Simplify utf8toutf16 functions

5 files changed, 30 insertions, 69 deletions

diff --git a/poppler/UTF.cc b/poppler/UTF.cc
index 6f67dccf..38fb0a28 100644
--- a/poppler/UTF.cc
+++ b/poppler/UTF.cc
@@ -430,14 +430,14 @@ int utf16CountUtf8Bytes(const uint16_t *utf16)
     return count;
 }
 
-int utf16ToUtf8(const uint16_t *utf16, int maxUtf16, char *utf8, int maxUtf8)
+std::string utf16ToUtf8(const uint16_t *utf16, int maxUtf16)
 {
     uint32_t codepoint = 0;
     uint32_t state = 0;
     int nIn = 0;
-    int nOut = 0;
-    char *p = utf8;
-    while (*utf16 && nIn < maxUtf16 && nOut < maxUtf8 - 1) {
+    char p[4];
+    std::string utf8;
+    while (*utf16 && nIn < maxUtf16) {
         decodeUtf16(&state, &codepoint, *utf16);
         if (state == UTF16_ACCEPT || state == UTF16_REJECT) {
             if (state == UTF16_REJECT || codepoint > UCS4_MAX) {
@@ -445,38 +445,17 @@ int utf16ToUtf8(const uint16_t *utf16, int maxUtf16, char *utf8, int maxUtf8)
                 state = 0;
             }
 
-            int bufSize = maxUtf8 - nOut;
-            int count = mapUTF8(codepoint, p, bufSize);
-            p += count;
-            nOut += count;
+            int count = mapUTF8(codepoint, p, 4);
+            utf8.append(std::string_view(p, count));
         }
         utf16++;
         nIn++;
     }
     // replace any trailing bytes too short for a valid UTF-8 with a replacement char
-    if (state != UTF16_ACCEPT && state != UTF16_REJECT && nOut < maxUtf8 - 1) {
-        int bufSize = maxUtf8 - nOut;
-        int count = mapUTF8(REPLACEMENT_CHAR, p, bufSize);
-        p += count;
-        nOut += count;
-        nOut++;
-    }
-    if (nOut > maxUtf8 - 1) {
-        nOut = maxUtf8 - 1;
+    if (state != UTF16_ACCEPT && state != UTF16_REJECT) {
+        int count = mapUTF8(REPLACEMENT_CHAR, p, 4);
+        utf8.append(std::string_view(p, count));
     }
-    utf8[nOut] = 0;
-    return nOut;
-}
-
-// Allocate utf8 string and convert utf16 into it.
-char *utf16ToUtf8(const uint16_t *utf16, int *len)
-{
-    const int n = utf16CountUtf8Bytes(utf16);
-    if (len) {
-        *len = n;
-    }
-    char *utf8 = (char *)gmalloc(n + 1);
-    utf16ToUtf8(utf16, INT_MAX, utf8, n + 1);
     return utf8;
 }
 
@@ -539,28 +518,25 @@ std::string TextStringToUtf8(const std::string &textStr)
 {
     int i, len;
     const char *s;
-    char *utf8;
+    std::string utf8;
 
     len = textStr.size();
     s = textStr.c_str();
     if (hasUnicodeByteOrderMark(textStr)) {
-        uint16_t *utf16;
+        std::vector<uint16_t> utf16;
         len = len / 2 - 1;
-        utf16 = new uint16_t[len + 1];
+        utf16.resize(len + 1);
         for (i = 0; i < len; i++) {
             utf16[i] = (s[2 + i * 2] & 0xff) << 8 | (s[3 + i * 2] & 0xff);
         }
         utf16[i] = 0;
-        utf8 = utf16ToUtf8(utf16);
-        delete[] utf16;
+        utf8 = utf16ToUtf8(utf16.data(), utf16.size());
     } else {
-        utf8 = (char *)gmalloc(len + 1);
+        utf8.resize(len + 1);
         for (i = 0; i < len; i++) {
             utf8[i] = pdfDocEncoding[s[i] & 0xff];
         }
         utf8[i] = 0;
     }
-    std::string utf8_string(utf8);
-    gfree(utf8);
-    return utf8_string;
+    return utf8;
 }
diff --git a/poppler/UTF.h b/poppler/UTF.h
index f09ce1cf..4486560f 100644
--- a/poppler/UTF.h
+++ b/poppler/UTF.h
@@ -134,14 +134,8 @@ int POPPLER_PRIVATE_EXPORT utf16CountUtf8Bytes(const uint16_t *utf16);
 //             maxUtf16 is set the the exact number of code units to convert.
 //  maxUtf16 - Maximum number of UTF-16 code units to convert. Conversion stops
 //             when either this count is reached or a null is encountered.
-//  utf8     - Output buffer to write the UTF-8 string to. Output will always be
-//             null terminated.
-//  maxUtf8  - Maximum size of the output buffer including space for null.
-// Returns number of UTF-8 bytes written (excluding NULL).
-int POPPLER_PRIVATE_EXPORT utf16ToUtf8(const uint16_t *utf16, int maxUtf16, char *utf8, int maxUtf8);
-
-// Allocate utf8 string and convert utf16 into it.
-char POPPLER_PRIVATE_EXPORT *utf16ToUtf8(const uint16_t *utf16, int *len = nullptr);
+// Returns utf8 string.
+std::string POPPLER_PRIVATE_EXPORT utf16ToUtf8(const uint16_t *utf16, int maxUtf16 = INT_MAX);
 
