| 1 | /**************************************************************************** |
| 2 | ** |
| 3 | ** Copyright (C) 2016 The Qt Company Ltd. |
| 4 | ** Contact: https://www.qt.io/licensing/ |
| 5 | ** |
| 6 | ** This file is part of the QtGui module of the Qt Toolkit. |
| 7 | ** |
| 8 | ** $QT_BEGIN_LICENSE:LGPL$ |
| 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 https://www.qt.io/terms-conditions. For further |
| 15 | ** information use the contact form at https://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 3 as published by the Free Software |
| 20 | ** Foundation and appearing in the file LICENSE.LGPL3 included in the |
| 21 | ** packaging of this file. Please review the following information to |
| 22 | ** ensure the GNU Lesser General Public License version 3 requirements |
| 23 | ** will be met: https://www.gnu.org/licenses/lgpl-3.0.html. |
| 24 | ** |
| 25 | ** GNU General Public License Usage |
| 26 | ** Alternatively, this file may be used under the terms of the GNU |
| 27 | ** General Public License version 2.0 or (at your option) the GNU General |
| 28 | ** Public license version 3 or any later version approved by the KDE Free |
| 29 | ** Qt Foundation. The licenses are as published by the Free Software |
| 30 | ** Foundation and appearing in the file LICENSE.GPL2 and LICENSE.GPL3 |
| 31 | ** included in the packaging of this file. Please review the following |
| 32 | ** information to ensure the GNU General Public License requirements will |
| 33 | ** be met: https://www.gnu.org/licenses/gpl-2.0.html and |
| 34 | ** https://www.gnu.org/licenses/gpl-3.0.html. |
| 35 | ** |
| 36 | ** $QT_END_LICENSE$ |
| 37 | ** |
| 38 | ****************************************************************************/ |
| 39 | |
| 40 | #ifndef QTEXTHTMLPARSER_P_H |
| 41 | #define QTEXTHTMLPARSER_P_H |
| 42 | |
| 43 | // |
| 44 | // W A R N I N G |
| 45 | // ------------- |
| 46 | // |
| 47 | // This file is not part of the Qt API. It exists purely as an |
| 48 | // implementation detail. This header file may change from version to |
| 49 | // version without notice, or even be removed. |
| 50 | // |
| 51 | // We mean it. |
| 52 | // |
| 53 | |
| 54 | #include <QtGui/private/qtguiglobal_p.h> |
| 55 | #include "QtGui/qbrush.h" |
| 56 | #include "QtGui/qcolor.h" |
| 57 | #include "QtGui/qfont.h" |
| 58 | #include "QtGui/qtextdocument.h" |
| 59 | #include "QtGui/qtextcursor.h" |
| 60 | |
| 61 | #include "QtCore/qlist.h" |
| 62 | |
| 63 | #include "private/qtextformat_p.h" |
| 64 | #include "private/qtextdocument_p.h" |
| 65 | #include "private/qcssparser_p.h" |
| 66 | |
| 67 | #ifndef QT_NO_TEXTHTMLPARSER |
| 68 | |
| 69 | QT_BEGIN_NAMESPACE |
| 70 | |
| 71 | enum QTextHTMLElements { |
| 72 | Html_unknown = -1, |
| 73 | Html_qt = 0, |
| 74 | Html_body, |
| 75 | |
| 76 | Html_a, |
| 77 | Html_em, |
| 78 | Html_i, |
| 79 | Html_big, |
| 80 | Html_small, |
| 81 | Html_strong, |
| 82 | Html_b, |
| 83 | Html_cite, |
| 84 | Html_address, |
| 85 | Html_var, |
| 86 | Html_dfn, |
| 87 | |
| 88 | Html_h1, |
| 89 | Html_h2, |
