Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
3 changed files
with
287 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,127 @@ | ||
/* | ||
* This program is free software; you can redistribute it and/or modify | ||
* it under the terms of the GNU General Public License as published by | ||
* the Free Software Foundation; either version 3 of the License, or | ||
* (at your option) any later version. | ||
* | ||
* Written (W) 2013 Evgeniy Andreev (gsomix) | ||
*/ | ||
|
||
#include <shogun/io/Parser.h> | ||
|
||
using namespace shogun; | ||
|
||
CParser::CParser() | ||
{ | ||
init(); | ||
} | ||
|
||
CParser::CParser(SGVector<char> text, CTokenizer* tokenizer) | ||
{ | ||
init(); | ||
|
||
set_text(text); | ||
set_tokenizer(tokenizer); | ||
} | ||
|
||
CParser::~CParser() | ||
{ | ||
m_text=SGVector<char>(); | ||
} | ||
|
||
bool CParser::has_next() | ||
{ | ||
if (m_tokenizer!=NULL) | ||
return m_tokenizer->has_next(); | ||
|
||
return false; | ||
} | ||
|
||
void CParser::skip_token() | ||
{ | ||
index_t start=0; | ||
m_tokenizer->next_token_idx(start); | ||
} | ||
|
||
SGVector<char> CParser::read_string() | ||
{ | ||
index_t start=0; | ||
index_t end=0; | ||
|
||
end=m_tokenizer->next_token_idx(start); | ||
|
||
SGVector<char> result=SGVector<char>(end-start); | ||
for (index_t i=start; i<end; i++) | ||
{ | ||
result[i-start]=m_text[i]; | ||
} | ||
|
||
return result; | ||
} | ||
|
||
bool CParser::read_bool() | ||
{ | ||
SGVector<char> token=read_string(); | ||
|
||
if (token.vlen>0) | ||
return (bool) strtol(token.vector, NULL, 10); | ||
else | ||
return (bool) 0L; | ||
} | ||
|
||
#define READ_INT_METHOD(fname, convf, sg_type) \ | ||
sg_type CParser::fname(int32_t base) \ | ||
{ \ | ||
SGVector<char> token=read_string(); \ | ||
\ | ||
if (token.vlen>0) \ | ||
return (sg_type) convf(token.vector, NULL, base); \ | ||
else \ | ||
return (sg_type) 0L; \ | ||
} | ||
|
||
READ_INT_METHOD(read_char, strtol, char) | ||
READ_INT_METHOD(read_byte, strtoul, uint8_t) | ||
READ_INT_METHOD(read_short, strtol, int16_t) | ||
READ_INT_METHOD(read_word, strtoul, uint16_t) | ||
READ_INT_METHOD(read_int, strtol, int32_t) | ||
READ_INT_METHOD(read_uint, strtoul, uint32_t) | ||
READ_INT_METHOD(read_long, strtoll, int64_t) | ||
READ_INT_METHOD(read_ulong, strtoull, uint64_t) | ||
#undef READ_INT_METHOD | ||
|
||
#define READ_REAL_METHOD(fname, convf, sg_type) \ | ||
sg_type CParser::fname() \ | ||
{ \ | ||
SGVector<char> token=read_string(); \ | ||
\ | ||
if (token.vlen>0) \ | ||
return (sg_type) convf(token.vector, NULL); \ | ||
else \ | ||
return (sg_type) 0L; \ | ||
} | ||
|
||
READ_REAL_METHOD(read_short_real, strtod, float32_t) | ||
READ_REAL_METHOD(read_real, strtod, float64_t) | ||
READ_REAL_METHOD(read_long_real, strtold, floatmax_t) | ||
#undef READ_REAL_METHOD | ||
|
||
void CParser::set_text(SGVector<char> text) | ||
{ | ||
m_text=text; | ||
if (m_tokenizer!=NULL) | ||
m_tokenizer->set_text(m_text); | ||
} | ||
|
||
void CParser::set_tokenizer(CTokenizer* tokenizer) | ||
{ | ||
m_tokenizer=tokenizer; | ||
if (m_tokenizer!=NULL) | ||
m_tokenizer->set_text(m_text); | ||
} | ||
|
||
void CParser::init() | ||
{ | ||
m_text=SGVector<char>(); | ||
m_tokenizer=NULL; | ||
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,92 @@ | ||
/* | ||
* This program is free software; you can redistribute it and/or modify | ||
* it under the terms of the GNU General Public License as published by | ||
* the Free Software Foundation; either version 3 of the License, or | ||
* (at your option) any later version. | ||
* | ||
* Written (W) 2013 Evgeniy Andreev (gsomix) | ||
*/ | ||
|
||
#ifndef __PARSER_H__ | ||
#define __PARSER_H__ | ||
|
||
#include <shogun/lib/SGVector.h> | ||
#include <shogun/lib/Tokenizer.h> | ||
|
||
namespace shogun | ||
{ | ||
/** @brief Class for reading from a string */ | ||
