Changeset 561 for trunk/src/corelib/codecs/qtextcodec.cpp

Timestamp:

Feb 11, 2010, 11:19:06 PM (15 years ago)

Author:

Dmitry A. Kuminov

Message:

trunk: Merged in qt 4.6.1 sources.

Location:

trunk

Files:

• . (modified) (1 prop)
• src/corelib/codecs/qtextcodec.cpp (modified) (28 diffs)

trunk/src/corelib/codecs/qtextcodec.cpp

@@ -2 +2 @@
 **
 ** Copyright (C) 2009 Nokia Corporation and/or its subsidiary(-ies).
-** Contact: Qt Software Information ([email protected])
+** All rights reserved.
+** Contact: Nokia Corporation ([email protected])
 **
 ** This file is part of the QtCore module of the Qt Toolkit.
@@ -21 +22 @@
 ** will be met: http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
 **
-** In addition, as a special exception, Nokia gives you certain
-** additional rights. These rights are described in the Nokia Qt LGPL
-** Exception version 1.0, included in the file LGPL_EXCEPTION.txt in this
-** package.
+** In addition, as a special exception, Nokia gives you certain additional
+** rights.  These rights are described in the Nokia Qt LGPL Exception
+** version 1.1, included in the file LGPL_EXCEPTION.txt in this package.
 **
 ** GNU General Public License Usage
@@ -34 +34 @@
 ** met: http://www.gnu.org/copyleft/gpl.html.
 **
-** If you are unsure which license is appropriate for your use, please
-** contact the sales department at qt-sales@nokia.com.
+** If you 
+** @nokia.com.
 ** $QT_END_LICENSE$
 **
@@ -91 +91 @@
 #include <ctype.h>
 #include <locale.h>
-#if defined (_XOPEN_UNIX) && !defined(Q_OS_QNX6) && !defined(Q_OS_OSF)
+#if defined (_XOPEN_UNIX) && !defined(Q_OS_QNX) && !defined(Q_OS_OSF)
 #include <langinfo.h>
 #endif
@@ -98 +98 @@
 #  define QT_NO_SETLOCALE
 #endif
+
+
+
 
 QT_BEGIN_NAMESPACE
@@ -106 +109 @@
 #endif
 
+
+
+
+
 
 static bool nameMatch(const QByteArray &name, const QByteArray &test)
@@ -118 +125 @@
     // if the letters and numbers are the same, we have a match
     while (*n != '\0') {
-        if (isalnum((uchar)*n)) {
+        if (*n)) {
             for (;;) {
                 if (*h == '\0')
                     return false;
-                if (isalnum((uchar)*h))
+                if (*h))
                     break;
                 ++h;
             }
-            if (tolower((uchar)*n) != tolower((uchar)*h))
+            if (*h))
                 return false;
             ++h;
@@ -132 +139 @@
         ++n;
     }
-    while (*h && !isalnum((uchar)*h))
+    while (*h && !*h))
            ++h;
     return (*h == '\0');
@@ -172 +179 @@
 
 static QList<QTextCodec*> *all = 0;
+
 static bool destroying_is_ok = false;
+
 
 static QTextCodec *localeMapper = 0;
@@ -194 +203 @@
         return;
 
+
     destroying_is_ok = true;
-
-    while (all->size())
-        delete all->takeFirst();
+#endif
+
+    for (QList<QTextCodec *>::const_iterator it = all->constBegin()
+            ; it != all->constEnd(); ++it) {
+        delete *it;
+    }
     delete all;
     all = 0;
     localeMapper = 0;
 
+
     destroying_is_ok = false;
+
 }
 
@@ -240 +255 @@
 
     const int wclen_auto = 4096;
-    WCHAR wc_auto[wclen_auto];
+     wc_auto[wclen_auto];
     int wclen = wclen_auto;
-    WCHAR *wc = wc_auto;
+     *wc = wc_auto;
     int len;
     QString sp;
@@ -282 +297 @@
                 wclen = MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED,
                                     mb, mblen, 0, 0);
-                wc = new WCHAR[wclen];
+                wc = new [wclen];
                 // and try again...
             }
@@ -348 +363 @@
     QString s;
     while((next = CharNextExA(CP_ACP, mb, 0)) != mb) {
-        WCHAR wc[2] ={0};
+         wc[2] ={0};
         int charlength = next - mb;
         int len = MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED|MB_ERR_INVALID_CHARS, mb, charlength, wc, 2);
@@ -687 +702 @@
 //     "vi", "vi_VN", 0 };
 
-static bool try_locale_list(const char * const locale[], const char * lang)
+static bool try_locale_list(const char * const locale[], const lang)
 {
     int i;
-    for(i=0; locale[i] && *locale[i] && strcmp(locale[i], lang); i++)
+    for(i=0; locale[i] && ; i++)
         ;
     return locale[i] != 0;
@@ -748 +763 @@
 
