| 1 | /**************************************************************************** |
| 2 | ** |
| 3 | ** Copyright (C) 2015 The Qt Company Ltd. |
| 4 | ** Contact: http://www.qt.io/licensing/ |
| 5 | ** |
| 6 | ** This file is part of the QtVersit module of the Qt Toolkit. |
| 7 | ** |
| 8 | ** $QT_BEGIN_LICENSE:LGPL21$ |
| 9 | ** Commercial License Usage |
| 10 | ** Licensees holding valid commercial Qt licenses may use this file in |
| 11 | ** accordance with the commercial license agreement provided with the |
| 12 | ** Software or, alternatively, in accordance with the terms contained in |
| 13 | ** a written agreement between you and The Qt Company. For licensing terms |
| 14 | ** and conditions see http://www.qt.io/terms-conditions. For further |
| 15 | ** information use the contact form at http://www.qt.io/contact-us. |
| 16 | ** |
| 17 | ** GNU Lesser General Public License Usage |
| 18 | ** Alternatively, this file may be used under the terms of the GNU Lesser |
| 19 | ** General Public License version 2.1 or version 3 as published by the Free |
| 20 | ** Software Foundation and appearing in the file LICENSE.LGPLv21 and |
| 21 | ** LICENSE.LGPLv3 included in the packaging of this file. Please review the |
| 22 | ** following information to ensure the GNU Lesser General Public License |
| 23 | ** requirements will be met: https://www.gnu.org/licenses/lgpl.html and |
| 24 | ** http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html. |
| 25 | ** |
| 26 | ** As a special exception, The Qt Company gives you certain additional |
| 27 | ** rights. These rights are described in The Qt Company LGPL Exception |
| 28 | ** version 1.1, included in the file LGPL_EXCEPTION.txt in this package. |
| 29 | ** |
| 30 | ** $QT_END_LICENSE$ |
| 31 | ** |
| 32 | ****************************************************************************/ |
| 33 | |
| 34 | #ifndef QVERSITREADER_P_H |
| 35 | #define QVERSITREADER_P_H |
| 36 | |
| 37 | // |
| 38 | // W A R N I N G |
| 39 | // ------------- |
| 40 | // |
| 41 | // This file is not part of the Qt API. It exists purely as an |
| 42 | // implementation detail. This header file may change from version to |
| 43 | // version without notice, or even be removed. |
| 44 | // |
| 45 | // We mean it. |
| 46 | // |
| 47 | |
| 48 | #include <QtCore/qbytearray.h> |
| 49 | #include <QtCore/qbytearraymatcher.h> |
| 50 | #include <QtCore/qhash.h> |
| 51 | #include <QtCore/qlist.h> |
| 52 | #include <QtCore/qmutex.h> |
| 53 | #include <QtCore/qpair.h> |
| 54 | #include <QtCore/qpointer.h> |
| 55 | #include <QtCore/qscopedpointer.h> |
| 56 | #include <QtCore/qstack.h> |
| 57 | #include <QtCore/qthread.h> |
| 58 | |
| 59 | #include <QtVersit/qversitreader.h> |
| 60 | #include <QtVersit/qversitdocument.h> |
| 61 | #include <QtVersit/qversitproperty.h> |
| 62 | |
| 63 | QT_FORWARD_DECLARE_CLASS(QBuffer) |
| 64 | QT_FORWARD_DECLARE_CLASS(QIODevice) |
| 65 | QT_FORWARD_DECLARE_CLASS(QTextCodec) |
| 66 | |
| 67 | QT_BEGIN_NAMESPACE_VERSIT |
| 68 | |
| 69 | // The maximum number of bytes allowed to stay in memory after being read. The smaller this is, |
| 70 | // the more time spent moving bytes around. The larger it is, the more memory is wasted. |
| 71 | static const int MAX_OLD_BYTES_TO_KEEP = 8192; |
| 72 | |
| 73 | /* |
| 74 | * An LByteArray has a subset of QByteArray's interface, plus an efficient chopLeft function |
| 75 | * |
| 76 | * It stores a QByteArray internally, plus a marker of where it starts and where it ends. |
| 77 | */ |
| 78 | class LByteArray |
| 79 | { |
| 80 | public: |
| 81 | LByteArray() : mStart(0), mEnd(0) {} |
| 82 | explicit LByteArray(const QByteArray& d) :mData(d), mStart(0), mEnd(d.size()) {} |
| 83 | LByteArray(const QByteArray& d, int start, int end) :mData(d), mStart(start), mEnd(end) {} |
| 84 | bool isEmpty() const { |
| 85 | return mEnd <= mStart; |
| 86 | } |
| 87 | char at(int i) const { |