 // Convert a UCS-4 string to pure ASCII (7bit)
 //   in       - UCS-4 string bytes
diff --git a/qt5/tests/check_utf_conversion.cpp b/qt5/tests/check_utf_conversion.cpp
index 88a32359..05ef519e 100644
--- a/qt5/tests/check_utf_conversion.cpp
+++ b/qt5/tests/check_utf_conversion.cpp
@@ -79,14 +79,13 @@ void TestUTFConversion::testUTF_data()
 
 void TestUTFConversion::testUTF()
 {
-    char utf8Buf[1000];
-    char *utf8String;
+    std::string utf8String;
     uint16_t utf16Buf[1000];
     uint16_t *utf16String;
     int len;
 
     QFETCH(QString, s);
-    char *str = strdup(s.toUtf8().constData());
+    QByteArray str = s.toUtf8();
 
     // UTF-8 to UTF-16
 
@@ -111,17 +110,13 @@ void TestUTFConversion::testUTF()
 
     len = utf16CountUtf8Bytes(s.utf16());
     QCOMPARE(len, (int)strlen(str));
-    Q_ASSERT(len < (int)sizeof(utf8Buf)); // if this fails, make utf8Buf larger
 
-    len = utf16ToUtf8(s.utf16(), INT_MAX, utf8Buf, sizeof(utf8Buf));
-    QVERIFY(compare(utf8Buf, str));
+    utf8String = utf16ToUtf8(s.utf16(), INT_MAX);
+    QVERIFY(compare(utf8String.c_str(), str));
     QCOMPARE(len, (int)strlen(str));
 
     utf8String = utf16ToUtf8(s.utf16());
-    QVERIFY(compare(utf8String, str));
-    free(utf8String);
-
-    free(str);
+    QVERIFY(compare(utf8String.c_str(), str));
 }
 
 void TestUTFConversion::testUnicodeToAscii7()
diff --git a/qt6/tests/check_utf_conversion.cpp b/qt6/tests/check_utf_conversion.cpp
index 62ce65eb..53fe4bec 100644
--- a/qt6/tests/check_utf_conversion.cpp
+++ b/qt6/tests/check_utf_conversion.cpp
@@ -77,14 +77,13 @@ void TestUTFConversion::testUTF_data()
 
 void TestUTFConversion::testUTF()
 {
-    char utf8Buf[1000];
-    char *utf8String;
+    std::string utf8String;
     uint16_t utf16Buf[1000];
     uint16_t *utf16String;
     int len;
 
     QFETCH(QString, s);
-    char *str = strdup(s.toUtf8().constData());
+    QByteArray str = s.toUtf8().constData();
 
     // UTF-8 to UTF-16
 
@@ -109,17 +108,13 @@ void TestUTFConversion::testUTF()
 
     len = utf16CountUtf8Bytes(s.utf16());
     QCOMPARE(len, (int)strlen(str));
-    Q_ASSERT(len < (int)sizeof(utf8Buf)); // if this fails, make utf8Buf larger
 
-    len = utf16ToUtf8(s.utf16(), INT_MAX, utf8Buf, sizeof(utf8Buf));
-    QVERIFY(compare(utf8Buf, str));
+    utf8String = utf16ToUtf8(s.utf16(), INT_MAX);
+    QVERIFY(compare(utf8String.c_str(), str));
     QCOMPARE(len, (int)strlen(str));
 
     utf8String = utf16ToUtf8(s.utf16());
-    QVERIFY(compare(utf8String, str));
-    free(utf8String);
-
-    free(str);
+    QVERIFY(compare(utf8String.c_str(), str));
 }
 
 void TestUTFConversion::testUnicodeToAscii7()
diff --git a/utils/Win32Console.cc b/utils/Win32Console.cc
index deac2458..181d3bb7 100644
--- a/utils/Win32Console.cc
+++ b/utils/Win32Console.cc
@@ -120,7 +120,8 @@ Win32Console::Win32Console(int *argc, char **argv[])
         argList = new char *[numArgs];
         privateArgList = new char *[numArgs];
         for (int i = 0; i < numArgs; i++) {
-            argList[i] = utf16ToUtf8((uint16_t *)(wargv[i]));
+            std::string arg = utf16ToUtf8((uint16_t *)(wargv[i]));
+            argList[i] = strdup(arg.c_str());
             // parseArgs will rearrange the argv list so we keep our own copy
             // to use for freeing all the strings
             privateArgList[i] = argList[i];
@@ -155,7 +156,7 @@ Win32Console::~Win32Console()
     flush(true);
     if (argList) {
         for (int i = 0; i < numArgs; i++)
-            gfree(privateArgList[i]);
+            free(privateArgList[i]);
         delete[] argList;
         delete[] privateArgList;
     }