Bootstrap: remove the UTF-16 and UTF-32 codecs

Unlike most of everything else in the Bootstrap lib, this is code that
couldn't be eliminated by the linker because they were referenced in one
static array. Maybe an exceptionally smart whole-program analysis could
do it, but GCC and Clang LTO modes don't do that now.

I removed the code that performed detection from HTML and from data too.
I could have left the detection of UTF-8 and "other" but this code
wasn't necessary. In particular, QTextStream couldn't benefit from it
because it already defaults to UTF-8, so the detection code would never
determine anything different from the input.

Drive-by removed QStringConverter::availableCodecs() too because it was
in the middle of functions #ifdef'ed out to.

This reduced the size of release-mode moc
   text    data     bss     dec     hex filename
1079858    5440     640 1085938  1091f2 original/moc
1074386    5200     640 1080226  107ba2 updated/moc
  -5472    -240       0   -5712         difference

Change-Id: I01ec3c774d9943adb903fffd17b7f114c42874ac
Reviewed-by: Lars Knoll <lars@knoll.priv.no>

4 files changed, 20 insertions, 3 deletions

diff --git a/src/corelib/text/qstring.cpp b/src/corelib/text/qstring.cpp
index 30054e5fe2..716b5823fb 100644
--- a/src/corelib/text/qstring.cpp
+++ b/src/corelib/text/qstring.cpp
@@ -3419,6 +3419,7 @@ QString &QString::assign(QAnyStringView s)
     return *this;
 }
 
+#ifndef QT_BOOTSTRAPPED
 QString &QString::assign_helper(const char32_t *data, qsizetype len)
 {
     // worst case: each char32_t requires a surrogate pair, so
@@ -3438,6 +3439,7 @@ QString &QString::assign_helper(const char32_t *data, qsizetype len)
     }
     return *this;
 }
+#endif
 
 /*!
   \fn QString &QString::remove(qsizetype position, qsizetype n)
@@ -6022,6 +6024,7 @@ QString QString::fromUtf8(QByteArrayView ba)
     return QUtf8::convertToUnicode(ba);
 }
 
+#ifndef QT_BOOTSTRAPPED
 /*!
     \since 5.3
     Returns a QString initialized with the first \a size characters
@@ -6083,7 +6086,7 @@ QString QString::fromUcs4(const char32_t *unicode, qsizetype size)
     QStringDecoder toUtf16(QStringDecoder::Utf32, QStringDecoder::Flag::Stateless);
     return toUtf16(QByteArrayView(reinterpret_cast<const char *>(unicode), size * 4));
 }
-
+#endif // !QT_BOOTSTRAPPED
 
 /*!
     Resizes the string to \a size characters and copies \a unicode
diff --git a/src/corelib/text/qstring.h b/src/corelib/text/qstring.h
index 854850a981..895ec4b5c0 100644
--- a/src/corelib/text/qstring.h
+++ b/src/corelib/text/qstring.h
@@ -1311,7 +1311,11 @@ QString QString::fromWCharArray(const wchar_t *string, qsizetype size)
     if constexpr (sizeof(wchar_t) == sizeof(QChar)) {
         return QString(reinterpret_cast<const QChar *>(string), size);
     } else {
+#ifdef QT_BOOTSTRAPPED
+        Q_UNREACHABLE_RETURN(QString());
+#else
         return fromUcs4(reinterpret_cast<const char32_t *>(string), size);
+#endif
     }
 }
 
diff --git a/src/corelib/text/qstringconverter.cpp b/src/corelib/text/qstringconverter.cpp
index 6ada8a842a..efa625e30b 100644
--- a/src/corelib/text/qstringconverter.cpp
+++ b/src/corelib/text/qstringconverter.cpp
@@ -942,6 +942,7 @@ int QUtf8::compareUtf8(QByteArrayView lhs, QByteArrayView rhs, Qt::CaseSensitivi
     return (end1 > src1) - (end2 > src2);
 }
 
+#ifndef QT_BOOTSTRAPPED
 QByteArray QUtf16::convertFromUnicode(QStringView in, QStringConverter::State *state, DataEndianness endian)
 {
     bool writeBom = !(state->internalState & HeaderDone) && state->flags & QStringConverter::Flag::WriteBom;
@@ -1251,6 +1252,7 @@ QChar *QUtf32::convertToUnicode(QChar *out, QByteArrayView in, QStringConverter:
 
     return out;
 }
+#endif // !QT_BOOTSTRAPPED
 
 #if defined(Q_OS_WIN) && !defined(QT_BOOTSTRAPPED)
 int QLocal8Bit::checkUtf8()
@@ -1647,6 +1649,7 @@ void QStringConverter::State::reset() noexcept
     }
 }
 
