24
|
1 |
/*
|
|
2 |
* Copyright (c) 2010 Nokia Corporation and/or its subsidiary(-ies).
|
|
3 |
* All rights reserved.
|
|
4 |
* This component and the accompanying materials are made available
|
|
5 |
* under the terms of "Eclipse Public License v1.0"
|
|
6 |
* which accompanies this distribution, and is available
|
|
7 |
* at the URL "http://www.eclipse.org/legal/epl-v10.html".
|
|
8 |
*
|
|
9 |
* Initial Contributors:
|
|
10 |
* Nokia Corporation - initial contribution.
|
|
11 |
*
|
|
12 |
* Contributors:
|
|
13 |
*
|
|
14 |
* Description:
|
|
15 |
*
|
|
16 |
*/
|
|
17 |
namespace analysis {
|
|
18 |
|
|
19 |
template<class Encoding>
|
|
20 |
StateMachineBreakIterator<Encoding>::StateMachineBreakIterator(
|
|
21 |
StateMachine<Encoding>& machine)
|
|
22 |
: machine_( machine ),
|
|
23 |
state_(),
|
|
24 |
current_(-1),
|
|
25 |
next_(-1),
|
|
26 |
text_(0)
|
|
27 |
{
|
|
28 |
}
|
|
29 |
|
|
30 |
template<class Encoding>
|
|
31 |
StateMachineBreakIterator<Encoding>::~StateMachineBreakIterator() {}
|
|
32 |
|
|
33 |
template<class Encoding>
|
|
34 |
void StateMachineBreakIterator<Encoding>::setText(const wchar_t* text)
|
|
35 |
{
|
|
36 |
// Let's point to the begining of new text
|
|
37 |
text_ = text;
|
|
38 |
cursor_ = 0;
|
|
39 |
|
|
40 |
// First boundary is in the beginning of buffer
|
|
41 |
current_ = 0;
|
|
42 |
// We haven't searched for next boundary yet
|
|
43 |
next_ = -1;
|
|
44 |
}
|
|
45 |
|
|
46 |
template<class Encoding>
|
|
47 |
bool StateMachineBreakIterator<Encoding>::hasNext()
|
|
48 |
{
|
|
49 |
prepareNext();
|
|
50 |
return next_ != -1;
|
|
51 |
}
|
|
52 |
|
|
53 |
template<class Encoding>
|
|
54 |
int StateMachineBreakIterator<Encoding>::current()
|
|
55 |
{
|
|
56 |
return current_;
|
|
57 |
}
|
|
58 |
|
|
59 |
template<class Encoding>
|
|
60 |
int StateMachineBreakIterator<Encoding>::next()
|
|
61 |
{
|
|
62 |
prepareNext();
|
|
63 |
current_ = next_;
|
|
64 |
next_ = -1;
|
|
65 |
return current_;
|
|
66 |
}
|
|
67 |
|
|
68 |
template<class Encoding>
|
|
69 |
void StateMachineBreakIterator<Encoding>::prepareNext()
|
|
70 |
{
|
|
71 |
// Implements longest matching word algorithm. The used
|
|
72 |
// state machine contains an entire dictionary. Each state
|
|
73 |
// transition interprets as incremental search in dictionary.
|
|
74 |
// Each final (or terminal) state, marks location, where the
|
|
75 |
// consumed states form a valid word. We try to find the
|
|
76 |
// longest matching word.
|
|
77 |
//
|
|
78 |
|
|
79 |
// Prepare next_ only, if new next_ hasn't been prepared before
|
|
80 |
if (next_ == -1 && text_ && text_[cursor_]) {
|
|
81 |
// Reset state machine
|
|
82 |
machine_.rootState(state_);
|
|
83 |
|
|
84 |
// lastBreak points to the end of last recognized word
|
|
85 |
int lastBreak = -1;
|
|
86 |
// Continue until EOF
|
|
87 |
while (text_[cursor_]) {
|
|
88 |
// Feed next character to the state machine
|
|
89 |
// and try to transit the state
|
|
90 |
if (!state_.next(text_[cursor_++])) {
|
|
91 |
// Check last final state
|
|
92 |
if (lastBreak != -1) {
|
|
93 |
// Final state marked a valid word
|
|
94 |
// This is word boundary we were lookign
|
|
95 |
cursor_ = lastBreak;
|
|
96 |
}
|
|
97 |
break;
|
|
98 |
} else if (state_.isFinal()) {
|
|
99 |
// Found a valid word! Mark the location
|
|
100 |
lastBreak = cursor_;
|
|
101 |
// Still, continue and try to find even a longer word
|
|
102 |
}
|
|
103 |
}
|
|
104 |
next_ = cursor_;
|
|
105 |
}
|
|
106 |
}
|
|
107 |
|
|
108 |
}
|