| 88 | return mData.at(i: mStart + i); |
| 89 | } |
| 90 | QByteArray toByteArray() const { |
| 91 | return mData.mid(index: mStart, len: mEnd-mStart); |
| 92 | } |
| 93 | /* Removes \a n bytes from the start of the QByteArray. */ |
| 94 | void chopLeft(int n) { |
| 95 | Q_ASSERT(size() >= n && n >= 0); |
| 96 | mStart += n; |
| 97 | } |
| 98 | QByteArray left(int n) { |
| 99 | Q_ASSERT(size() >= n && n >= 0); |
| 100 | return mData.mid(index: mStart, len: n); |
| 101 | } |
| 102 | int indexOf(const QByteArray& needle) { |
| 103 | int index = mData.indexOf(a: needle, from: mStart) - mStart; |
| 104 | if (index < size()) |
| 105 | return index; |
| 106 | return -1; |
| 107 | } |
| 108 | int size() const { |
| 109 | return mEnd - mStart; |
| 110 | } |
| 111 | const char* constData() const { |
| 112 | return mData.constData() + mStart; |
| 113 | } |
| 114 | bool contains(const QByteArray& ba) const { |
| 115 | int i = mData.indexOf(a: ba, from: mStart); |
| 116 | return i > 0 && i <= mEnd - ba.length(); |
| 117 | } |
| 118 | bool endsWith(const QByteArray& ba) const { |
| 119 | // Loop backwards from ba and from mData (starting from index mEnd) |
| 120 | if (ba.size() > size()) |
| 121 | return false; |
| 122 | return memcmp(s1: mData.constData()+mEnd-ba.size(), s2: ba.constData(), n: ba.size()) == 0; |
| 123 | } |
| 124 | LByteArray& operator=(const QByteArray& ba) { |
| 125 | mData = ba; |
| 126 | mStart = 0; |
| 127 | mEnd = mData.size(); |
| 128 | return *this; |
| 129 | } |
| 130 | bool operator==(const QByteArray& ba) { |
| 131 | return toByteArray() == ba; |
| 132 | } |
| 133 | bool operator!=(const QByteArray& ba) { |
| 134 | return toByteArray() != ba; |
| 135 | } |
| 136 | |
| 137 | private: |
| 138 | /* Clears the memory of bytes before the start marker */ |
| 139 | void dropOldData() { |
| 140 | if (mStart > MAX_OLD_BYTES_TO_KEEP && mEnd >= mStart) { |
| 141 | mData.remove(index: 0, len: mStart); |
| 142 | mEnd -= mStart; |
| 143 | mStart = 0; |
| 144 | } |
| 145 | } |
| 146 | void setBounds(int start, int end) { |
| 147 | mStart = start; |
| 148 | mEnd = end; |
| 149 | } |
| 150 | QByteArray mData; |
| 151 | int mStart; |
| 152 | int mEnd; |
| 153 | friend class LineReader; |
| 154 | }; |
| 155 | |
| 156 | class Q_VERSIT_EXPORT LineReader |
| 157 | { |
| 158 | public: |
| 159 | LineReader(QIODevice* device, QTextCodec* codec); |
| 160 | LineReader(QIODevice* device); |
| 161 | LineReader(QIODevice* device, QTextCodec* codec, int chunkSize); |
| 162 | void init(); |
| 163 | void pushLine(const QByteArray& line); |
| 164 | int odometer() const; |
| 165 | bool atEnd() const; |
| 166 | QTextCodec* codec() const; |
| 167 | bool isCodecCertain() const; |
| 168 | bool isCodecUtf8Compatible() const; |
| 169 | void setCodecUtf8Incompatible(); |
| 170 | LByteArray readLine(); |
| 171 | |
| 172 | private: |
| 173 | void readOneLine(LByteArray* cursor); |
| 174 | bool tryReadLine(LByteArray* cursor, bool atEnd); |
| 175 | |
| 176 | QIODevice* const mDevice; |
| 177 | QTextCodec* mCodec; |
| 178 | bool mIsCodecCertain; |
| 179 | bool mIsCodecUtf8Compatible; |
| 180 | int mChunkSize; // How many bytes to read in one go. |
| 181 | QList<QByteArrayMatcher> mCrlfList; |
| 182 | QStack<QByteArray> mPushedLines; // Stores a lines that has been "pushed" in front by pushLine |
| 183 | LByteArray mBuffer; |
| 184 | int mOdometer; |
| 185 | int mSearchFrom; |
| 186 | }; |
| 187 | |
| 188 | class Q_VERSIT_EXPORT QVersitReaderPrivate : public QThread |
| 189 | { |
| 190 | Q_OBJECT |
| 191 | |
| 192 | public: // Constructors and destructor |
| 193 | QVersitReaderPrivate(); |
| 194 | ~QVersitReaderPrivate(); |
| 195 | |
| 196 | static QHash<QPair<QVersitDocument::VersitType,QString>, QVersitProperty::ValueType>* |
| 197 | valueTypeMap(); |
| 198 | void init(QVersitReader* reader); |
| 199 | |
| 200 | signals: |
| 201 | void stateChanged(QVersitReader::State state); |
