Do not exclude Unicode noncharacters from rtl_convertUnicodeToText

For one, that broke round-tripping with e.g. UTF-8 (see the test case added to
Test::testComplex in sal/qa/rtl/textenc/rtl_textcvt.cxx) which did not treat
noncharacters as invalid.

For another, <https://unicode.org/faq/private_use.html#nonchar7> is meanwhile
quite clear on the matter:

"Q: Are noncharacters prohibited in interchange?

"A: This question has led to some controversy, because the Unicode Standard has
been somewhat ambiguous about the status of noncharacters. The formal wording of
the definition of 'noncharacter' in the standard has always indicated that
noncharacters 'should never be interchanged.' That led some people to assume
that the definition actually meant 'shall not be interchanged' and that
therefore the presence of a noncharacter in any Unicode string immediately
rendered that string malformed according to the standard. But the intended use
of noncharacters requires the ability to exchange them in a limited context, at
least across APIs and even through data files and other means of 'interchange',
so that they can be processed as intended. The choice of the word 'should' in
the original definition was deliberate, and indicated that one should not try to
interchange noncharacters precisely because their interpretation is strictly
internal to whatever implementation uses them, so they have no publicly
interchangeable semantics. But other informative wording in the text of the core
specification and in the character names list was differently and more strongly
worded, leading to contradictory interpretations.

"Given this ambiguity of intent, in 2013 the UTC issued Corrigendum #9, which
deleted the phrase 'and that should never be interchanged' from the definition
of noncharacters, to make it clear that prohibition from interchange is not part
of the formal definition of noncharacters. Corrigendum #9 has been incorporated
into the core specification for Unicode 7.0.

"Q: Are noncharacters invalid in Unicode strings and UTFs?

"A: Absolutely not. Noncharacters do not cause a Unicode string to be ill-formed
in any UTF. This can be seen explicitly in the table above, where every
noncharacter code point has a well-formed representation in UTF-32, in UTF-16,
and in UTF-8. An implementation which converts noncharacter code points between
one UTF representation and another must preserve these values correctly. The
fact that they are called 'noncharacters' and are not intended for open
interchange does not mean that they are somehow illegal or invalid code points
which make strings containing them invalid."

Change-Id: I4fcc0156e3d2fd305a7c7bb0c7b3dbef846c9e64
Reviewed-on: https://gerrit.libreoffice.org/78598
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

11 files changed, 110 insertions, 50 deletions

diff --git a/sal/qa/rtl/textenc/rtl_textcvt.cxx b/sal/qa/rtl/textenc/rtl_textcvt.cxx
index 2f5359b32c77..af9ccca345e7 100644
--- a/sal/qa/rtl/textenc/rtl_textcvt.cxx
+++ b/sal/qa/rtl/textenc/rtl_textcvt.cxx
@@ -457,6 +457,8 @@ public:
 
     void testInvalidUtf8();
 
+    void testInvalidUnicode();
+
     void testSRCBUFFERTOSMALL();
 
     void testMime();
@@ -471,6 +473,7 @@ public:
     CPPUNIT_TEST(testComplexCut);
     CPPUNIT_TEST(testInvalidUtf7);
     CPPUNIT_TEST(testInvalidUtf8);
+    CPPUNIT_TEST(testInvalidUnicode);
     CPPUNIT_TEST(testSRCBUFFERTOSMALL);
     CPPUNIT_TEST(testMime);
     CPPUNIT_TEST(testWindows);
@@ -2336,6 +2339,15 @@ void Test::testComplex() {
               true,
               false,
               RTL_UNICODETOTEXT_FLAGS_UNDEFINED_ERROR },
+            { RTL_TEXTENCODING_UTF8,
+              RTL_CONSTASCII_STRINGPARAM("\xEF\xBF\xBF"),
+              {0xFFFF},
+              1,
+              false,
+              true,
+              true,
+              false,
+              RTL_UNICODETOTEXT_FLAGS_UNDEFINED_ERROR },
 
             // Test Java UTF-8:
 
