Refactoring to address some issues from #5895 (handling of unsigned numbers). Also adding a TODO testcase since the real issue (FP) is still not fixed

Makefile

@@ -214,6 +214,7 @@ TESTOBJ =     test/options.o \
               test/testtimer.o \
               test/testtoken.o \
               test/testtokenize.o \
+              test/testtokenlist.o \
               test/testtype.o \
               test/testuninitvar.o \
               test/testunusedfunctions.o \
@@ -573,6 +574,9 @@ test/testtoken.o: test/testtoken.cpp lib/cxx11emu.h test/testsuite.h lib/errorlo
 test/testtokenize.o: test/testtokenize.cpp lib/cxx11emu.h test/testsuite.h lib/errorlogger.h lib/config.h lib/suppressions.h test/redirect.h lib/library.h lib/mathlib.h lib/token.h lib/valueflow.h lib/tokenize.h lib/tokenlist.h lib/settings.h lib/standards.h lib/timer.h lib/path.h lib/preprocessor.h
 	$(CXX) ${INCLUDE_FOR_TEST} $(CPPFLAGS) $(CFG) $(CXXFLAGS) $(UNDEF_STRICT_ANSI) -std=c++0x -c -o test/testtokenize.o test/testtokenize.cpp
 
+test/testtokenlist.o: test/testtokenlist.cpp lib/cxx11emu.h test/testsuite.h lib/errorlogger.h lib/config.h lib/suppressions.h test/redirect.h lib/library.h lib/mathlib.h lib/token.h lib/valueflow.h lib/settings.h lib/standards.h lib/timer.h lib/tokenlist.h
+	$(CXX) ${INCLUDE_FOR_TEST} $(CPPFLAGS) $(CFG) $(CXXFLAGS) $(UNDEF_STRICT_ANSI) -std=c++0x -c -o test/testtokenlist.o test/testtokenlist.cpp
+
 test/testtype.o: test/testtype.cpp lib/cxx11emu.h lib/preprocessor.h lib/config.h lib/tokenize.h lib/errorlogger.h lib/suppressions.h lib/tokenlist.h lib/symboldatabase.h lib/token.h lib/valueflow.h lib/mathlib.h lib/checktype.h lib/check.h lib/settings.h lib/library.h lib/standards.h lib/timer.h test/testsuite.h test/redirect.h test/testutils.h
 	$(CXX) ${INCLUDE_FOR_TEST} $(CPPFLAGS) $(CFG) $(CXXFLAGS) $(UNDEF_STRICT_ANSI) -std=c++0x -c -o test/testtype.o test/testtype.cpp
 

lib/checkcondition.cpp

@@ -525,7 +525,8 @@ static std::string invertOperatorForOperandSwap(std::string s)
     return s;
 }
 
-static bool checkIntRelation(const std::string &op, const MathLib::bigint value1, const MathLib::bigint value2)
+template <typename T>
+static bool checkIntRelation(const std::string &op, const T value1, const T value2)
 {
     return (op == "==" && value1 == value2) ||
            (op == "!=" && value1 != value2) ||
@@ -543,7 +544,9 @@ static bool checkFloatRelation(const std::string &op, const double value1, const
            (op == "<=" && value1 <= value2);
 }
 
-template<class T> static T getvalue(const int test, const T value1, const T value2)
+
+template<class T>
+static inline T getvalue(const int test, const T value1, const T value2)
 {
     // test:
     // 1 => return value that is less than both value1 and value2
@@ -553,7 +556,7 @@ template<class T> static T getvalue(const int test, const T value1, const T valu
     // 5 => return value that is larger than both value1 and value2
     switch (test) {
     case 1: {
-        T ret = std::min(value1, value2);
+        const T ret = std::min(value1, value2);
         if ((ret - (T)1) < ret)
             return ret - (T)1;
         else if ((ret / (T)2) < ret)
@@ -565,11 +568,19 @@ template<class T> static T getvalue(const int test, const T value1, const T valu
     case 2:
         return value1;
     case 3:
+        if (std::numeric_limits<T>::is_integer) {
+            const T min = std::min(value1, value2);
+            if (min== std::numeric_limits<T>::max())
+                return min;
+            else
+                return min+1; // see #5895
+        } else {
             return (value1 + value2) / (T)2;
+        }
     case 4:
         return value2;
     case 5: {
-        T ret = std::max(value1, value2);
+        const T ret = std::max(value1, value2);
         if ((ret + (T)1) > ret)
             return ret + (T)1;
         else if ((ret / (T)2) > ret)
@@ -672,6 +683,13 @@ void CheckCondition::checkIncorrectLogicOperator()
                 if (isfloat && (op1 == "==" || op1 == "!=" || op2 == "==" || op2 == "!="))
                     continue;
 