+#ifndef QT_BOOTSTRAPPED
 static QChar *fromUtf16(QChar *out, QByteArrayView in, QStringConverter::State *state)
 {
     return QUtf16::convertToUnicode(out, in, state, DetectEndianness);
@@ -1706,6 +1709,7 @@ static char *toUtf32LE(char *out, QStringView in, QStringConverter::State *state
 {
     return QUtf32::convertFromUnicode(out, in, state, LittleEndianness);
 }
+#endif // !QT_BOOTSTRAPPED
 
 char *QLatin1::convertFromUnicode(char *out, QStringView in, QStringConverter::State *state) noexcept
 {
@@ -1747,11 +1751,13 @@ static char *toLocal8Bit(char *out, QStringView in, QStringConverter::State *sta
 static qsizetype fromUtf8Len(qsizetype l) { return l + 1; }
 static qsizetype toUtf8Len(qsizetype l) { return 3*(l + 1); }
 
+#ifndef QT_BOOTSTRAPPED
 static qsizetype fromUtf16Len(qsizetype l) { return l/2 + 2; }
 static qsizetype toUtf16Len(qsizetype l) { return 2*(l + 1); }
 
 static qsizetype fromUtf32Len(qsizetype l) { return l/2 + 2; }
 static qsizetype toUtf32Len(qsizetype l) { return 4*(l + 1); }
+#endif
 
 static qsizetype fromLatin1Len(qsizetype l) { return l + 1; }
 static qsizetype toLatin1Len(qsizetype l) { return l + 1; }
@@ -1887,12 +1893,14 @@ static qsizetype toLatin1Len(qsizetype l) { return l + 1; }
 const QStringConverter::Interface QStringConverter::encodingInterfaces[QStringConverter::LastEncoding + 1] =
 {
     { "UTF-8", QUtf8::convertToUnicode, fromUtf8Len, QUtf8::convertFromUnicode, toUtf8Len },
+#ifndef QT_BOOTSTRAPPED
     { "UTF-16", fromUtf16, fromUtf16Len, toUtf16, toUtf16Len },
     { "UTF-16LE", fromUtf16LE, fromUtf16Len, toUtf16LE, toUtf16Len },
     { "UTF-16BE", fromUtf16BE, fromUtf16Len, toUtf16BE, toUtf16Len },
     { "UTF-32", fromUtf32, fromUtf32Len, toUtf32, toUtf32Len },
     { "UTF-32LE", fromUtf32LE, fromUtf32Len, toUtf32LE, toUtf32Len },
     { "UTF-32BE", fromUtf32BE, fromUtf32Len, toUtf32BE, toUtf32Len },
+#endif
     { "ISO-8859-1", QLatin1::convertToUnicode, fromLatin1Len, QLatin1::convertFromUnicode, toLatin1Len },
     { "Locale", fromLocal8Bit, fromUtf8Len, toLocal8Bit, toUtf8Len }
 };
@@ -2240,6 +2248,7 @@ std::optional<QStringConverter::Encoding> QStringConverter::encodingForName(cons
     return std::nullopt;
 }
 
+#ifndef QT_BOOTSTRAPPED
 /*!
    Returns the encoding for the content of \a data if it can be determined.
    \a expectedFirstCharacter can be passed as an additional hint to help determine
@@ -2401,7 +2410,6 @@ QStringList QStringConverter::availableCodecs()
     return result;
 }
 
-
 /*!
     Tries to determine the encoding of the HTML in \a data by looking at leading byte
     order marks or a charset specifier in the HTML meta tag and returns a QStringDecoder
@@ -2425,7 +2433,7 @@ QStringDecoder QStringDecoder::decoderForHtml(QByteArrayView data)
 
     return QStringDecoder(Utf8);
 }
-
+#endif // !QT_BOOTSTRAPPED
 
 /*!
     Returns the canonical name for encoding \a e.
diff --git a/src/corelib/text/qstringconverter_base.h b/src/corelib/text/qstringconverter_base.h
index 19b6399c05..d6b6fcb484 100644
--- a/src/corelib/text/qstringconverter_base.h
+++ b/src/corelib/text/qstringconverter_base.h
@@ -89,12 +89,14 @@ public:
 
     enum Encoding {
         Utf8,
+#ifndef QT_BOOTSTRAPPED
         Utf16,
         Utf16LE,
         Utf16BE,
         Utf32,
         Utf32LE,
         Utf32BE,
+#endif
         Latin1,
         System,
         LastEncoding = System