1 /**

     2  * Relational pipes

     3  * Copyright © 2020 František Kučera (Frantovo.cz, GlobalCode.info)

     4  *

     5  * This program is free software: you can redistribute it and/or modify

     6  * it under the terms of the GNU General Public License as published by

     7  * the Free Software Foundation, version 3 of the License.

     8  *

     9  * This program is distributed in the hope that it will be useful,

    10  * but WITHOUT ANY WARRANTY; without even the implied warranty of

    11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

    12  * GNU General Public License for more details.

    13  *

    14  * You should have received a copy of the GNU General Public License

    15  * along with this program. If not, see <http://www.gnu.org/licenses/>.

    16  */

    17 #pragma once

    18 

    19 #include <sstream>

    20 #include <codecvt>

    21 #include <arpa/inet.h>

    22 

    23 #include "UnescapingINIHandler.h"

    24 

    25 using namespace std;

    26 using namespace relpipe::writer;

    27 

    28 namespace relpipe {

    29 namespace in {

    30 namespace ini {

    31 namespace lib {

    32 

    33 class JavaPropertiesUnescapingINIContentHandler : public UnescapingINIContentHandler {

    34 private:

    35 	wstring_convert < codecvt_utf8<wchar_t>> convertor; // INI parser works with UTF-8

    36 

    37 	bool readHex(const char* hexadecimal, size_t hexLength, uint8_t* resultBuffer, size_t binLength) {

    38 		if (hexLength != binLength * 2) return false;

    39 

    40 		for (size_t i = 0; i < binLength; i++) {

    41 			uint8_t value = 0;

    42 			char a = hexadecimal[i * 2];

    43 			char b = hexadecimal[i * 2 + 1];

    44 

    45 			if (a >= '0' && a <= '9') value += (a - '0')*16;

    46 			else if (a >= 'a' && a <= 'f') value += (a - 'a' + 10)*16;

    47 			else if (a >= 'A' && a <= 'F') value += (a - 'A' + 10)*16;

    48 			else return false;

    49 

    50 			if (b >= '0' && b <= '9') value += b - '0';

    51 			else if (b >= 'a' && b <= 'f') value += b - 'a' + 10;

    52 			else if (b >= 'A' && b <= 'F') value += b - 'A' + 10;

    53 			else return false;

    54 

    55 			if (resultBuffer) resultBuffer[i] = value;

    56 		}

    57 		return true;

    58 	}

    59 

    60 protected:

    61 

    62 	virtual std::string unescape(const std::string& s) {

    63 		std::stringstream result;

    64 		for (int i = 0, length = s.size(); i < length; i++) {

    65 			char ch = s[i];

    66 			if (i + 1 < length && ch == ESC) {

    67 				ch = s[i + 1];

    68 				if (ch == 'u') {

    69 					// TODO: simplify, clean-up, verify (but seems working)

    70 					i++;

    71 					int hexLength = 4;

    72 					if (i + hexLength < length) {

    73 						uint16_t u16;

    74 						bool hexOK = readHex(s.c_str() + i + 1, hexLength, (uint8_t*) & u16, sizeof (u16));

    75 						if (hexOK) result << convertor.to_bytes(ntohs(u16));

    76 						else throw std::logic_error(std::string("Invalid unicode escape sequence: invalid HEX"));

    77 						i += hexLength;

    78 					} else {

    79 						throw std::logic_error(std::string("Invalid unicode escape sequence: missing characters"));

    80 					}

    81 

    82 				} else if (ch == ESC && !lastEscaphingPhase) put(result, ESC, i).put(ESC); // copy and skip even the second \ to avoid its misinterpretation in the next cycle

    83 				else if (ch == ESC && lastEscaphingPhase) put(result, ESC, i); // unescape \\ to \.

    84 				else if (lastEscaphingPhase) throw std::logic_error(std::string("Unsupported escape sequence: ") + ch);

    85 				else result.put(ESC); // keep the escape sequence for later unescaping phase

    86 			} else if (ch == ESC) {

    87 				throw std::logic_error(std::string("Missing escape sequence")); // this should not happen

    88 			} else {

    89 				result.put(ch);

    90 			}

    91 		}

    92 		return result.str();

    93 	}

    94 

    95 public:

    96 

    97 	JavaPropertiesUnescapingINIContentHandler(INIContentHandler& output, bool lastEscaphingPhase) : UnescapingINIContentHandler(output, lastEscaphingPhase, true) {

    98 	}

    99 

   100 };

   101 

   102 }

   103 }

   104 }

   105 }