Java Code Examples for com.ibm.icu.lang.UCharacter#getIntPropertyValue()

/**
 * Check codePoint is FullWidth or not according to Unicode Standard version 12.0.0.
 * See http://unicode.org/reports/tr11/
 */
public static boolean isFullWidth(int codePoint) {
	int value = UCharacter.getIntPropertyValue(codePoint, UProperty.EAST_ASIAN_WIDTH);
	switch (value) {
		case UCharacter.EastAsianWidth.NEUTRAL:
			return false;
		case UCharacter.EastAsianWidth.AMBIGUOUS:
			return false;
		case UCharacter.EastAsianWidth.HALFWIDTH:
			return false;
		case UCharacter.EastAsianWidth.FULLWIDTH:
			return true;
		case UCharacter.EastAsianWidth.NARROW:
			return false;
		case UCharacter.EastAsianWidth.WIDE:
			return true;
		default:
			throw new RuntimeException("unknown UProperty.EAST_ASIAN_WIDTH: " + value);
	}
}
 

public boolean handles(int c, int breakType) {
    if (breakType == BreakIterator.KIND_WORD || breakType == BreakIterator.KIND_LINE) {
        int script = UCharacter.getIntPropertyValue(c, UProperty.SCRIPT);
        return (script == UScript.KHMER);
    }
    return false;
}
 

@Override
public boolean handles(int c, int breakType) {
    if (breakType == BreakIterator.KIND_WORD || breakType == BreakIterator.KIND_LINE) {
        int script = UCharacter.getIntPropertyValue(c, UProperty.SCRIPT);
        return (script == UScript.MYANMAR);
    }
    return false;
}
 

public synchronized void handleChar(int c, int breakType) {
    if (breakType >= 0 && breakType < fHandled.length && c != DONE32) {
        if (!fHandled[breakType].contains(c)) {
            int script = UCharacter.getIntPropertyValue(c, UProperty.SCRIPT);
            fHandled[breakType].applyIntPropertyValue(UProperty.SCRIPT, script);
        }
    }
}
 

public boolean handles(int c, int breakType) {
    if (breakType == BreakIterator.KIND_WORD || breakType == BreakIterator.KIND_LINE) {
        int script = UCharacter.getIntPropertyValue(c, UProperty.SCRIPT);
        return (script == UScript.LAO);
    }
    return false;
}
 

public boolean handles(int c, int breakType) {
    if (breakType == BreakIterator.KIND_WORD || breakType == BreakIterator.KIND_LINE) {
        int script = UCharacter.getIntPropertyValue(c, UProperty.SCRIPT);
        return (script == UScript.THAI);
    }
    return false;
}
 

private static UnicodeSet getIntPropInclusions(int prop) {
    assert(UProperty.INT_START <= prop && prop < UProperty.INT_LIMIT);
    int inclIndex = UCharacterProperty.SRC_COUNT + prop - UProperty.INT_START;
    if (inclusions[inclIndex] != null) {
        return inclusions[inclIndex];
    }
    int src = UCharacterProperty.INSTANCE.getSource(prop);
    UnicodeSet incl = getInclusionsForSource(src);

    UnicodeSet intPropIncl = new UnicodeSet(0, 0);
    int numRanges = incl.getRangeCount();
    int prevValue = 0;
    for (int i = 0; i < numRanges; ++i) {
        int rangeEnd = incl.getRangeEnd(i);
        for (int c = incl.getRangeStart(i); c <= rangeEnd; ++c) {
            // TODO: Get a UCharacterProperty.IntProperty to avoid the property dispatch.
            int value = UCharacter.getIntPropertyValue(c, prop);
            if (value != prevValue) {
                intPropIncl.add(c);
                prevValue = value;
            }
        }
    }

    // Compact for caching.
    return inclusions[inclIndex] = intPropIncl.compact();
}
 

void calcChainedFollowPos(RBBINode tree) {

           List<RBBINode> endMarkerNodes = new ArrayList<RBBINode>();
           List<RBBINode> leafNodes      = new ArrayList<RBBINode>();

            // get a list of all endmarker nodes.
           tree.findNodes(endMarkerNodes, RBBINode.endMark);

           // get a list all leaf nodes
           tree.findNodes(leafNodes, RBBINode.leafChar);

           // Collect all leaf nodes that can start matches for rules
           // with inbound chaining enabled, which is the union of the
           // firstPosition sets from each of the rule root nodes.

