import vi][import es]");
        UnicodeSet tailoredSet = viescoll.getTailoredSet();
        UnicodeSet importTailoredSet = importviescoll.getTailoredSet();
        if (!tailoredSet.equals(importTailoredSet)) {
            warnln("Tailored set not equal");
        }
        for (UnicodeSetIterator it = new UnicodeSetIterator(tailoredSet); it.next(); ) {
            String t = it.getString();
            CollationKey sk1 = viescoll.getCollationKey(t);
            CollationKey sk2 = importviescoll.getCollationKey(t);
            if (!sk1.equals(sk2)) {
                warnln("Collation key's not equal for " + t);
            }
        }
    } catch (Exception e) {
        // Android patch: Add --omitCollationRules to genrb.
        logln("ERROR: in creation of rule based collator");
    // Android patch end.
    }
}

Example 87 with UnicodeSet

use of android.icu.text.UnicodeSet in project j2objc by google.

the class AlphabeticIndexTest method TestFirstCharacters.

@Test
public void TestFirstCharacters() {
    AlphabeticIndex alphabeticIndex = new AlphabeticIndex(Locale.ENGLISH);
    RuleBasedCollator collator = alphabeticIndex.getCollator();
    collator.setStrength(Collator.IDENTICAL);
    Collection<String> firsts = alphabeticIndex.getFirstCharactersInScripts();
    // Verify that each script is represented exactly once.
    // Exclude pseudo-scripts like Common (no letters).
    // Exclude scripts like Braille and Sutton SignWriting
    // because they only have symbols, not letters.
    UnicodeSet missingScripts = new UnicodeSet("[^[:inherited:][:unknown:][:common:][:Braille:][:SignWriting:]]");
    String last = "";
    for (String index : firsts) {
        if (collator.compare(last, index) >= 0) {
            errln("Characters not in order: " + last + " !< " + index);
        }
        int script = getFirstRealScript(index);
        if (script == UScript.UNKNOWN) {
            continue;
        }
        UnicodeSet s = new UnicodeSet().applyIntPropertyValue(UProperty.SCRIPT, script);
        if (missingScripts.containsNone(s)) {
            errln("2nd character in script: " + index + "\t" + new UnicodeSet(missingScripts).retainAll(s).toPattern(false));
        }
        missingScripts.removeAll(s);
    }
    if (missingScripts.size() != 0) {
        String missingScriptNames = "";
        UnicodeSet missingChars = new UnicodeSet(missingScripts);
        for (; ; ) {
            int c = missingChars.charAt(0);
            if (c < 0) {
                break;
            }
            int script = UScript.getScript(c);
            missingScriptNames += " " + UCharacter.getPropertyValueName(UProperty.SCRIPT, script, UProperty.NameChoice.SHORT);
            missingChars.removeAll(new UnicodeSet().applyIntPropertyValue(UProperty.SCRIPT, script));
        }
        errln("Missing character from:" + missingScriptNames + " -- " + missingScripts);
    }
}

Example 88 with UnicodeSet

use of android.icu.text.UnicodeSet in project j2objc by google.

the class UnicodeSetTest method TestComparison.

@Test
public void TestComparison() {
    UnicodeSet set1 = new UnicodeSet("[a-b d-g {ch} {zh}]").freeze();
    UnicodeSet set2 = new UnicodeSet("[c-e {ch}]").freeze();
    UnicodeSet set3 = new UnicodeSet("[d m-n z {dh}]").freeze();
    // compareTo(UnicodeSet)
    // do indirectly, by sorting
    List<UnicodeSet> unsorted = Arrays.asList(set3, set2, set1);
    List<UnicodeSet> goalShortest = Arrays.asList(set2, set3, set1);
    List<UnicodeSet> goalLongest = Arrays.asList(set1, set3, set2);
    List<UnicodeSet> goalLex = Arrays.asList(set1, set2, set3);
    List<UnicodeSet> sorted = new ArrayList(new TreeSet<UnicodeSet>(unsorted));
    assertNotEquals("compareTo-shorter-first", unsorted, sorted);
    assertEquals("compareTo-shorter-first", goalShortest, sorted);
    TreeSet<UnicodeSet> sorted1 = new TreeSet<UnicodeSet>(new Comparator<UnicodeSet>() {