 #if !defined(Q_OS_WIN32) && !defined(Q_OS_WINCE) && !defined(Q_OS_OS2)
-static QTextCodec *checkForCodec(const char *name) {
+static QTextCodec *checkForCodec(const name) {
     QTextCodec *c = QTextCodec::codecForName(name);
     if (!c) {
-        const char *at = strchr(name, '@');
-        if (at) {
-            QByteArray n(name, at - name);
-            c = QTextCodec::codecForName(n.data());
+        const int index = name.indexOf('@');
+        if (index != -1) {
+            c = QTextCodec::codecForName(name.left(index));
         }
     }
@@ -776 +790 @@
 #endif
 
-#if defined (_XOPEN_UNIX) && !defined(Q_OS_QNX6) && !defined(Q_OS_OSF)
+#if defined (_XOPEN_UNIX) && !defined(Q_OS_QNX) && !defined(Q_OS_OSF)
     if (!localeMapper) {
         char *charset = nl_langinfo (CODESET);
@@ -797 +811 @@
         // to fall back to environment variables.
 #if !defined(QT_NO_SETLOCALE)
-        char * ctype = qstrdup(setlocale(LC_CTYPE, 0));
+        c);
 #else
-        char * ctype = qstrdup("");
+        c;
 #endif
 
         // Get the first nonempty value from $LC_ALL, $LC_CTYPE, and $LANG
         // environment variables.
-        char * lang = qstrdup(qgetenv("LC_ALL").constData());
-        if (!lang || lang[0] == 0 || strcmp(lang, "C") == 0) {
-            if (lang) delete [] lang;
-            lang = qstrdup(qgetenv("LC_CTYPE").constData());
+        QByteArray lang = qgetenv("LC_ALL");
+        if (lang.isEmpty() || lang == "C") {
+            lang = qgetenv("LC_CTYPE");
         }
-        if (!lang || lang[0] == 0 || strcmp(lang, "C") == 0) {
-            if (lang) delete [] lang;
-            lang = qstrdup(qgetenv("LANG").constData());
+        if (lang.isEmpty() || lang == "C") {
+            lang = qgetenv("LANG");
         }
 
@@ -824 +836 @@
 
         // 1. CODESET from ctype if it contains a .CODESET part (e.g. en_US.ISO8859-15)
-        char * codeset = ctype ? strchr(ctype, '.') : 0;
-        if (codeset && *codeset == '.')
-            localeMapper = checkForCodec(codeset + 1);
+        ;
+        if ()
+            localeMapper = checkForCodec();
 
         // 2. CODESET from lang if it contains a .CODESET part
-        codeset = lang ? strchr(lang, '.') : 0;
-        if (!localeMapper && codeset && *codeset == '.')
-            localeMapper = checkForCodec(codeset + 1);
+        if (!localeMapper) {
+            indexOfDot = lang.indexOf('.');
+            if (indexOfDot != -1)
+                localeMapper = checkForCodec( lang.mid(indexOfDot + 1) );
+        }
 
         // 3. ctype (maybe the locale is named "ISO-8859-1" or something)
-        if (!localeMapper && ctype && *ctype != 0 && strcmp (ctype, "C") != 0)
+        if (!localeMapper && )
             localeMapper = checkForCodec(ctype);
 
         // 4. locale (ditto)
-        if (!localeMapper && lang && *lang != 0)
+        if (!localeMapper && )
             localeMapper = checkForCodec(lang);
 
         // 5. "@euro"
-        if ((!localeMapper && ctype && strstr(ctype, "@euro")) || (lang && strstr(lang, "@euro")))
+        if ((!localeMapper && ctype))
             localeMapper = checkForCodec("ISO 8859-15");
 
         // 6. guess locale from ctype unless ctype is "C"
         // 7. guess locale from lang
-        char * try_by_name = ctype;
-        if (ctype && *ctype != 0 && strcmp (ctype, "C") != 0)
-            try_by_name = lang;
+        const QByteArray &try_by_name = (!ctype.isEmpty() && ctype != "C") ? lang : ctype;
 
         // Now do the guessing.
-        if (lang && *lang && !localeMapper && try_by_name && *try_by_name) {
+        if () {
             if (try_locale_list(iso8859_15locales, lang))
                 localeMapper = QTextCodec::codecForName("ISO 8859-15");
@@ -885 +897 @@
         }
 
-        delete [] ctype;
-        delete [] lang;
     }
 
@@ -906 +916 @@
         return;
 
+
     if (destroying_is_ok)
         qWarning("QTextCodec: Creating new codec during codec cleanup");
+
     all = new QList<QTextCodec*>;
     // create the cleanup object to cleanup all codecs on exit
@@ -944 +956 @@
     (void)new QSjisCodec;
     (void)new QEucKrCodec;
+
     (void)new QBig5Codec;
     (void)new QBig5hkscsCodec;
@@ -979 +992 @@
 }
 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
 QTextCodec::ConverterState::~ConverterState()
 {
@@ -1131 +1164 @@
 
