1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
|
#include "split.h"
#include "tokenizer.h"
namespace {
class TSimpleTokenHandler: public ITokenHandler {
public:
TSimpleTokenHandler(TVector<TUtf16String>* outTokens, const TTokenizerSplitParams& params)
: Tokens(outTokens)
, Params(params)
{
}
void OnToken(const TWideToken& token, size_t, NLP_TYPE type) override {
if (!Params.HandledMask.SafeTest(type)) {
return;
}
Tokens->push_back(TUtf16String(token.Token, token.Leng));
}
private:
TVector<TUtf16String>* Tokens;
TTokenizerSplitParams Params;
};
class TSimpleSentenceHandler: public ITokenHandler {
public:
TSimpleSentenceHandler(TVector<TUtf16String>* sentences)
: Sentences(sentences)
{
}
void OnToken(const TWideToken& token, size_t, NLP_TYPE type) override {
CurToken += token.Text();
if (type == NLP_SENTBREAK || type == NLP_PARABREAK) {
Flush();
}
}
void Flush() {
if (!CurToken.empty()) {
Sentences->push_back(CurToken);
CurToken = TUtf16String();
}
}
private:
TUtf16String CurToken;
TVector<TUtf16String>* Sentences;
};
}
const TTokenizerSplitParams::THandledMask TTokenizerSplitParams::WORDS(NLP_WORD);
const TTokenizerSplitParams::THandledMask TTokenizerSplitParams::NOT_PUNCT(NLP_WORD, NLP_INTEGER, NLP_FLOAT, NLP_MARK);
TVector<TUtf16String> SplitIntoTokens(const TUtf16String& text, const TTokenizerSplitParams& params) {
TVector<TUtf16String> words;
TSimpleTokenHandler handler(&words, params);
TNlpTokenizer tokenizer(handler, params.BackwardCompatibility);
TTokenizerOptions opts { params.SpacePreserve, params.TokenizerLangMask, params.UrlDecode };
tokenizer.Tokenize(text.data(), text.size(), opts);
return words;
}
TVector<TUtf16String> SplitIntoSentences(const TUtf16String& text, const TTokenizerSplitParams& params) {
TVector<TUtf16String> sentences;
TSimpleSentenceHandler handler(&sentences);
TNlpTokenizer tokenizer(handler, params.BackwardCompatibility);
TTokenizerOptions opts { params.SpacePreserve, params.TokenizerLangMask, params.UrlDecode };
tokenizer.Tokenize(text.data(), text.size(), opts);
handler.Flush();
return sentences;
}
|
