src/Iconv.cpp
author František Kučera <franta-hg@frantovo.cz>
Wed, 20 Nov 2019 14:56:41 +0100
branchv_0
changeset 18 9d566568d37c
permissions -rw-r--r--
Iconv and TextCodec classes for converting text encodings
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
18
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     1
/**
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     2
 * Relational pipes (library)
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     3
 * Copyright © 2019 František Kučera (Frantovo.cz, GlobalCode.info)
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     4
 *
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     5
 * This program is free software: you can redistribute it and/or modify
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     6
 * it under the terms of the:
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     7
 *  - GNU Lesser General Public License as published by the Free Software Foundation;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     8
 *    version 3 of the License or (at your option)
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     9
 *  - GNU General Public License as published by the Free Software Foundation;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    10
 *    version 2 of the License.
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    11
 *
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    12
 * This program is distributed in the hope that it will be useful,
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    13
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    14
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    15
 * GNU General Public License for more details.
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    16
 *
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    17
 * You should have received a copy of the GNU General Public License
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    18
 * along with this program. If not, see <http://www.gnu.org/licenses/>.
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    19
 */
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    20
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    21
#include "../include/relpipe/common/text/Iconv.h"
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    22
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    23
namespace relpipe {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    24
namespace common {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    25
namespace text {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    26
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    27
class Iconv::IconvInternal {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    28
public:
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    29
	static const size_t BUFFER_SIZE_MINIMUM;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    30
	static const size_t BUFFER_SIZE_DEFAULT;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    31
	const size_t bufferSize;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    32
	iconv_t cd;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    33
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    34
	IconvInternal(std::string to, std::string from, const size_t bufferSize) : bufferSize(std::max(bufferSize, BUFFER_SIZE_MINIMUM)) {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    35
		cd = iconv_open(to.c_str(), from.c_str());
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    36
		if (errno) throw std::string("iconv_open() error: ") + strerror(errno); // TODO: custom exception
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    37
	}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    38
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    39
	virtual ~IconvInternal() {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    40
		iconv_close(cd);
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    41
	}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    42
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    43
};
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    44
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    45
const size_t Iconv::IconvInternal::BUFFER_SIZE_MINIMUM = 2;  // TODO: enough?
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    46
const size_t Iconv::IconvInternal::BUFFER_SIZE_DEFAULT = 20;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    47
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    48
Iconv::Iconv(std::string to, std::string from) : Iconv(to, from, IconvInternal::BUFFER_SIZE_DEFAULT) {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    49
}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    50
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    51
Iconv::Iconv(std::string to, std::string from, size_t bufferSize) {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    52
	internal = new Iconv::IconvInternal(to, from, bufferSize);
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    53
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    54
}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    55
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    56
Iconv::~Iconv() {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    57
	delete internal;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    58
}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    59
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    60
std::string Iconv::convert(std::string originalText) {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    61
	std::stringstream result;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    62
	size_t inBytesLeft = originalText.size();
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    63
	size_t outBytesLeft = internal->bufferSize;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    64
	char outBuffer[internal->bufferSize];
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    65
	char* outBuf = (char*) outBuffer;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    66
	char* inBuf = (char*) originalText.c_str();
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    67
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    68
	do {
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    69
		size_t nconv = iconv(internal->cd, &inBuf, &inBytesLeft, &outBuf, &outBytesLeft);
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    70
		if (nconv < 0) throw std::string("iconv() error: ") + strerror(errno); // TODO: custom exception
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    71
		// TODO: throw exception if locale is not initialized and platform default "" is used (avoid infinite loop)
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    72
		result.write(outBuffer, internal->bufferSize - outBytesLeft);
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    73
		outBytesLeft = internal->bufferSize;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    74
		outBuf = (char*) outBuffer;
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    75
	} while (inBytesLeft > 0);
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    76
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    77
	return result.str();
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    78
}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    79
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    80
}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    81
}
9d566568d37c Iconv and TextCodec classes for converting text encodings
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    82
}