        public int compare(UnicodeSet o1, UnicodeSet o2) {
            // TODO Auto-generated method stub
            return o1.compareTo(o2, ComparisonStyle.LONGER_FIRST);
        }
    });
    sorted1.addAll(unsorted);
    sorted = new ArrayList(sorted1);
    assertNotEquals("compareTo-longer-first", unsorted, sorted);
    assertEquals("compareTo-longer-first", goalLongest, sorted);
    sorted1 = new TreeSet<UnicodeSet>(new Comparator<UnicodeSet>() {

        public int compare(UnicodeSet o1, UnicodeSet o2) {
            // TODO Auto-generated method stub
            return o1.compareTo(o2, ComparisonStyle.LEXICOGRAPHIC);
        }
    });
    sorted1.addAll(unsorted);
    sorted = new ArrayList(sorted1);
    assertNotEquals("compareTo-lex", unsorted, sorted);
    assertEquals("compareTo-lex", goalLex, sorted);
    // compare(String, int)
    // make a list of interesting combinations
    List<String> sources = Arrays.asList("\u0000", "a", "b", "\uD7FF", "\uD800", "\uDBFF", "\uDC00", "\uDFFF", "\uE000", "\uFFFD", "\uFFFF");
    TreeSet<String> target = new TreeSet<String>();
    for (String s : sources) {
        target.add(s);
        for (String t : sources) {
            target.add(s + t);
            for (String u : sources) {
                target.add(s + t + u);
            }
        }
    }
    // now compare all the combinations. If any of them is a code point, use it.
    int maxErrorCount = 0;
    compare: for (String last : target) {
        for (String curr : target) {
            int lastCount = Character.codePointCount(last, 0, last.length());
            int currCount = Character.codePointCount(curr, 0, curr.length());
            int comparison;
            if (lastCount == 1) {
                comparison = UnicodeSet.compare(last.codePointAt(0), curr);
            } else if (currCount == 1) {
                comparison = UnicodeSet.compare(last, curr.codePointAt(0));
            } else {
                continue;
            }
            if (comparison != last.compareTo(curr)) {
                // repeat for debugging
                if (lastCount == 1) {
                    comparison = UnicodeSet.compare(last.codePointAt(0), curr);
                } else if (currCount == 1) {
                    comparison = UnicodeSet.compare(last, curr.codePointAt(0));
                }
                if (maxErrorCount++ > 10) {
                    errln(maxErrorCount + " Failure in comparing " + last + " & " + curr + "\tOmitting others...");
                    break compare;
                }
                errln(maxErrorCount + " Failure in comparing " + last + " & " + curr);
            }
        }
    }
    // compare(Iterable<T>, Iterable<T>)
    int max = 10;
    List<String> test1 = new ArrayList<String>(max);
    List<String> test2 = new ArrayList<String>(max);
    for (int i = 0; i <= max; ++i) {
        test1.add("a" + i);
        // add in reverse order
        test2.add("a" + (max - i));
    }
    assertNotEquals("compare iterable test", test1, test2);
    TreeSet<CharSequence> sortedTest1 = new TreeSet<CharSequence>(test1);
    TreeSet<CharSequence> sortedTest2 = new TreeSet<CharSequence>(test2);
    assertEquals("compare iterable test", sortedTest1, sortedTest2);
}

Example 89 with UnicodeSet

use of android.icu.text.UnicodeSet in project j2objc by google.

the class UnicodeSetTest method TestPropertyAccess.