@@ -2960,6 +2972,24 @@ void Test::testInvalidUtf8() {
     }
 }
 
+void Test::testInvalidUnicode() {
+    auto const converter = rtl_createUnicodeToTextConverter(RTL_TEXTENCODING_UTF8);
+    CPPUNIT_ASSERT(converter != nullptr);
+    sal_Unicode const input[] = {0xDC00}; // lone low surrogate
+    char buf[TEST_STRING_SIZE];
+    sal_uInt32 info;
+    sal_Size converted;
+    auto const size = rtl_convertUnicodeToText(
+        converter, nullptr, input, SAL_N_ELEMENTS(input), buf, TEST_STRING_SIZE,
+        (RTL_UNICODETOTEXT_FLAGS_UNDEFINED_ERROR | RTL_UNICODETOTEXT_FLAGS_INVALID_ERROR
+         | RTL_UNICODETOTEXT_FLAGS_FLUSH),
+        &info, &converted);
+    CPPUNIT_ASSERT_EQUAL(sal_Size(0), size);
+    CPPUNIT_ASSERT_EQUAL(RTL_UNICODETOTEXT_INFO_ERROR | RTL_UNICODETOTEXT_INFO_INVALID, info);
+    CPPUNIT_ASSERT_EQUAL(sal_Size(1), converted);
+    rtl_destroyTextToUnicodeConverter(converter);
+}
+
 void Test::testSRCBUFFERTOSMALL() {
     rtl_TextToUnicodeConverter cv = rtl_createTextToUnicodeConverter(
         RTL_TEXTENCODING_EUC_JP);
diff --git a/sal/textenc/convertbig5hkscs.cxx b/sal/textenc/convertbig5hkscs.cxx
index eaed0c7a36d2..77484666982e 100644
--- a/sal/textenc/convertbig5hkscs.cxx
+++ b/sal/textenc/convertbig5hkscs.cxx
@@ -21,6 +21,7 @@
 
 #include <cassert>
 
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 #include <sal/types.h>
 
@@ -333,6 +334,11 @@ sal_Size ImplConvertUnicodeToBig5Hkscs(void const * pData,
                 nHighSurrogate = static_cast<sal_Unicode>(nChar);
                 continue;
             }
+            else if (ImplIsLowSurrogate(nChar))
+            {
+                bUndefined = false;
+                goto bad_input;
+            }
         }
         else if (ImplIsLowSurrogate(nChar))
             nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
@@ -342,11 +348,7 @@ sal_Size ImplConvertUnicodeToBig5Hkscs(void const * pData,
             goto bad_input;
         }
 
-        if (ImplIsLowSurrogate(nChar) || ImplIsNoncharacter(nChar))
-        {
-            bUndefined = false;
-            goto bad_input;
-        }
+        assert(rtl::isUnicodeScalarValue(nChar));
 
         if (nChar < 0x80)
             if (pDestBufPtr != pDestBufEnd)
diff --git a/sal/textenc/converteuctw.cxx b/sal/textenc/converteuctw.cxx
index abc214402636..edb3c07fa934 100644
--- a/sal/textenc/converteuctw.cxx
+++ b/sal/textenc/converteuctw.cxx
@@ -19,6 +19,9 @@
 
 #include <sal/config.h>
 
+#include <cassert>
+
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 #include <sal/types.h>
 
@@ -342,6 +345,11 @@ sal_Size ImplConvertUnicodeToEucTw(void const * pData,
                 nHighSurrogate = static_cast<sal_Unicode>(nChar);
                 continue;
             }
+            else if (ImplIsLowSurrogate(nChar))
+            {
+                bUndefined = false;
+                goto bad_input;
+            }
         }
         else if (ImplIsLowSurrogate(nChar))
             nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
@@ -351,11 +359,7 @@ sal_Size ImplConvertUnicodeToEucTw(void const * pData,
             goto bad_input;
         }
 
-        if (ImplIsLowSurrogate(nChar) || ImplIsNoncharacter(nChar))
-        {
-            bUndefined = false;
-            goto bad_input;
-        }
+        assert(rtl::isUnicodeScalarValue(nChar));
 
         if (nChar < 0x80)
             if (pDestBufPtr != pDestBufEnd)
diff --git a/sal/textenc/convertgb18030.cxx b/sal/textenc/convertgb18030.cxx
index 9aa64970421b..88f5a999d87d 100644
--- a/sal/textenc/convertgb18030.cxx
+++ b/sal/textenc/convertgb18030.cxx
@@ -19,6 +19,9 @@
 
 #include <sal/config.h>
 
+#include <cassert>
+
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 #include <sal/types.h>
 
@@ -332,6 +335,11 @@ sal_Size ImplConvertUnicodeToGb18030(void const * pData,
                 nHighSurrogate = static_cast<sal_Unicode>(nChar);
                 continue;
             }
+            else if (ImplIsLowSurrogate(nChar))
+            {
+                bUndefined = false;
+                goto bad_input;
+            }
         }
         else if (ImplIsLowSurrogate(nChar))
             nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
@@ -341,11 +349,7 @@ sal_Size ImplConvertUnicodeToGb18030(void const * pData,
             goto bad_input;
         }
 
-        if (ImplIsLowSurrogate(nChar) || ImplIsNoncharacter(nChar))
-        {
-            bUndefined = false;
-            goto bad_input;
-        }
+        assert(rtl::isUnicodeScalarValue(nChar));
 
         if (nChar < 0x80)
             if (pDestBufPtr != pDestBufEnd)
diff --git a/sal/textenc/convertisciidevangari.cxx b/sal/textenc/convertisciidevangari.cxx
index 9e6583119252..b8566ed6a51e 100644
--- a/sal/textenc/convertisciidevangari.cxx
+++ b/sal/textenc/convertisciidevangari.cxx
@@ -7,10 +7,16 @@
  * file, You can obtain one at http://mozilla.org/MPL/2.0/.
  */
 
