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 

    18 #include <vector>

    19 #include <regex>

    20 

    21 #include "INIReader.h"

    22 

    23 class INIReaderImpl : public INIReader {

    24 private:

    25 	std::istream& input;

    26 	std::vector<INIContentHandler*> handlers;

    27 public:

    28 

    29 	INIReaderImpl(std::istream& input) : input(input) {

    30 	}

    31 

    32 	void addHandler(INIContentHandler* handler) override {

    33 		handlers.push_back(handler);

    34 	}

    35 

    36 	void process() override {

    37 

    38 		for (INIContentHandler* handler : handlers) handler->startDocument();

    39 

    40 		std::regex whitespacePattrern("\\s*");

    41 		std::regex commentPattrern("\\s*(;|#)\\s*(.*)");

    42 		std::regex sectionPattrern("\\s*\\[\\s*([^\\]]+)\\s*\\]\\s*");

    43 		std::regex entryQuotesPattrern(/***/"\\s*(([^=\\]]+?[^=\\s\\]]*)(\\[([^\\]]+)\\])?)\\s*=\\s*\"([^']+)\"\\s*((;|#)\\s*(.*))?");

    44 		std::regex entryApostrophesPattrern("\\s*(([^=\\]]+?[^=\\s\\]]*)(\\[([^\\]]+)\\])?)\\s*=\\s*'([^']+)'\\s*((;|#)\\s*(.*))?");

    45 		std::regex entryPlainPattrern("\\s*(([^=\\]]+?[^=\\s\\]]*)(\\[([^\\]]+)\\])?)\\s*=\\s*(.*)");

    46 

    47 		std::smatch match;

    48 		bool inSection = false;

    49 		std::string line;

    50 		int lineNumber = 0;

    51 		int eventNumber = 0;

    52 

    53 

    54 		while (std::getline(input, line)) {

    55 			lineNumber++;

    56 

    57 			if (std::regex_match(line, match, whitespacePattrern)) {

    58 				// TODO: support also whitespace

    59 			} else if (std::regex_match(line, match, commentPattrern)) {

    60 				// TODO: support also comments + emit also the comment style (;/#)

    61 			} else if (std::regex_match(line, match, sectionPattrern)) {

    62 				if (inSection) for (INIContentHandler* handler : handlers) handler->endSection();

    63 				inSection = true;

    64 				INIContentHandler::SectionStartEvent event;

    65 				event.lineNumber = lineNumber;

    66 				event.eventNumber = ++eventNumber;

    67 				event.name = match[1];

    68 				// TODO: support also comments + emit also the comment style (;/#)

    69 				for (INIContentHandler* handler : handlers) handler->startSection(event);

    70 			} else if (std::regex_match(line, match, entryQuotesPattrern) || std::regex_match(line, match, entryApostrophesPattrern) || std::regex_match(line, match, entryPlainPattrern)) {

    71 				INIContentHandler::EntryEvent event;

    72 				event.lineNumber = lineNumber;

    73 				event.eventNumber = ++eventNumber;

    74 				event.key = match[2];

    75 				event.subKey = match[4];

    76 				event.fullKey = match[1];

    77 				event.value = match[5];

    78 				if (match.size() == 9) event.comment = match[8];

    79 				// TODO: emit also the quote style ('/"/) and surrounding whitespace

    80 				for (INIContentHandler* handler : handlers) handler->entry(event);

    81 			} else {

    82 				// TODO: warning, error, or support unknown content

    83 			}

    84 

    85 			// TODO: probably switch to state-machine approach instead of regular expressions

    86 			// TODO: warning/error handler

    87 			// TODO: support also multiline content (\ + \n)

    88 			// TODO: support also quoted or multiline keys?

    89 			// TODO: support also escaped characters

    90 			// TODO: support also Java .properties and manifest.mf formats?

    91 			// TODO: support also nested sections – hierarchy

    92 			// TODO: support also nested keys e.g. key.sub.subsub.subsubsub=value – translate them to nested sections

    93 			// TODO: support also option for alternative key-value separator (: instead of =)

    94 			// TODO: support also other encodings (currently only UTF-8 is supported)

    95 		}

    96 

    97 		if (inSection) for (INIContentHandler* handler : handlers) handler->endSection();

    98 

    99 		for (INIContentHandler* handler : handlers) handler->endDocument();

   100 	}

   101 };

   102 

   103 INIReader* INIReader::create(std::istream& input) {

   104 	return new INIReaderImpl(input);

   105 }