| 90 | Html_h3, |
| 91 | Html_h4, |
| 92 | Html_h5, |
| 93 | Html_h6, |
| 94 | Html_p, |
| 95 | Html_center, |
| 96 | |
| 97 | Html_font, |
| 98 | |
| 99 | Html_ul, |
| 100 | Html_ol, |
| 101 | Html_li, |
| 102 | |
| 103 | Html_code, |
| 104 | Html_tt, |
| 105 | Html_kbd, |
| 106 | Html_samp, |
| 107 | |
| 108 | Html_img, |
| 109 | Html_br, |
| 110 | Html_hr, |
| 111 | |
| 112 | Html_sub, |
| 113 | Html_sup, |
| 114 | |
| 115 | Html_pre, |
| 116 | Html_blockquote, |
| 117 | Html_head, |
| 118 | Html_div, |
| 119 | Html_span, |
| 120 | Html_dl, |
| 121 | Html_dt, |
| 122 | Html_dd, |
| 123 | Html_u, |
| 124 | Html_s, |
| 125 | Html_nobr, |
| 126 | |
| 127 | // tables |
| 128 | Html_table, |
| 129 | Html_tr, |
| 130 | Html_td, |
| 131 | Html_th, |
| 132 | Html_thead, |
| 133 | Html_tbody, |
| 134 | , |
| 135 | Html_caption, |
| 136 | |
| 137 | // misc... |
| 138 | Html_html, |
| 139 | Html_style, |
| 140 | Html_title, |
| 141 | Html_meta, |
| 142 | Html_link, |
| 143 | Html_script, |
| 144 | |
| 145 | Html_NumElements |
| 146 | }; |
| 147 | |
| 148 | struct QTextHtmlElement |
| 149 | { |
| 150 | const char name[11]; |
| 151 | QTextHTMLElements id; |
| 152 | enum DisplayMode { DisplayBlock, DisplayInline, DisplayTable, DisplayNone } displayMode; |
| 153 | }; |
| 154 | |
| 155 | class QTextHtmlParser; |
| 156 | |
| 157 | struct QTextHtmlParserNode { |
| 158 | enum WhiteSpaceMode { |
| 159 | WhiteSpaceNormal, |
| 160 | WhiteSpacePre, |
| 161 | WhiteSpaceNoWrap, |
| 162 | WhiteSpacePreWrap, |
| 163 | WhiteSpacePreLine, |
| 164 | WhiteSpaceModeUndefined = -1 |
| 165 | }; |
| 166 | |
| 167 | QTextHtmlParserNode(); |
| 168 | QString tag; |
| 169 | QString text; |
| 170 | QStringList attributes; |
| 171 | int parent; |
| 172 | QList<int> children; |
| 173 | QTextHTMLElements id; |
| 174 | QTextCharFormat charFormat; |
| 175 | QTextBlockFormat blockFormat; |
| 176 | uint cssFloat : 2; |
| 177 | uint hasOwnListStyle : 1; |
| 178 | uint hasOwnLineHeightType : 1; |
| 179 | uint hasLineHeightMultiplier : 1; |
| 180 | uint hasCssListIndent : 1; |
| 181 | uint isEmptyParagraph : 1; |
| 182 | uint isTextFrame : 1; |
| 183 | uint isRootFrame : 1; |
| 184 | uint displayMode : 3; // QTextHtmlElement::DisplayMode |
| 185 | uint hasHref : 1; |
| 186 | QTextListFormat::Style listStyle; |
| 187 | QString textListNumberPrefix; |
| 188 | QString textListNumberSuffix; |
| 189 | QString imageName; |
| 190 | QString imageAlt; |
| 191 | qreal imageWidth; |
| 192 | qreal imageHeight; |
| 193 | QTextLength width; |
| 194 | QTextLength height; |
| 195 | qreal tableBorder; |
| 196 | int tableCellRowSpan; |
| 197 | int tableCellColSpan; |
| 198 | qreal tableCellSpacing; |
| 199 | qreal tableCellPadding; |
| 200 | qreal tableCellBorder[4]; |
| 201 | QBrush tableCellBorderBrush[4]; |
| 202 | QTextFrameFormat::BorderStyle tableCellBorderStyle[4]; |
| 203 | QBrush borderBrush; |
| 204 | QTextFrameFormat::BorderStyle borderStyle; |
| 205 | bool borderCollapse; |
| 206 | int userState; |
| 207 | |
| 208 | int cssListIndent; |
| 209 | |