+#include <sal/config.h>
+
+#include <cassert>
+
 #include "converter.hxx"
 #include "unichars.hxx"
 #include "convertisciidevangari.hxx"
 #include "convertsinglebytetobmpunicode.hxx"
+
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 
 using namespace sal::detail::textenc;
@@ -261,6 +267,11 @@ sal_Size UnicodeToIsciiDevanagari::convert(sal_Unicode const* pSrcBuf, sal_Size
                 cHighSurrogate = static_cast< sal_Unicode >(c);
                 continue;
             }
+            else if (ImplIsLowSurrogate(c))
+            {
+                bUndefined = false;
+                goto bad_input;
+            }
         }
         else if (ImplIsLowSurrogate(c))
         {
@@ -271,11 +282,7 @@ sal_Size UnicodeToIsciiDevanagari::convert(sal_Unicode const* pSrcBuf, sal_Size
             bUndefined = false;
             goto bad_input;
         }
-        if (ImplIsLowSurrogate(c) || ImplIsNoncharacter(c))
-        {
-            bUndefined = false;
-            goto bad_input;
-        }
+        assert(rtl::isUnicodeScalarValue(c));
 
         //halant + halant     E8 E8  -> halant + ZWNJ   094D 200C
         //halant + nukta    E8 E9   halant + ZWJ    094D 200D
diff --git a/sal/textenc/convertiso2022cn.cxx b/sal/textenc/convertiso2022cn.cxx
index 76aee21f04b4..9e89c27486db 100644
--- a/sal/textenc/convertiso2022cn.cxx
+++ b/sal/textenc/convertiso2022cn.cxx
@@ -19,6 +19,9 @@
 
 #include <sal/config.h>
 
+#include <cassert>
+
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 #include <sal/types.h>
 
@@ -558,6 +561,11 @@ sal_Size ImplConvertUnicodeToIso2022Cn(void const * pData,
                 nHighSurrogate = static_cast<sal_Unicode>(nChar);
                 continue;
             }
+            else if (ImplIsLowSurrogate(nChar))
+            {
+                bUndefined = false;
+                goto bad_input;
+            }
         }
         else if (ImplIsLowSurrogate(nChar))
             nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
@@ -567,11 +575,7 @@ sal_Size ImplConvertUnicodeToIso2022Cn(void const * pData,
             goto bad_input;
         }
 
-        if (ImplIsLowSurrogate(nChar) || ImplIsNoncharacter(nChar))
-        {
-            bUndefined = false;
-            goto bad_input;
-        }
+        assert(rtl::isUnicodeScalarValue(nChar));
 
         if (nChar == 0x0A || nChar == 0x0D) // LF, CR
         {
diff --git a/sal/textenc/convertiso2022jp.cxx b/sal/textenc/convertiso2022jp.cxx
index 565c09ab36f5..ef2abeefcdb4 100644
--- a/sal/textenc/convertiso2022jp.cxx
+++ b/sal/textenc/convertiso2022jp.cxx
@@ -19,6 +19,9 @@
 
 #include <sal/config.h>
 
+#include <cassert>
+
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 #include <sal/types.h>
 
@@ -377,6 +380,11 @@ sal_Size ImplConvertUnicodeToIso2022Jp(void const * pData,
                 nHighSurrogate = static_cast<sal_Unicode>(nChar);
                 continue;
             }
+            else if (ImplIsLowSurrogate(nChar))
+            {
+                bUndefined = false;
+                goto bad_input;
+            }
         }
         else if (ImplIsLowSurrogate(nChar))
             nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
@@ -386,11 +394,7 @@ sal_Size ImplConvertUnicodeToIso2022Jp(void const * pData,
             goto bad_input;
         }
 
-        if (ImplIsLowSurrogate(nChar) || ImplIsNoncharacter(nChar))
-        {
-            bUndefined = false;
-            goto bad_input;
-        }
+        assert(rtl::isUnicodeScalarValue(nChar));
 
         if (nChar == 0x0A || nChar == 0x0D) // LF, CR
         {
diff --git a/sal/textenc/convertiso2022kr.cxx b/sal/textenc/convertiso2022kr.cxx
index 6169969812d2..5c7971ba5af6 100644
--- a/sal/textenc/convertiso2022kr.cxx
+++ b/sal/textenc/convertiso2022kr.cxx
@@ -19,6 +19,9 @@
 
 #include <sal/config.h>
 
+#include <cassert>
+
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 #include <sal/types.h>
 
