Warning, file /office/tellico/src/utils/xmlhandler.cpp was not indexed or was modified since last indexation (in which case cross-reference links may be missing, inaccurate or erroneous).
0001 /*************************************************************************** 0002 Copyright (C) 2010-2020 Robby Stephenson <robby@periapsis.org> 0003 ***************************************************************************/ 0004 0005 /*************************************************************************** 0006 * * 0007 * This program is free software; you can redistribute it and/or * 0008 * modify it under the terms of the GNU General Public License as * 0009 * published by the Free Software Foundation; either version 2 of * 0010 * the License or (at your option) version 3 or any later version * 0011 * accepted by the membership of KDE e.V. (or its successor approved * 0012 * by the membership of KDE e.V.), which shall act as a proxy * 0013 * defined in Section 14 of version 3 of the license. * 0014 * * 0015 * This program is distributed in the hope that it will be useful, * 0016 * but WITHOUT ANY WARRANTY; without even the implied warranty of * 0017 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * 0018 * GNU General Public License for more details. * 0019 * * 0020 * You should have received a copy of the GNU General Public License * 0021 * along with this program. If not, see <http://www.gnu.org/licenses/>. * 0022 * * 0023 ***************************************************************************/ 0024 0025 #include "xmlhandler.h" 0026 #include "../tellico_debug.h" 0027 0028 #include <QRegularExpression> 0029 #include <QTextStream> 0030 #include <QXmlStreamReader> 0031 #include <QTextCodec> 0032 0033 using Tellico::XMLHandler; 0034 0035 bool XMLHandler::setUtf8XmlEncoding(QString& text_) { 0036 static const QRegularExpression rx(QLatin1String("encoding\\s*=\\s*\"([\\w-]+)\"")); 0037 QTextStream stream(&text_); 0038 // the xml header might still indicate an encoding other than utf-8 0039 // so read the first line and ensure it is set to utf-8 0040 QString firstLine = stream.readLine(); 0041 QRegularExpressionMatch match = rx.match(firstLine); 0042 if(match.hasMatch() && 0043 match.capturedRef(1).compare(QLatin1String("utf-8"), Qt::CaseInsensitive) != 0) { 0044 firstLine.replace(rx, QStringLiteral("encoding=\"utf-8\"")); 0045 text_ = firstLine + QLatin1Char('\n') + stream.readAll(); 0046 return true; 0047 } 0048 return false; 0049 } 0050 0051 QString XMLHandler::readXMLData(const QByteArray& data_) { 0052 // need to recognize encoding from the data, like QXmlInputSource::fromRawData() used to do 0053 QXmlStreamReader reader(data_); 0054 while(!reader.isStartDocument() && !reader.atEnd()) { 0055 reader.readNext(); 0056 } 0057 QStringRef enc = reader.documentEncoding(); 0058 if(enc.isEmpty() || enc.compare(QLatin1String("utf-8"), Qt::CaseInsensitive) == 0) { 0059 // default to utf8 and no need to parse to change embedded encoding 0060 return QString::fromUtf8(data_); 0061 } 0062 0063 QTextCodec* codec = QTextCodec::codecForName(enc.toUtf8()); 0064 if(!codec) { 0065 return QString::fromUtf8(data_); 0066 } 0067 QString text = codec->toUnicode(data_); 0068 // since we always process XML files as utf-8, make sure the embedded encoding is set to utf-8 0069 if(!setUtf8XmlEncoding(text)) { 0070 myDebug() << "Found non utf-8 encoding but did not change the embedded declaration" << enc; 0071 // output up to first 100 characters for debugging purposes 0072 myDebug() << text.left(100); 0073 } 0074 return text; 0075 }