| 210 | WhiteSpaceMode wsm; |
| 211 | |
| 212 | inline bool isListStart() const |
| 213 | { return id == Html_ol || id == Html_ul; } |
| 214 | inline bool isTableCell() const |
| 215 | { return id == Html_td || id == Html_th; } |
| 216 | inline bool isBlock() const |
| 217 | { return displayMode == QTextHtmlElement::DisplayBlock; } |
| 218 | |
| 219 | inline bool isNotSelfNesting() const |
| 220 | { return id == Html_p || id == Html_li; } |
| 221 | |
| 222 | inline bool allowedInContext(int parentId) const |
| 223 | { |
| 224 | switch (id) { |
| 225 | case Html_dd: |
| 226 | case Html_dt: return (parentId == Html_dl); |
| 227 | case Html_tr: return (parentId == Html_table |
| 228 | || parentId == Html_thead |
| 229 | || parentId == Html_tbody |
| 230 | || parentId == Html_tfoot |
| 231 | ); |
| 232 | case Html_th: |
| 233 | case Html_td: return (parentId == Html_tr); |
| 234 | case Html_thead: |
| 235 | case Html_tbody: |
| 236 | case Html_tfoot: return (parentId == Html_table); |
| 237 | case Html_caption: return (parentId == Html_table); |
| 238 | case Html_body: return parentId != Html_head; |
| 239 | default: break; |
| 240 | } |
| 241 | return true; |
| 242 | } |
| 243 | |
| 244 | inline bool mayNotHaveChildren() const |
| 245 | { return id == Html_img || id == Html_hr || id == Html_br || id == Html_meta; } |
| 246 | |
| 247 | void initializeProperties(const QTextHtmlParserNode *parent, const QTextHtmlParser *parser); |
| 248 | |
| 249 | inline int uncollapsedMargin(int mar) const { return margin[mar]; } |
| 250 | |
| 251 | bool isNestedList(const QTextHtmlParser *parser) const; |
| 252 | |
| 253 | void parseStyleAttribute(const QString &value, const QTextDocument *resourceProvider); |
| 254 | |
| 255 | #if QT_CONFIG(cssparser) |
| 256 | void applyCssDeclarations(const QList<QCss::Declaration> &declarations, |
| 257 | const QTextDocument *resourceProvider); |
| 258 | |
| 259 | void setListStyle(const QList<QCss::Value> &cssValues); |
| 260 | # endif |
| 261 | |
| 262 | void applyForegroundImage(qint64 cacheKey, const QTextDocument *resourceProvider); |
| 263 | void applyBackgroundImage(const QString &url, const QTextDocument *resourceProvider); |
| 264 | |
| 265 | bool hasOnlyWhitespace() const; |
| 266 | |
| 267 | int margin[4]; |
| 268 | int padding[4]; |
| 269 | |
| 270 | friend class QTextHtmlParser; |
| 271 | }; |
| 272 | Q_DECLARE_TYPEINFO(QTextHtmlParserNode, Q_MOVABLE_TYPE); |
| 273 | |
| 274 | |
| 275 | class QTextHtmlParser |
| 276 | { |
| 277 | public: |
| 278 | enum Margin { |
| 279 | MarginTop, |
| 280 | MarginRight, |
| 281 | MarginBottom, |
| 282 | MarginLeft |
| 283 | }; |
| 284 | ~QTextHtmlParser() |
| 285 | { |
| 286 | qDeleteAll(nodes); |
| 287 | } |
| 288 | |
| 289 | inline const QTextHtmlParserNode &at(int i) const { return *nodes.at(i); } |
| 290 | inline QTextHtmlParserNode &operator[](int i) { return *nodes[i]; } |
| 291 | inline int count() const { return nodes.count(); } |
| 292 | inline int last() const { return nodes.count()-1; } |
| 293 | int depth(int i) const; |
| 294 | int topMargin(int i) const; |