@@ -355,6 +358,11 @@ sal_Size ImplConvertUnicodeToIso2022Kr(void const * pData,
                     nHighSurrogate = static_cast<sal_Unicode>(nChar);
                     continue;
                 }
+                else if (ImplIsLowSurrogate(nChar))
+                {
+                    bUndefined = false;
+                    goto bad_input;
+                }
             }
             else if (ImplIsLowSurrogate(nChar))
                 nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
@@ -364,11 +372,7 @@ sal_Size ImplConvertUnicodeToIso2022Kr(void const * pData,
                 goto bad_input;
             }
 
-            if (ImplIsLowSurrogate(nChar) || ImplIsNoncharacter(nChar))
-            {
-                bUndefined = false;
-                goto bad_input;
-            }
+            assert(rtl::isUnicodeScalarValue(nChar));
 
             if (nChar == 0x0A || nChar == 0x0D) // LF, CR
             {
diff --git a/sal/textenc/convertsinglebytetobmpunicode.cxx b/sal/textenc/convertsinglebytetobmpunicode.cxx
index b883411cbffa..6dc7891a3001 100644
--- a/sal/textenc/convertsinglebytetobmpunicode.cxx
+++ b/sal/textenc/convertsinglebytetobmpunicode.cxx
@@ -19,8 +19,10 @@
 
 #include <sal/config.h>
 
+#include <cassert>
 #include <cstddef>
 
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 #include <sal/types.h>
 
@@ -113,16 +115,18 @@ sal_Size rtl_textenc_convertBmpUnicodeToSingleByte(
                 highSurrogate = static_cast< sal_Unicode >(c);
                 continue;
             }
+            else if (ImplIsLowSurrogate(c))
+            {
+                undefined = false;
+                goto bad_input;
+            }
         } else if (ImplIsLowSurrogate(c)) {
             c = ImplCombineSurrogates(highSurrogate, c);
         } else {
             undefined = false;
             goto bad_input;
         }
-        if (ImplIsLowSurrogate(c) || ImplIsNoncharacter(c)) {
-            undefined = false;
-            goto bad_input;
-        }
+        assert(rtl::isUnicodeScalarValue(c));
         // Linearly searching through the ranges if probably fastest, assuming
         // that most converted characters belong to the ASCII subset:
         for (std::size_t i = 0; i < entries; ++i) {
diff --git a/sal/textenc/tcvtutf8.cxx b/sal/textenc/tcvtutf8.cxx
index a2430ff7c075..950d810e8b85 100644
--- a/sal/textenc/tcvtutf8.cxx
+++ b/sal/textenc/tcvtutf8.cxx
@@ -19,7 +19,10 @@
 
 #include <sal/config.h>
 
+#include <cassert>
+
 #include <sal/types.h>
+#include <rtl/character.hxx>
 #include <rtl/textcvt.h>
 
 #include "converter.hxx"
@@ -347,15 +350,17 @@ sal_Size ImplConvertUnicodeToUtf8(
                 nHighSurrogate = static_cast<sal_Unicode>(nChar);
                 continue;
             }
+            else if (ImplIsLowSurrogate(nChar) && !bJavaUtf8)
+            {
+                goto bad_input;
+            }
         }
         else if (ImplIsLowSurrogate(nChar) && !bJavaUtf8)
             nChar = ImplCombineSurrogates(nHighSurrogate, nChar);
         else
             goto bad_input;
 
-        if ((ImplIsLowSurrogate(nChar) && !bJavaUtf8)
-            || ImplIsNoncharacter(nChar))
-            goto bad_input;
+        assert(bJavaUtf8 ? nChar <= 0xFFFF : rtl::isUnicodeScalarValue(nChar));
 
         if (nChar <= 0x7F && (!bJavaUtf8 || nChar != 0))
             if (pDestBufPtr != pDestBufEnd)
diff --git a/sal/textenc/unichars.hxx b/sal/textenc/unichars.hxx
index 0bcd6f710518..e11e47493183 100644
--- a/sal/textenc/unichars.hxx
+++ b/sal/textenc/unichars.hxx
@@ -29,14 +29,6 @@
 
 #define RTL_TEXTENC_UNICODE_REPLACEMENT_CHARACTER 0xFFFD
 
-inline bool ImplIsNoncharacter(sal_uInt32 nUtf32)
-{
-    return (nUtf32 >= 0xFDD0 && nUtf32 <= 0xFDEF)
-           || (nUtf32 & 0xFFFF) >= 0xFFFE
-           || !rtl::isUnicodeCodePoint(nUtf32);
-}
-    // All code points that are noncharacters, as of Unicode 3.1.1.
-
 bool ImplIsControlOrFormat(sal_uInt32 nUtf32);
 
 inline bool ImplIsHighSurrogate(sal_uInt32 nUtf32)