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 |
#ifndef TINYANALYSIS_INL_
|
|
18 |
#define TINYANALYSIS_INL_
|
|
19 |
|
|
20 |
#include "CLucene.h"
|
|
21 |
|
|
22 |
/*
|
|
23 |
* TODO:
|
|
24 |
*
|
|
25 |
* Move more functionality in here
|
|
26 |
*
|
|
27 |
*/
|
|
28 |
|
|
29 |
namespace analysis {
|
|
30 |
namespace tiny {
|
|
31 |
namespace cl {
|
|
32 |
|
|
33 |
template<int SIZE>
|
|
34 |
wchar_t ReaderBuffer<SIZE>::operator[](int i)
|
|
35 |
{
|
|
36 |
if ( i < read_ - SIZE ) throw TooOldIndexException();
|
|
37 |
// fill
|
|
38 |
while (i >= read_) {
|
|
39 |
int c = reader_.read();
|
|
40 |
buf_[cut_++] = ( c == -1 ? '\0' : c );
|
|
41 |
cut_ %= SIZE;
|
|
42 |
if (read_++ >= SIZE) {
|
|
43 |
offset_++;
|
|
44 |
}
|
|
45 |
}
|
|
46 |
// guaranteed that i < read
|
|
47 |
return buf_[i % SIZE];
|
|
48 |
}
|
|
49 |
|
|
50 |
template<int SIZE>
|
|
51 |
ReaderBuffer<SIZE>::ReaderBuffer(lucene::util::Reader& reader)
|
|
52 |
: reader_(reader), read_(0), cut_(0), offset_(0) {}
|
|
53 |
|
|
54 |
template<int SIZE>
|
|
55 |
typename ReaderBuffer<SIZE>::iterator ReaderBuffer<SIZE>::at(int i) {
|
|
56 |
return iterator(*this, i);
|
|
57 |
}
|
|
58 |
|
|
59 |
template<int SIZE>
|
|
60 |
typename ReaderBuffer<SIZE>::iterator ReaderBuffer<SIZE>::begin() {
|
|
61 |
return iterator(*this, 0);
|
|
62 |
}
|
|
63 |
|
|
64 |
}
|
|
65 |
|
|
66 |
template <typename Iterator>
|
|
67 |
void Token<Iterator>::copyTo(lucene::analysis::Token* token) {
|
|
68 |
token->resetTermTextLen();
|
|
69 |
token->growBuffer(utf16size()+1);
|
|
70 |
token->setPositionIncrement(1);
|
|
71 |
Utf16Writer<wchar_t*> out(token->_termText);
|
|
72 |
Iterator i = begin_;
|
|
73 |
token->setStartOffset(i);
|
|
74 |
for (int n = length_; n; n--) {
|
|
75 |
out<<*i; ++i;
|
|
76 |
}
|
|
77 |
out<<L'\0';
|
|
78 |
token->setEndOffset(i);
|
|
79 |
}
|
|
80 |
|
|
81 |
}
|
|
82 |
}
|
|
83 |
|
|
84 |
#endif /* TINYANALYSIS_INL_ */
|