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.

     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 <stdexcept>

    21 #include <string>

    22 

    23 namespace relpipe {

    24 namespace tr {

    25 namespace sql {

    26 

    27 /**

    28  * An SQL script may consist of several SQL statements separated by „;“ character.

    29  * But this character may also occur in literal values or quoted identifiers where it does not separate SQL statements.

    30  * This scanner reads SQL script character by character and separates particular SQL statements.

    31  * It recognizes comments, literals, quoted values and other parts (see enum State),

    32  * so it is able to distinguish which „;“ separates SQL statements from other „;“ that should be ignored.

    33  * 

    34  * Usage:

    35  * Append characters using append() and when it returns true,

    36  * call getAndReset() and use the returned statement.

    37  * If there are any remaining characters on the input, start again with appending…

    38  */

    39 class SqlInputScanner {

    40 private:

    41 

    42 	enum class State {

    43 		/** SQL code like keywords, functions, operators or identifiers (not quoted ones) */

    44 		SQL,

    45 		/** classic SQL comment: -- terminated by the line end */

    46 		COMMENT_SINGLE,

    47 		/** comment like this one */

    48 		COMMENT_MULTI,

    49 		/** quoted identifier like "column_name" */

    50 		IDENTIFIER,

    51 		/** string literal like 'some value' */

    52 		STRING_VALUE

    53 	};

    54 

    55 	State state = State::SQL;

    56 	bool includeSeparators = true;

    57 	std::wstringstream current;

    58 	wchar_t last = L'\0';

    59 

    60 public:

    61 	/**

    62 	 * @param ch next character from the input stream

    63 	 * @return whether the accumulated SQL statement is complete – then call getAndReset(), otherwise continue appending

    64 	 */

    65 	bool append(wchar_t ch);

    66 

    67 	/**

    68 	 * Returns current buffer and empties it, so append() can be called again (building new SQL statement)

    69 	 * @return accumulated SQL statement

    70 	 */

    71 	std::wstring getAndReset();

    72 

    73 	/**

    74 	 * @param includeSeparators whether „;“ separators should be appended to the buffer

    75 	 * and then returned from getAndReset() as part of the SQL statement

    76 	 */

    77 	void setIncludeSeparators(bool includeSeparators) {

    78 		this->includeSeparators = includeSeparators;

    79 	}

    80 };

    81 

    82 bool SqlInputScanner::append(wchar_t ch) {

    83 	if (state != State::SQL || (ch != L';') || (includeSeparators && ch == L';')) current << ch;

    84 

    85 	if (state == State::SQL) {

    86 		if (ch == L'"') state = State::IDENTIFIER;

    87 		else if (ch == L'\'') state = State::STRING_VALUE;

    88 		else if (ch == L'-' && last == L'-') state = State::COMMENT_SINGLE;

    89 		else if (ch == L'*' && last == L'/') state = State::COMMENT_MULTI;

    90 	} else if (state == State::COMMENT_SINGLE) {

    91 		if (ch == L'\n') state = State::SQL;

    92 	} else if (state == State::COMMENT_MULTI) {

    93 		if (ch == L'/' && last == L'*') state = State::SQL;

    94 	} else if (state == State::IDENTIFIER) {

    95 		if (ch == L'"') state = State::SQL;

    96 	} else if (state == State::STRING_VALUE) {

    97 		if (ch == L'\'') state = State::SQL;

    98 	} else {

    99 		throw std::domain_error("Unsupported SqlInputScanner state (bug in code)");

   100 	}

   101 

   102 	last = ch;

   103 	return state == State::SQL && ch == L';';

   104 };

   105 

   106 std::wstring SqlInputScanner::getAndReset() {

   107 	std::wstring str = current.str();

   108 	current.str(L"");

   109 	current.clear();

   110 	return str;

   111 }

   112 

   113 }

   114 }

   115 }