MCPcopy Create free account
hub / github.com/catboost/catboost / Tokenize

Method Tokenize

library/cpp/tokenizer/tokenizer.cpp:14–39  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

12#include <util/stream/file.h>
13
14void TNlpTokenizer::Tokenize(const wchar16* str,
15 size_t size,
16 const TTokenizerOptions& opts) {
17 bool semicolonBreaksSentence = opts.LangMask == TLangMask(LANG_GRE);
18 TSentBreakFilter sentBreakFilter(opts.LangMask);
19 THolder<TNlpParser> parser;
20 switch (opts.Version) {
21 case 2:
22 parser = MakeHolder<TVersionedNlpParser<2>>(TokenHandler, sentBreakFilter, Buffer, opts.SpacePreserve,
23 BackwardCompatible, semicolonBreaksSentence, opts.UrlDecode);
24 break;
25 case 3:
26 parser = MakeHolder<TVersionedNlpParser<3>>(TokenHandler, sentBreakFilter, Buffer, opts.SpacePreserve,
27 BackwardCompatible, semicolonBreaksSentence, opts.UrlDecode, opts.KeepAffixes);
28 break;
29 default:
30 parser = MakeHolder<TDefaultNlpParser>(TokenHandler, sentBreakFilter, Buffer, opts.SpacePreserve,
31 BackwardCompatible, semicolonBreaksSentence, opts.UrlDecode);
32 break;
33 }
34 try {
35 parser->Execute(str, size, &TextStart);
36 } catch (const ITokenHandler::TAllDoneException&) {
37 // do nothing
38 }
39}
40
41#ifndef CATBOOST_OPENSOURCE
42void TNlpTokenizer::Tokenize(const char* text,

Callers 2

SplitIntoTokensFunction · 0.45
SplitIntoSentencesFunction · 0.45

Calls 4

CharToWideFunction · 0.85
TokenizeFunction · 0.85
ExecuteMethod · 0.45
DataMethod · 0.45

Tested by

no test coverage detected