 /*!
-    \enum QTextCodec::ConversionFlag
-
-    \value DefaultConversion  No flag is set.
-    \value ConvertInvalidToNull  If this flag is set, each invalid input
-                                 character is output as a null character.
-    \value IgnoreHeader  Ignore any Unicode byte-order mark and don't generate any.
-
-    \omitvalue FreeFunction
-*/
-
-/*!
-    \fn QTextCodec::ConverterState::ConverterState(ConversionFlags flags)
-
-    Constructs a ConverterState object initialized with the given \a flags.
-*/
-
-/*!
-    \fn QTextCodec::ConverterState::~ConverterState()
-
-    Destroys the ConverterState object.
-*/
-
-/*!
     \nonreentrant
 
@@ -1176 +1186 @@
 QTextCodec::~QTextCodec()
 {
+
     if (!destroying_is_ok)
         qWarning("QTextCodec::~QTextCodec: Called by application");
+
     if (all)
         all->removeAll(this);
@@ -1308 +1320 @@
     own mechanism for setting the locale.
 
-    Setting this codec is not supported on DOS based Windows.
-
     \sa codecForLocale()
 */
 void QTextCodec::setCodecForLocale(QTextCodec *c)
 {
-#ifdef Q_WS_WIN
-    if (QSysInfo::WindowsVersion& QSysInfo::WV_DOS_based)
-        return;
-#endif
     codecForLocaleSet = true;
     localeMapper = c;
@@ -1770 +1776 @@
     \since 4.4
 
-    Tries to detect the encoding of the provided snippet of HTML in the given byte array, \a ba,
-    and returns a QTextCodec instance that is capable of decoding the html to unicode.
-    If the codec cannot be detected from the content provided, \a defaultCodec is returned.
+    Tries to detect the encoding of the provided snippet of HTML in
+    the given byte array, \a ba, by checking the BOM (Byte Order Mark)
+    and the content-type meta header and returns a QTextCodec instance
+    that is capable of decoding the html to unicode.  If the codec
+    cannot be detected from the content provided, \a defaultCodec is
+    returned.
+
+    \sa codecForUtfText()
 */
 QTextCodec *QTextCodec::codecForHtml(const QByteArray &ba, QTextCodec *defaultCodec)
@@ -1780 +1791 @@
     QTextCodec *c = 0;
 
-    if (ba.size() > 1 && (((uchar)ba[0] == 0xfe && (uchar)ba[1] == 0xff)
-                          || ((uchar)ba[0] == 0xff && (uchar)ba[1] == 0xfe))) {
-        c = QTextCodec::codecForMib(1015); // utf16
-    } else if (ba.size() > 2
-             && (uchar)ba[0] == 0xef
-             && (uchar)ba[1] == 0xbb
-             && (uchar)ba[2] == 0xbf) {
-        c = QTextCodec::codecForMib(106); // utf-8
-    } else {
+    c = QTextCodec::codecForUtfText(ba, c);
+    if (!c) {
         QByteArray header = ba.left(512).toLower();
         if ((pos = header.indexOf("http-equiv=")) != -1) {
-            pos = header.indexOf("charset=", pos) + int(strlen("charset="));
-            if (pos != -1) {
-                int pos2 = header.indexOf('\"', pos+1);
-                QByteArray cs = header.mid(pos, pos2-pos);
-                //            qDebug("found charset: %s", cs.data());
-                c = QTextCodec::codecForName(cs);
+            if ((pos = header.lastIndexOf("meta ", pos)) != -1) {
+                pos = header.indexOf("charset=", pos) + int(strlen("charset="));
+                if (pos != -1) {
+                    int pos2 = header.indexOf('\"', pos+1);
+                    QByteArray cs = header.mid(pos, pos2-pos);
+                    //            qDebug("found charset: %s", cs.data());
+                    c = QTextCodec::codecForName(cs);
+                }
             }
         }
@@ -1807 +1813 @@
 
 /*!
-  \overload
-
-  If the codec cannot be detected, this overload returns a Latin-1 QTextCodec.
+    \overload
+
+    Tries to detect the encoding of the provided snippet of HTML in
+    the given byte array, \a ba, by checking the BOM (Byte Order Mark)
+    and the content-type meta header and returns a QTextCodec instance
+    that is capable of decoding the html to unicode. If the codec cannot
+    be detected, this overload returns a Latin-1 QTextCodec.
 */
 QTextCodec *QTextCodec::codecForHtml(const QByteArray &ba)
 {
     return codecForHtml(ba, QTextCodec::codecForMib(/*Latin 1*/ 4));
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
 }