Teach qmake about digit-grouping apostrophes in numeric literals

It was previously understanding them as character literal delimiters,
with unfortunate consequences if a numeric literal contained an odd
number of them. Recognize that an apostrophe with a digit on each side
of it isn't the opening quote of a character literal (unless the digit
before it is preceded by a u). Extend the findMocs test to trigger the
bug, prior to the fix; verified it passes with the fix.

Fixes: QTBUG-98845
Change-Id: I5db3ac59aaeade7c2d6c1fb680ba97261ec0e8a9
Reviewed-by: Jörg Bornemann <[email protected]>

qmake/generators/makefiledeps.cpp

@@ -365,13 +365,21 @@ static bool matchWhileUnsplitting(const char *buffer, int buffer_len, int start,
     return true;
 }
 
-/* Advance from an opening quote at buffer[offset] to the matching close quote. */
+/* Advance from an opening quote at buffer[offset] to the matching close quote.
+   If an apostrophe turns out to be a digit-separator in a numeric literal,
+   rather than the start of a character literal, treat it as both the open and
+   the close quote of the "string" that isn't there.
+*/
 static int scanPastString(char *buffer, int buffer_len, int offset, int *lines)
 {
     // http://en.cppreference.com/w/cpp/language/string_literal
     // It might be a C++11 raw string.
     bool israw = false;
-    if (buffer[offset] == '"' && offset > 0) {
+
+    Q_ASSERT(offset < buffer_len);
+    if (offset <= 0) {
+        // skip, neither of these special cases applies here
+    } else if (buffer[offset] == '"') {
         int explore = offset - 1;
         bool prefix = false; // One of L, U, u or u8 may appear before R
         bool saw8 = false; // Partial scan of u8
@@ -415,6 +423,19 @@ static int scanPastString(char *buffer, int buffer_len, int offset, int *lines)
             && (isalnum(buffer[explore]) || buffer[explore] == '_')) {
             israw = false;
         }
+
+    } else {
+        // Is this apostrophe a digit separator rather than the start of a
+        // character literal ? If so, there was no string to scan past, so
+        // treat the apostrophe as both open and close.
+        Q_ASSERT(buffer[offset] == '\'' && offset > 0);
+        // Wrap std::isdigit() to package the casting to unsigned char.
+        const auto isDigit = [](unsigned char c) { return std::isdigit(c); };
+        if (isDigit(buffer[offset - 1]) && offset + 1 < buffer_len && isDigit(buffer[offset + 1])) {
+            // One exception: u8'0' is a perfectly good character literal.
+            if (offset < 2 || buffer[offset - 1] != '8' || buffer[offset - 2] != 'u')
+                return offset;
+        }
     }
 
     if (israw) {

tests/auto/tools/qmake/testdata/findMocs/digitseparated.h

@@ -0,0 +1,8 @@
+#include <QObject>
+
+class AfterDigitSeparator : public QObject
+{
+    Q_OBJECT
+public:
+    AfterDigitSeparator() : QObject(nullptr) {}
+};

tests/auto/tools/qmake/testdata/findMocs/findMocs.pro

@@ -1,5 +1,6 @@
 DESTDIR	= ./
 
 HEADERS += object1.h object2.h object3.h object4.h \
-        object5.h object6.h object7.h object8.h object9.h
+        object5.h object6.h object7.h object8.h object9.h \
+        digitseparated.h
 SOURCES += main.cpp

tests/auto/tools/qmake/testdata/findMocs/main.cpp

@@ -1,7 +1,6 @@
 // Copyright (C) 2016 The Qt Company Ltd.
 // SPDX-License-Identifier: LicenseRef-Qt-Commercial OR GPL-3.0-only WITH Qt-GPL-exception-1.0
 
-
 #include <moc_object1.cpp>
 #include <moc_object2.cpp>
 #include <moc_object3.cpp>
@@ -12,4 +11,9 @@
 #include     "object8.h"
 #include <moc_object9.cpp>
 
-int main() { return 0; }
+int main() { return 0'000; }
+/* Included *after* the use of a numeric literal with an *odd* number of digit
+   separator tick marks in it (and no subsequent apostrophe to end the
+   "single-quoted character literal" that the old parser though it was thus left
+   in).  */
+#include <moc_digitseparated.cpp>