@Test
public void TestPropertyAccess() {
    int count = 0;
    // test to see that all of the names work
    for (int propNum = UProperty.BINARY_START; propNum < UProperty.INT_LIMIT; ++propNum) {
        count++;
        // Skipping tests in the non-exhaustive mode to shorten the test time ticket#6475
        if (TestFmwk.getExhaustiveness() <= 5 && count % 5 != 0) {
            continue;
        }
        if (propNum >= UProperty.BINARY_LIMIT && propNum < UProperty.INT_START) {
            // skip the gap
            propNum = UProperty.INT_START;
        }
        for (int nameChoice = UProperty.NameChoice.SHORT; nameChoice <= UProperty.NameChoice.LONG; ++nameChoice) {
            String propName;
            try {
                propName = UCharacter.getPropertyName(propNum, nameChoice);
                if (propName == null) {
                    // allow non-existent short names
                    if (nameChoice == UProperty.NameChoice.SHORT)
                        continue;
                    throw new NullPointerException();
                }
            } catch (RuntimeException e1) {
                errln("Can't get property name for: " + "Property (" + propNum + ")" + ", NameChoice: " + nameChoice + ", " + e1.getClass().getName());
                continue;
            }
            logln("Property (" + propNum + "): " + propName);
            for (int valueNum = UCharacter.getIntPropertyMinValue(propNum); valueNum <= UCharacter.getIntPropertyMaxValue(propNum); ++valueNum) {
                String valueName;
                try {
                    valueName = UCharacter.getPropertyValueName(propNum, valueNum, nameChoice);
                    if (valueName == null) {
                        // allow non-existent short names
                        if (nameChoice == UProperty.NameChoice.SHORT)
                            continue;
                        if ((propNum == UProperty.CANONICAL_COMBINING_CLASS || propNum == UProperty.LEAD_CANONICAL_COMBINING_CLASS || propNum == UProperty.TRAIL_CANONICAL_COMBINING_CLASS) && !isCccValue(valueNum)) {
                            // Otherwise they are just integer values.
                            continue;
                        } else {
                            throw new NullPointerException();
                        }
                    }
                } catch (RuntimeException e1) {
                    errln("Can't get property value name for: " + "Property (" + propNum + "): " + propName + ", " + "Value (" + valueNum + ") " + ", NameChoice: " + nameChoice + ", " + e1.getClass().getName());
                    continue;
                }
                logln("Value (" + valueNum + "): " + valueName);
                UnicodeSet testSet;
                try {
                    testSet = new UnicodeSet("[:" + propName + "=" + valueName + ":]");
                } catch (RuntimeException e) {
                    errln("Can't create UnicodeSet for: " + "Property (" + propNum + "): " + propName + ", " + "Value (" + valueNum + "): " + valueName + ", " + e.getClass().getName());
                    continue;
                }
                UnicodeSet collectedErrors = new UnicodeSet();
                for (UnicodeSetIterator it = new UnicodeSetIterator(testSet); it.next(); ) {
                    int value = UCharacter.getIntPropertyValue(it.codepoint, propNum);
                    if (value != valueNum) {
                        collectedErrors.add(it.codepoint);
                    }
                }
                if (collectedErrors.size() != 0) {
                    errln("Property Value Differs: " + "Property (" + propNum + "): " + propName + ", " + "Value (" + valueNum + "): " + valueName + ", " + "Differing values: " + collectedErrors.toPattern(true));
                }
            }
        }
    }
}

Example 90 with UnicodeSet

use of android.icu.text.UnicodeSet in project j2objc by google.

the class UnicodeSetTest method testForSpanGaps.

@Test
public void testForSpanGaps() {
    String[] items = { "a", "b", "c", "{ab}", "{bc}", "{cd}", "{abc}", "{bcd}" };
    final int limit = 1 << items.length;
    // build long string for testing
    StringBuilder longBuffer = new StringBuilder();
    for (int i = 1; i < limit; ++i) {
        longBuffer.append("x");
        longBuffer.append(getCombinations(items, i));
    }
    String longString = longBuffer.toString();
    longString = longString.replace("{", "").replace("}", "");
    long start = System.nanoTime();
    for (int i = 1; i < limit; ++i) {
        UnicodeSet us = new UnicodeSet("[" + getCombinations(items, i) + "]");
        int problemFound = checkSpan(longString, us, SpanCondition.SIMPLE);
        if (problemFound >= 0) {
            assertEquals("Testing " + longString + ", found gap at", -1, problemFound);
            break;
        }
    }
    long end = System.nanoTime();
    logln("Time for SIMPLE   :\t" + (end - start));
    start = System.nanoTime();
    for (int i = 1; i < limit; ++i) {
        UnicodeSet us = new UnicodeSet("[" + getCombinations(items, i) + "]");
        int problemFound = checkSpan(longString, us, SpanCondition.CONTAINED);
        if (problemFound >= 0) {
            assertEquals("Testing " + longString + ", found gap at", -1, problemFound);
            break;
        }
    }
    end = System.nanoTime();
    logln("Time for CONTAINED:\t" + (end - start));
}