]>
Commit | Line | Data |
---|---|---|
b75a7d8f | 1 | /* |
374ca955 | 2 | ********************************************************************** |
51004dcb | 3 | * Copyright (c) 2001-2012, International Business Machines |
374ca955 | 4 | * Corporation and others. All Rights Reserved. |
b75a7d8f A |
5 | ********************************************************************** |
6 | * Date Name Description | |
7 | * 07/18/01 aliu Creation. | |
8 | ********************************************************************** | |
9 | */ | |
10 | ||
11 | #include "unicode/unifilt.h" | |
12 | #include "unicode/rep.h" | |
4388f060 | 13 | #include "unicode/utf16.h" |
b75a7d8f A |
14 | |
15 | U_NAMESPACE_BEGIN | |
374ca955 A |
16 | UOBJECT_DEFINE_ABSTRACT_RTTI_IMPLEMENTATION(UnicodeFilter) |
17 | ||
b75a7d8f | 18 | |
374ca955 A |
19 | /* Define this here due to the lack of another file. |
20 | It can't be defined in the header */ | |
21 | UnicodeMatcher::~UnicodeMatcher() {} | |
22 | ||
23 | UnicodeFilter::~UnicodeFilter() {} | |
b75a7d8f A |
24 | |
25 | /** | |
51004dcb A |
26 | * UnicodeFunctor API. |
27 | * Note that UnicodeMatcher is a base class of UnicodeFilter. | |
b75a7d8f A |
28 | */ |
29 | UnicodeMatcher* UnicodeFilter::toMatcher() const { | |
51004dcb | 30 | return const_cast<UnicodeFilter *>(this); |
b75a7d8f A |
31 | } |
32 | ||
374ca955 A |
33 | void UnicodeFilter::setData(const TransliterationRuleData*) {} |
34 | ||
b75a7d8f A |
35 | /** |
36 | * Default implementation of UnicodeMatcher::matches() for Unicode | |
37 | * filters. Matches a single code point at offset (either one or | |
38 | * two 16-bit code units). | |
39 | */ | |
40 | UMatchDegree UnicodeFilter::matches(const Replaceable& text, | |
41 | int32_t& offset, | |
42 | int32_t limit, | |
43 | UBool incremental) { | |
44 | UChar32 c; | |
45 | if (offset < limit && | |
46 | contains(c = text.char32At(offset))) { | |
4388f060 | 47 | offset += U16_LENGTH(c); |
b75a7d8f A |
48 | return U_MATCH; |
49 | } | |
50 | if (offset > limit && | |
51 | contains(c = text.char32At(offset))) { | |
52 | // Backup offset by 1, unless the preceding character is a | |
53 | // surrogate pair -- then backup by 2 (keep offset pointing at | |
54 | // the lead surrogate). | |
55 | --offset; | |
56 | if (offset >= 0) { | |
4388f060 | 57 | offset -= U16_LENGTH(text.char32At(offset)) - 1; |
b75a7d8f A |
58 | } |
59 | return U_MATCH; | |
60 | } | |
61 | if (incremental && offset == limit) { | |
62 | return U_PARTIAL_MATCH; | |
63 | } | |
64 | return U_MISMATCH; | |
65 | } | |
66 | ||
67 | U_NAMESPACE_END | |
68 | ||
69 | //eof |