| 295 | int bottomMargin(int i) const; |
| 296 | inline int leftMargin(int i) const { return margin(i, MarginLeft); } |
| 297 | inline int rightMargin(int i) const { return margin(i, MarginRight); } |
| 298 | |
| 299 | inline int topPadding(int i) const { return at(i).padding[MarginTop]; } |
| 300 | inline int bottomPadding(int i) const { return at(i).padding[MarginBottom]; } |
| 301 | inline int leftPadding(int i) const { return at(i).padding[MarginLeft]; } |
| 302 | inline int rightPadding(int i) const { return at(i).padding[MarginRight]; } |
| 303 | |
| 304 | inline qreal tableCellBorder(int i, int edge) const { return at(i).tableCellBorder[edge]; } |
| 305 | inline QTextFrameFormat::BorderStyle tableCellBorderStyle(int i, int edge) const { return at(i).tableCellBorderStyle[edge]; } |
| 306 | inline QBrush tableCellBorderBrush(int i, int edge) const { return at(i).tableCellBorderBrush[edge]; } |
| 307 | |
| 308 | void dumpHtml(); |
| 309 | |
| 310 | void parse(const QString &text, const QTextDocument *resourceProvider); |
| 311 | |
| 312 | static int lookupElement(const QString &element); |
| 313 | protected: |
| 314 | QTextHtmlParserNode *newNode(int parent); |
| 315 | QList<QTextHtmlParserNode *> nodes; |
| 316 | QString txt; |
| 317 | int pos, len; |
| 318 | |
| 319 | bool textEditMode; |
| 320 | |
| 321 | void parse(); |
| 322 | void parseTag(); |
| 323 | void parseCloseTag(); |
| 324 | void parseExclamationTag(); |
| 325 | QString parseEntity(); |
| 326 | QString parseWord(); |
| 327 | QTextHtmlParserNode *resolveParent(); |
| 328 | void resolveNode(); |
| 329 | QStringList parseAttributes(); |
| 330 | void applyAttributes(const QStringList &attributes); |
| 331 | void eatSpace(); |
| 332 | inline bool hasPrefix(QChar c, int lookahead = 0) const |
| 333 | {return pos + lookahead < len && txt.at(pos) == c; } |
| 334 | int margin(int i, int mar) const; |
| 335 | |
| 336 | bool nodeIsChildOf(int i, QTextHTMLElements id) const; |
| 337 | |
| 338 | |
| 339 | #if QT_CONFIG(cssparser) |
| 340 | QList<QCss::Declaration> declarationsForNode(int node) const; |
| 341 | void resolveStyleSheetImports(const QCss::StyleSheet &sheet); |
| 342 | void importStyleSheet(const QString &href); |
| 343 | |
| 344 | struct ExternalStyleSheet |
| 345 | { |
| 346 | inline ExternalStyleSheet() {} |
| 347 | inline ExternalStyleSheet(const QString &_url, const QCss::StyleSheet &_sheet) |
| 348 | : url(_url), sheet(_sheet) {} |
| 349 | QString url; |
| 350 | QCss::StyleSheet sheet; |
| 351 | }; |
| 352 | friend class QTypeInfo<ExternalStyleSheet>; |
| 353 | QList<ExternalStyleSheet> externalStyleSheets; |
| 354 | QList<QCss::StyleSheet> inlineStyleSheets; |
| 355 | # endif |
| 356 | |
| 357 | const QTextDocument *resourceProvider; |
| 358 | }; |
| 359 | #if QT_CONFIG(cssparser) |
| 360 | Q_DECLARE_TYPEINFO(QTextHtmlParser::ExternalStyleSheet, Q_MOVABLE_TYPE); |
| 361 | #endif |
| 362 | |
| 363 | QT_END_NAMESPACE |
| 364 | |
| 365 | #endif // QT_NO_TEXTHTMLPARSER |
| 366 | |
| 367 | #endif // QTEXTHTMLPARSER_P_H |
| 368 | |