From b436814eb57a2176eafb191e326d6419d938f840 Mon Sep 17 00:00:00 2001
From: eelke <eelke@eelkeklein.nl>
Date: Sat, 6 Jan 2018 21:18:28 +0100
Subject: [PATCH] THe SqlLexer also now recognizes casts.

---
 core/SqlLexer.cpp                | 29 +++++++++++++++++++++++++++++
 core/SqlLexer.h                  |  3 ++-
 tests/auto/mycase/tst_SqlLexer.h | 24 ++++++++++++++++++++++++
 3 files changed, 55 insertions(+), 1 deletion(-)

diff --git a/core/SqlLexer.cpp b/core/SqlLexer.cpp
index e211996..ecb94e6 100644
--- a/core/SqlLexer.cpp
+++ b/core/SqlLexer.cpp
@@ -23,6 +23,24 @@ QChar SqlLexer::peekChar()
 	return result;
 }
 
+
+//+ - * / < > = ~ ! @ # % ^ & | ` ?
+//There are a few restrictions on your choice of name:
+//   -- and /* cannot appear anywhere in an operator name, since they will be taken as the start of a comment.
+//   A multicharacter operator name cannot end in + or -, unless the name also contains at least one of these characters:
+//   ~ ! @ # % ^ & | ` ?
+//   For example, @- is an allowed operator name, but *- is not. This restriction allows PostgreSQL to parse SQL-compliant commands without requiring spaces between tokens.
+//   The use of => as an operator name is deprecated. It may be disallowed altogether in a future release.
+//The operator != is mapped to <> on input, so these two names are always equivalent.
+template <typename C>
+inline bool isOperatorChar(C c)
+{
+	return c == '+' || c == '-' || c == '*' || c == '/' || c == '<' || c == '>' || c == '='
+		|| c == '~' || c == '!' || c == '@' || c == '#' || c == '%' || c == '^' || c == '&'
+		|| c == '|' || c == '`' || c == '?';
+}
+
+
 /**
  * @brief NextBasicToken
  * @param in
@@ -75,6 +93,17 @@ bool SqlLexer::nextBasicToken(int &startpos, int &length, BasicTokenType &tokent
 			else if (c == '$') {
                 return parseDollarQuote(startpos, length, tokentype, out);                
             }
+			else if (c == ':') {
+				c = peekChar();
+				if (c == ':') {
+					nextChar();
+					length = m_pos - startpos;
+					tokentype = BasicTokenType::Cast;
+					QStringRef sr(&m_block, startpos, length);
+					out = sr.toString();
+					return true;
+				}
+			}
 			else {
 				// Undetermined symbol
 				for (;;) {
diff --git a/core/SqlLexer.h b/core/SqlLexer.h
index 18bbd35..c9684d8 100644
--- a/core/SqlLexer.h
+++ b/core/SqlLexer.h
@@ -13,7 +13,8 @@ enum class BasicTokenType {
 	QuotedString,
 	DollarQuote, // Return the dollar quote tag, do not consume the entire string (potentially long)
 	QuotedIdentifier,
-    Parameter
+	Parameter,
+	Cast
 };
 
 enum class LexerState {
diff --git a/tests/auto/mycase/tst_SqlLexer.h b/tests/auto/mycase/tst_SqlLexer.h
index 25d6535..76dd1f7 100644
--- a/tests/auto/mycase/tst_SqlLexer.h
+++ b/tests/auto/mycase/tst_SqlLexer.h
@@ -36,3 +36,27 @@ TEST(SqlLexer, lexer_quote_in_string)
 	ASSERT_THAT(tokentype, Eq(BasicTokenType::QuotedString));
 }
 
+TEST(SqlLexer, lexer_cast)
+{
+	QString input = "'1'::integer";
+	SqlLexer lexer(input, LexerState::Null);
+
+	int startpos, length;
+	BasicTokenType tokentype;
+	QString out;
+	lexer.nextBasicToken(startpos, length, tokentype, out);
+
+	ASSERT_THAT(startpos, Eq(0));
+	ASSERT_THAT(length, Eq(3));
+	ASSERT_THAT(tokentype, Eq(BasicTokenType::QuotedString));
+	lexer.nextBasicToken(startpos, length, tokentype, out);
+
+	ASSERT_THAT(startpos, Eq(3));
+	ASSERT_THAT(length, Eq(2));
+	ASSERT_THAT(tokentype, Eq(BasicTokenType::Cast));
+	lexer.nextBasicToken(startpos, length, tokentype, out);
+
+	ASSERT_THAT(startpos, Eq(5));
+	ASSERT_THAT(length, Eq(7));
+	ASSERT_THAT(tokentype, Eq(BasicTokenType::Symbol));
+}