           List<RBBINode> ruleRootNodes = new ArrayList<RBBINode>();
           addRuleRootNodes(ruleRootNodes, tree);

           Set<RBBINode> matchStartNodes = new HashSet<RBBINode>();
           for (RBBINode node: ruleRootNodes) {
               if (node.fChainIn) {
                   matchStartNodes.addAll(node.fFirstPosSet);
               }
           }

           // Iterate over all leaf nodes,
           //
           for (RBBINode tNode : leafNodes) {
               RBBINode endNode = null;

               // Identify leaf nodes that correspond to overall rule match positions.
               //   These include an endMarkerNode in their followPos sets.
               for (RBBINode endMarkerNode : endMarkerNodes) {
                   if (tNode.fFollowPos.contains(endMarkerNode)) {
                       endNode = tNode;
                       break;
                   }
               }
               if (endNode == null) {
                   // node wasn't an end node.  Try again with the next.
                   continue;
               }

               // We've got a node that can end a match.

               // Line Break Specific hack:  If this node's val correspond to the $CM char class,
               //                            don't chain from it.
               // TODO:  Add rule syntax for this behavior, get specifics out of here and
               //        into the rule file.
               if (fRB.fLBCMNoChain) {
                   int c = this.fRB.fSetBuilder.getFirstChar(endNode.fVal);
                   if (c != -1) {
                       // c == -1 occurs with sets containing only the {eof} marker string.
                       int cLBProp = UCharacter.getIntPropertyValue(c, UProperty.LINE_BREAK);
                       if (cLBProp == UCharacter.LineBreak.COMBINING_MARK) {
                           continue;
                       }
                   }
               }


               // Now iterate over the nodes that can start a match, looking for ones
               //   with the same char class as our ending node.
               for (RBBINode startNode : matchStartNodes) {
                   if (startNode.fType != RBBINode.leafChar) {
                       continue;
                   }

                   if (endNode.fVal == startNode.fVal) {
                       // The end val (character class) of one possible match is the
                       //   same as the start of another.

                       // Add all nodes from the followPos of the start node to the
                       //  followPos set of the end node, which will have the effect of
                       //  letting matches transition from a match state at endNode
                       //  to the second char of a match starting with startNode.
                       endNode.fFollowPos.addAll(startNode.fFollowPos);
                   }
               }
           }
       }
 

@Override
public boolean contains(int ch) {
    return UCharacter.getIntPropertyValue(ch, prop) == value;
}
 

private LanguageBreakEngine getLanguageBreakEngine(int c) {

        // We have a dictionary character.
        // Does an already instantiated break engine handle it?
        for (LanguageBreakEngine candidate : fBreakEngines.values()) {
            if (candidate.handles(c, fBreakType)) {
                return candidate;
            }
        }

        // if we don't have an existing engine, build one.
        int script = UCharacter.getIntPropertyValue(c, UProperty.SCRIPT);
        if (script == UScript.KATAKANA || script == UScript.HIRAGANA) {
            // Katakana, Hiragana and Han are handled by the same dictionary engine.
            // Fold them together for mapping from script -> engine.
            script = UScript.HAN;
        }

        LanguageBreakEngine eng = fBreakEngines.get(script);
        /*
        if (eng != null && !eng.handles(c, fBreakType)) {
            fUnhandledBreakEngine.handleChar(c, getBreakType());
            eng = fUnhandledBreakEngine;
        } else  */  {
            try {
                switch (script) {
                case UScript.THAI:
                    eng = new ThaiBreakEngine();
                    break;
                case UScript.LAO:
                    eng = new LaoBreakEngine();
                    break;
                case UScript.MYANMAR:
                    eng = new BurmeseBreakEngine();
                    break;
                case UScript.KHMER:
                    eng = new KhmerBreakEngine();
                    break;
                case UScript.HAN:
                    if (getBreakType() == KIND_WORD) {
                        eng = new CjkBreakEngine(false);
                    }
                    else {
                        fUnhandledBreakEngine.handleChar(c, getBreakType());
                        eng = fUnhandledBreakEngine;
                    }
                    break;
                case UScript.HANGUL:
                    if (getBreakType() == KIND_WORD) {
                        eng = new CjkBreakEngine(true);
                    } else {
                        fUnhandledBreakEngine.handleChar(c, getBreakType());
                        eng = fUnhandledBreakEngine;
                    }
                    break;
                default:
                    fUnhandledBreakEngine.handleChar(c, getBreakType());
                    eng = fUnhandledBreakEngine;
                    break;
                }
            } catch (IOException e) {
                eng = null;
            }
        }

        if (eng != null && eng != fUnhandledBreakEngine) {
            LanguageBreakEngine existingEngine = fBreakEngines.putIfAbsent(script, eng);
            if (existingEngine != null) {
                // There was a race & another thread was first to register an engine for this script.
                // Use theirs and discard the one we just created.
                eng = existingEngine;
            }
            // assert eng.handles(c, fBreakType);
        }
        return eng;
    }
 

@Override
public boolean contains(int ch) {
    return UCharacter.getIntPropertyValue(ch, prop) == value;
}
 

public boolean has(int codePoint, int value) {
    return UCharacter.getIntPropertyValue(codePoint, propertyId) == value;
}