From 487c24bd4620ce2764bdb826e988037bd623419d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?M=C3=A5rten=20Nordheim?= <marten.nordheim@qt.io>
Date: Tue, 28 Nov 2023 17:01:17 +0100
Subject: [PATCH] QLocal8Bit::convert{To,From}Unicode[win]: use more of state

Like other backends we should increment the invalid character count
when we output a replacement character.
And we should also output the NULL character if requested!
The downside here is that convertFromUnicode doesn't even have the
ability to do so. So instead I added a comment explaining why it is not
handled there.

Task-number: QTBUG-118318
Pick-to: 6.5
Change-Id: I57ba631aa59454e77007ab353277b7e8c2b5526a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
(cherry picked from commit 3c8d71796499a2ca641758dc1e9af4a778c8ef41)
(cherry picked from commit 4483e814407b4dedbcea828e30d19784a8b7e462)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
---
 src/corelib/text/qstringconverter.cpp | 31 ++++++++++++++++++++++-----
 1 file changed, 26 insertions(+), 5 deletions(-)

diff --git a/src/corelib/text/qstringconverter.cpp b/src/corelib/text/qstringconverter.cpp
index 63b544a4b46..5b5e51f62c5 100644
--- a/src/corelib/text/qstringconverter.cpp
+++ b/src/corelib/text/qstringconverter.cpp
@@ -1268,8 +1268,16 @@ QString QLocal8Bit::convertToUnicode_sys(QByteArrayView in, quint32 codePage,
     const char *mb = in.data();
     qsizetype mblen = in.size();
 
-    if (state && state->flags & QStringConverter::Flag::Stateless)
+    Q_ASSERT(state);
+    qsizetype &invalidChars = state->invalidChars;
+    using Flag = QStringConverter::Flag;
+    const bool useNullForReplacement = !!(state->flags & Flag::ConvertInvalidToNull);
+    const char16_t replacementCharacter = useNullForReplacement ? QChar::Null
+                                                                : QChar::ReplacementCharacter;
+    if (state->flags & Flag::Stateless) {
+        Q_ASSERT(state->remainingChars == 0);
         state = nullptr;
+    }
 
     if (!mb || !mblen)
         return QString();
@@ -1320,7 +1328,8 @@ QString QLocal8Bit::convertToUnicode_sys(QByteArrayView in, quint32 codePage,
             // We couldn't decode any of the characters in the saved state,
             // so output replacement characters
             for (int i = 0; i < state->remainingChars; ++i)
-                out[i] = QChar::ReplacementCharacter;
+                out[i] = replacementCharacter;
+            invalidChars += state->remainingChars;
             out += state->remainingChars;
             outlen -= state->remainingChars;
             state->remainingChars = 0;
@@ -1406,7 +1415,8 @@ QString QLocal8Bit::convertToUnicode_sys(QByteArrayView in, quint32 codePage,
                 std::tie(out, outlen) = growOut(1);
                 if (!out)
                     return {};
-                *out = QChar::ReplacementCharacter;
+                *out = replacementCharacter;
+                ++invalidChars;
                 ++out;
                 --outlen;
                 ++mb;
@@ -1435,7 +1445,8 @@ QString QLocal8Bit::convertToUnicode_sys(QByteArrayView in, quint32 codePage,
     if (!state && mblen > 0) {
         // We have trailing character(s) that could not be converted, and
         // nowhere to cache them
-        sp.resize(sp.size() + mblen, QChar::ReplacementCharacter);
+        sp.resize(sp.size() + mblen, replacementCharacter);
+        invalidChars += mblen;
     }
     return sp;
 }
@@ -1452,8 +1463,18 @@ QByteArray QLocal8Bit::convertFromUnicode_sys(QStringView in, quint32 codePage,
     qsizetype uclen = in.size();
 
     Q_ASSERT(state);
-    if (state->flags & QStringConverter::Flag::Stateless) // temporary
+    // The Windows API has a *boolean* out-parameter that says if a replacement
+    // character was used, but it gives us no way to know _how many_ were used.
+    // Since we cannot simply scan the string for replacement characters
+    // (which is potentially a question mark, and thus a valid character),
+    // we simply do not track the number of invalid characters here.
+    // auto &invalidChars = state->invalidChars;
+
+    using Flag = QStringConverter::Flag;
+    if (state->flags & Flag::Stateless) { // temporary
+        Q_ASSERT(state->remainingChars == 0);
         state = nullptr;
+    }
 
     if (!ch)
         return QByteArray();