| 202 | void resultsAvailable(); |
| 203 | |
| 204 | protected: // From QThread |
| 205 | void run(); |
| 206 | |
| 207 | public: // New functions |
| 208 | void read(); |
| 209 | |
| 210 | // mutexed getters and setters. |
| 211 | void setState(QVersitReader::State); |
| 212 | QVersitReader::State state() const; |
| 213 | void setError(QVersitReader::Error); |
| 214 | QVersitReader::Error error() const; |
| 215 | void setCanceling(bool cancelling); |
| 216 | bool isCanceling(); |
| 217 | |
| 218 | bool parseVersitDocument(LineReader* lineReader, QVersitDocument* document); |
| 219 | bool parseVersitDocumentBody(LineReader* lineReader, QVersitDocument* document); |
| 220 | |
| 221 | QVersitProperty parseNextVersitProperty( |
| 222 | QVersitDocument::VersitType versitType, |
| 223 | LineReader* lineReader); |
| 224 | |
| 225 | void parseVCard21Property( |
| 226 | LByteArray* text, |
| 227 | QVersitProperty* property, |
| 228 | LineReader* lineReader); |
| 229 | |
| 230 | void parseVCard30Property( |
| 231 | QVersitDocument::VersitType versitType, |
| 232 | LByteArray* text, |
| 233 | QVersitProperty* property, |
| 234 | LineReader* lineReader); |
| 235 | |
| 236 | bool setVersionFromProperty( |
| 237 | QVersitDocument* document, |
| 238 | const QVersitProperty& property) const; |
| 239 | |
| 240 | bool unencode( |
| 241 | QByteArray* value, |
| 242 | QVersitProperty* property, |
| 243 | LineReader* lineReader) const; |
| 244 | |
| 245 | QString decodeCharset( |
| 246 | const QByteArray& value, |
| 247 | QVersitProperty* property, |
| 248 | LineReader* lineReader, |
| 249 | QTextCodec** codec) const; |
| 250 | |
| 251 | void decodeQuotedPrintable(QByteArray* text) const; |
| 252 | |
| 253 | |
| 254 | /* These functions operate on a cursor describing a single line */ |
| 255 | QPair<QStringList,QString> extractPropertyGroupsAndName(LByteArray* line, QTextCodec* codec) |
| 256 | const; |
| 257 | QMultiHash<QString,QString> (LByteArray* line, QTextCodec* codec) |
| 258 | const; |
| 259 | QMultiHash<QString,QString> (LByteArray* line, QTextCodec* codec) |
| 260 | const; |
| 261 | |
| 262 | // "Private" functions |
| 263 | QList<QByteArray> (LByteArray* line, QTextCodec *codec) const; |
| 264 | QList<QByteArray> (const QByteArray& text, const QByteArray& separator, |
| 265 | QTextCodec *codec) const; |
| 266 | QByteArray (const QByteArray& text, int startPosition, int length=-1) const; |
| 267 | QString paramName(const QByteArray& parameter, QTextCodec* codec) const; |
| 268 | QString paramValue(const QByteArray& parameter, QTextCodec* codec) const; |
| 269 | template <class T> static bool containsAt(const T& text, const QByteArray& ba, int index); |
| 270 | bool splitStructuredValue(QVersitProperty* property, |
| 271 | bool hasEscapedBackslashes) const; |
| 272 | static QStringList splitValue(const QString& string, |
| 273 | const QChar& sep, |
| 274 | QString::SplitBehavior behaviour, |
| 275 | bool hasEscapedBackslashes); |
| 276 | static void removeBackSlashEscaping(QString* text); |
| 277 | |
| 278 | // Data |
| 279 | public: |
| 280 | QPointer<QIODevice> mIoDevice; |
| 281 | QScopedPointer<QBuffer> mInputBytes; // Holds the data set by setData() |
| 282 | QList<QVersitDocument> mVersitDocuments; |
| 283 | int mDocumentNestingLevel; // Depth in parsing nested Versit documents |
| 284 | QTextCodec* mDefaultCodec; |
| 285 | QVersitReader::State mState; |
| 286 | QVersitReader::Error mError; |
| 287 | bool mIsCanceling; |
| 288 | mutable QMutex mMutex; |
| 289 | |
| 290 | private: |
| 291 | /* key is the document type and property name, value is the type of property it is. |
| 292 | If there is no entry, assume it is a PlainType */ |
| 293 | static QHash<QPair<QVersitDocument::VersitType,QString>, QVersitProperty::ValueType>* mValueTypeMap; |
| 294 | }; |
| 295 | |
| 296 | QT_END_NAMESPACE_VERSIT |
| 297 | |
| 298 | #endif // QVERSITREADER_P_H |
| 299 | |