class CParser : public CSGObject | ||
{ | ||
public: | ||
/** default constructor */ | ||
CParser(); | ||
|
||
/** constructor | ||
* | ||
* @param string the text to parse | ||
* @param tokenizer tokenizer | ||
*/ | ||
CParser(SGVector<char> string, CTokenizer* tokenizer); | ||
|
||
/** destructor */ | ||
virtual ~CParser(); | ||
|
||
/** check for next line in the stream | ||
* | ||
* @return true if there is next line, false - otherwise | ||
*/ | ||
virtual bool has_next(); | ||
|
||
/** skip next token */ | ||
virtual void skip_token(); | ||
|
||
/** read string */ | ||
virtual SGVector<char> read_string(); | ||
|
||
/** read one of the several base data types. */ | ||
//@{ | ||
virtual bool read_bool(); | ||
virtual char read_char(int32_t base=10); | ||
virtual uint8_t read_byte(int32_t base=10); | ||
virtual int16_t read_short(int32_t base=10); | ||
virtual uint16_t read_word(int32_t base=10); | ||
virtual int32_t read_int(int32_t base=10); | ||
virtual uint32_t read_uint(int32_t base=10); | ||
virtual int64_t read_long(int32_t base=10); | ||
virtual uint64_t read_ulong(int32_t base=10); | ||
virtual float32_t read_short_real(); | ||
virtual float64_t read_real(); | ||
virtual floatmax_t read_long_real(); | ||
//@} | ||
|
||
/** set tokenizer | ||
* | ||
* @param tokenizer tokenizer | ||
*/ | ||
void set_tokenizer(CTokenizer* tokenizer); | ||
|
||
/** set the char array that requires tokenization | ||
* | ||
* @param txt the text to tokenize | ||
*/ | ||
void set_text(SGVector<char> text); | ||
|
||
/** @return object name */ | ||
virtual const char* get_name() const { return "Parser"; } | ||
|
||
private: | ||
/** class initialization */ | ||
void init(); | ||
|
||
private: | ||
/** text to tokenizer */ | ||
SGVector<char> m_text; | ||
|
||
/** tokenizer */ | ||
CTokenizer* m_tokenizer; | ||
}; | ||
|
||
} | ||
|
||
#endif /** __STRING_READER_H__ */ |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,68 @@ | ||
#include <shogun/lib/DelimiterTokenizer.h> | ||
#include <shogun/lib/SGVector.h> | ||
#include <shogun/io/Parser.h> | ||
|
||
#include <gtest/gtest.h> | ||
|
||
using namespace shogun; | ||
|
||
TEST(ParserTest, tokenization) | ||
{ | ||
int32_t ntokens=5; | ||
const char* text=" This is the ultimate test! "; | ||
const char* tokens[]={"This", "is", "the", "ultimate", "test!"}; | ||
SGVector<char> cv(const_cast<char* >(text), 30, false); | ||
|
||
CDelimiterTokenizer* tokenizer=new CDelimiterTokenizer(); | ||
tokenizer->init_for_whitespace(); | ||
tokenizer->set_skip_delimiters(true); | ||
|
||
CParser* reader=new CParser(cv, tokenizer); | ||
|
||
SGVector<char> token; | ||
int32_t num_tokens=0; | ||
while (reader->has_next()) | ||
{ | ||
token=reader->read_string(); | ||
|
||
EXPECT_EQ(strlen(tokens[num_tokens]), token.vlen); | ||
for (int32_t i=0; i<token.vlen; i++) | ||
{ | ||
EXPECT_EQ(tokens[num_tokens][i], token[i]); | ||
} | ||
num_tokens++; | ||
} | ||
EXPECT_EQ(num_tokens, ntokens); | ||
|
||
SG_UNREF(reader); | ||
SG_UNREF(tokenizer); | ||
} | ||
|
||
TEST(ParserTest, tokenization_reals) | ||
{ | ||
int32_t ntokens=5; | ||
const char* text="1.0, 1.1, 1.2, 1.3, 1.4\n"; | ||
float64_t tokens[]={1.0, 1.1, 1.2, 1.3, 1.4}; | ||
SGVector<char> cv(const_cast<char* >(text), 24, false); | ||
|
||
CDelimiterTokenizer* tokenizer=new CDelimiterTokenizer(); | ||
tokenizer->delimiters[' ']=1; | ||
tokenizer->delimiters[',']=1; | ||
tokenizer->delimiters['\n']=1; | ||
tokenizer->set_skip_delimiters(true); | ||
|
||
CParser* reader=new CParser(cv, tokenizer); | ||
|
||
float64_t tmp=0; | ||
int32_t num_tokens=0; | ||
while (reader->has_next()) | ||
{ | ||
tmp=reader->read_real(); | ||
EXPECT_EQ(tokens[num_tokens], tmp); | ||
num_tokens++; | ||
} | ||
EXPECT_EQ(num_tokens, ntokens); | ||
|
||
SG_UNREF(reader); | ||
SG_UNREF(tokenizer); | ||
} |