+                const double d1 = (isfloat) ? MathLib::toDoubleNumber(value1) : 0;
+                const double d2 = (isfloat) ? MathLib::toDoubleNumber(value2) : 0;
+                const MathLib::bigint i1 = (isfloat) ? 0 : MathLib::toLongNumber(value1);
+                const MathLib::bigint i2 = (isfloat) ? 0 : MathLib::toLongNumber(value2);
+                const bool useUnsignedInt = (std::numeric_limits<MathLib::bigint>::max()==i1)||(std::numeric_limits<MathLib::bigint>::max()==i2);
+                const MathLib::biguint u1 = (useUnsignedInt) ? MathLib::toLongNumber(value1) : 0;
+                const MathLib::biguint u2 = (useUnsignedInt) ? MathLib::toLongNumber(value2) : 0;
                 // evaluate if expression is always true/false
                 bool alwaysTrue = true, alwaysFalse = true;
                 bool firstTrue = true, secondTrue = true;
@@ -684,14 +702,14 @@ void CheckCondition::checkIncorrectLogicOperator()
                     // 5 => testvalue is larger than both value1 and value2
                     bool result1, result2;
                     if (isfloat) {
-                        const double d1 = MathLib::toDoubleNumber(value1);
-                        const double d2 = MathLib::toDoubleNumber(value2);
                         const double testvalue = getvalue<double>(test, d1, d2);
                         result1 = checkFloatRelation(op1, testvalue, d1);
                         result2 = checkFloatRelation(op2, testvalue, d2);
+                    } else if (useUnsignedInt) {
+                        const MathLib::biguint testvalue = getvalue<MathLib::biguint>(test, u1, u2);
+                        result1 = checkIntRelation(op1, testvalue, u1);
+                        result2 = checkIntRelation(op2, testvalue, u2);
                     } else {
-                        const MathLib::bigint i1 = MathLib::toLongNumber(value1);
-                        const MathLib::bigint i2 = MathLib::toLongNumber(value2);
                         const MathLib::bigint testvalue = getvalue<MathLib::bigint>(test, i1, i2);
                         result1 = checkIntRelation(op1, testvalue, i1);
                         result2 = checkIntRelation(op2, testvalue, i2);

lib/tokenlist.cpp

@@ -132,7 +132,7 @@ void TokenList::addtoken(const std::string & str, const unsigned int lineno, con
     std::string str2;
     if (MathLib::isHex(str) || MathLib::isOct(str) || MathLib::isBin(str)) {
         std::ostringstream str2stream;
-        str2stream << MathLib::toLongNumber(str);
+        str2stream << MathLib::MathLib::toULongNumber(str);
         str2 = str2stream.str();
     } else if (str.compare(0, 5, "_Bool") == 0) {
         str2 = "bool";

test/testcondition.cpp

@@ -48,6 +48,7 @@ private:
         TEST_CASE(incorrectLogicOperator7); // opposite expressions: (expr || !expr)
         TEST_CASE(secondAlwaysTrueFalseWhenFirstTrueError);
         TEST_CASE(incorrectLogicOp_condSwapping);
+        TEST_CASE(testBug5895);
 
         TEST_CASE(modulo);
 
@@ -656,7 +657,6 @@ private:
               "    if (x == 1.0 && x == 3.0)\n"
               "        a++;\n"
               "}");
-        //ASSERT_EQUALS("[test.cpp:2]: (warning) Logical conjunction always evaluates to false: x == 1.0 && x == 3.0.\n", errout.str());
         ASSERT_EQUALS("", errout.str()); // float comparisons with == and != are not checked right now - such comparison is a bad idea
 
         check("void f(float x) {\n"
@@ -1408,6 +1408,14 @@ private:
               "}");
         ASSERT_EQUALS("", errout.str());
     }
+
+    void testBug5895() {
+        check("void png_parse(uint64_t init, int buf_size) {\n"
+              "    if (init == 0x89504e470d0a1a0a || init == 0x8a4d4e470d0a1a0a)\n"
+              "        ;\n"
+              "}");
+        TODO_ASSERT_EQUALS("", "[test.cpp:2]: (style) Redundant condition: If init == 9894494448401390090, the comparison init == 9965707617509186058 is always true.\n", errout.str());
+    }
 };
 
 REGISTER_TEST(TestCondition)

test/testfiles.pri

@@ -50,6 +50,7 @@ SOURCES += $${BASEPATH}/test64bit.cpp \
            $${BASEPATH}/testtimer.cpp \
            $${BASEPATH}/testtoken.cpp \
            $${BASEPATH}/testtokenize.cpp \
+           $${BASEPATH}/testtokenlist.cpp \
            $${BASEPATH}/testtype.cpp \
            $${BASEPATH}/testuninitvar.cpp \
            $${BASEPATH}/testunusedfunctions.cpp \