src/lib/BasicUnescapingINIHandler.h
author František Kučera <franta-hg@frantovo.cz>
Wed, 25 Nov 2020 21:50:26 +0100
branchv_0
changeset 26 80e129ec3408
permissions -rw-r--r--
new INI parser
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
26
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     1
/**
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     2
 * Relational pipes
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     3
 * Copyright © 2020 František Kučera (Frantovo.cz, GlobalCode.info)
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     4
 *
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     5
 * This program is free software: you can redistribute it and/or modify
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     6
 * it under the terms of the GNU General Public License as published by
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     7
 * the Free Software Foundation, version 3 of the License.
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     8
 *
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     9
 * This program is distributed in the hope that it will be useful,
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    10
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    11
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    12
 * GNU General Public License for more details.
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    13
 *
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    14
 * You should have received a copy of the GNU General Public License
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    15
 * along with this program. If not, see <http://www.gnu.org/licenses/>.
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    16
 */
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    17
#pragma once
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    18
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    19
#include <sstream>
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    20
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    21
#include "UnescapingINIHandler.h"
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    22
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    23
using namespace std;
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    24
using namespace relpipe::writer;
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    25
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    26
namespace relpipe {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    27
namespace in {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    28
namespace ini {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    29
namespace lib {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    30
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    31
class BasicUnescapingINIContentHandler : public UnescapingINIContentHandler {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    32
protected:
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    33
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    34
	virtual std::string unescape(const std::string& s) {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    35
		std::stringstream result;
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    36
		for (int i = 0, length = s.size(); i < length; i++) {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    37
			char ch = s[i];
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    38
			if (i + 1 < length && ch == ESC) {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    39
				ch = s[i + 1];
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    40
				if (ch == 'n') put(result, '\n', i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    41
				else if (ch == 'r') put(result, '\r', i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    42
				else if (ch == 't') put(result, '\t', i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    43
				else if (ch == 's') put(result, ' ', i); // TODO: Reconsider what is „basic“ escaping and should be supported.
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    44
				else if (ch == '"') put(result, ch, i); //        The delimiters (\n,]",') are already unescaped during the first stage in the INIReader while parsing (the delimiter relevant to given environment is unescaped, e.g. \" in "quoted" value).
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    45
				else if (ch == '\'') put(result, ch, i); //       So it does not necessary to do it here. But someone might write a="xxx\'zzz" however it is superfluous because a="xxx'zzz" will also work.
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    46
				else if (ch == ']') put(result, ch, i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    47
				else if (ch == ':') put(result, ch, i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    48
				else if (ch == ';') put(result, ch, i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    49
				else if (ch == '#') put(result, ch, i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    50
				else if (ch == '=') put(result, ch, i);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    51
				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
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    52
				else if (ch == ESC && lastEscaphingPhase) put(result, ESC, i); // unescape \\ to \.
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    53
				else if (lastEscaphingPhase) throw std::logic_error(std::string("Unsupported escape sequence: ") + ch);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    54
				else result.put(ESC); // keep the escape sequence for later unescaping phase
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    55
			} else if (ch == ESC) {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    56
				throw std::logic_error(std::string("Missing escape sequence")); // this should not happen
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    57
			} else {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    58
				result.put(ch);
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    59
			}
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    60
		}
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    61
		return result.str();
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    62
	}
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    63
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    64
public:
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    65
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    66
	BasicUnescapingINIContentHandler(INIContentHandler& output, bool lastEscaphingPhase) : UnescapingINIContentHandler(output, lastEscaphingPhase) {
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    67
	}
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    68
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    69
};
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    70
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    71
}
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    72
}
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    73
}
80e129ec3408 new INI parser
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    74
}