| 1 | /****************************************************************************
|
|---|
| 2 | **
|
|---|
| 3 | ** Copyright (C) 2009 Nokia Corporation and/or its subsidiary(-ies).
|
|---|
| 4 | ** Contact: Qt Software Information ([email protected])
|
|---|
| 5 | **
|
|---|
| 6 | ** This file is part of the QtCore module of the Qt Toolkit.
|
|---|
| 7 | **
|
|---|
| 8 | ** $QT_BEGIN_LICENSE:LGPL$
|
|---|
| 9 | ** Commercial Usage
|
|---|
| 10 | ** Licensees holding valid Qt Commercial licenses may use this file in
|
|---|
| 11 | ** accordance with the Qt Commercial License Agreement provided with the
|
|---|
| 12 | ** Software or, alternatively, in accordance with the terms contained in
|
|---|
| 13 | ** a written agreement between you and Nokia.
|
|---|
| 14 | **
|
|---|
| 15 | ** GNU Lesser General Public License Usage
|
|---|
| 16 | ** Alternatively, this file may be used under the terms of the GNU Lesser
|
|---|
| 17 | ** General Public License version 2.1 as published by the Free Software
|
|---|
| 18 | ** Foundation and appearing in the file LICENSE.LGPL included in the
|
|---|
| 19 | ** packaging of this file. Please review the following information to
|
|---|
| 20 | ** ensure the GNU Lesser General Public License version 2.1 requirements
|
|---|
| 21 | ** will be met: http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
|
|---|
| 22 | **
|
|---|
| 23 | ** In addition, as a special exception, Nokia gives you certain
|
|---|
| 24 | ** additional rights. These rights are described in the Nokia Qt LGPL
|
|---|
| 25 | ** Exception version 1.0, included in the file LGPL_EXCEPTION.txt in this
|
|---|
| 26 | ** package.
|
|---|
| 27 | **
|
|---|
| 28 | ** GNU General Public License Usage
|
|---|
| 29 | ** Alternatively, this file may be used under the terms of the GNU
|
|---|
| 30 | ** General Public License version 3.0 as published by the Free Software
|
|---|
| 31 | ** Foundation and appearing in the file LICENSE.GPL included in the
|
|---|
| 32 | ** packaging of this file. Please review the following information to
|
|---|
| 33 | ** ensure the GNU General Public License version 3.0 requirements will be
|
|---|
| 34 | ** met: http://www.gnu.org/copyleft/gpl.html.
|
|---|
| 35 | **
|
|---|
| 36 | ** If you are unsure which license is appropriate for your use, please
|
|---|
| 37 | ** contact the sales department at [email protected].
|
|---|
| 38 | ** $QT_END_LICENSE$
|
|---|
| 39 | **
|
|---|
| 40 | ****************************************************************************/
|
|---|
| 41 |
|
|---|
| 42 | #include <QRegExp>
|
|---|
| 43 | #include <QString>
|
|---|
| 44 |
|
|---|
| 45 | #include "qxmlutils_p.h"
|
|---|
| 46 |
|
|---|
| 47 | QT_BEGIN_NAMESPACE
|
|---|
| 48 |
|
|---|
| 49 | /* TODO:
|
|---|
| 50 | * - isNameChar() doesn't have to be public, it's only needed in
|
|---|
| 51 | * qdom.cpp -- refactor fixedXmlName() to use isNCName()
|
|---|
| 52 | * - A lot of functions can be inlined.
|
|---|
| 53 | */
|
|---|
| 54 |
|
|---|
| 55 | class QXmlCharRange
|
|---|
| 56 | {
|
|---|
| 57 | public:
|
|---|
| 58 | ushort min;
|
|---|
| 59 | ushort max;
|
|---|
| 60 | };
|
|---|
| 61 | typedef const QXmlCharRange *RangeIter;
|
|---|
| 62 |
|
|---|
| 63 | /*!
|
|---|
| 64 | Performs a binary search between \a begin and \a end inclusive, to check whether \a
|
|---|
| 65 | c is contained. Remember that the QXmlCharRange instances must be in numeric order.
|
|---|
| 66 | */
|
|---|
| 67 | bool QXmlUtils::rangeContains(RangeIter begin, RangeIter end, const QChar c)
|
|---|
| 68 | {
|
|---|
| 69 | const ushort cp(c.unicode());
|
|---|
| 70 |
|
|---|
| 71 | // check the first two ranges "manually" as characters in that
|
|---|
| 72 | // range are checked very often and we avoid the binary search below.
|
|---|
| 73 |
|
|---|
| 74 | if (cp <= begin->max)
|
|---|
| 75 | return cp >= begin->min;
|
|---|
| 76 |
|
|---|
| 77 | ++begin;
|
|---|
| 78 |
|
|---|
| 79 | if (begin == end)
|
|---|
| 80 | return false;
|
|---|
| 81 |
|
|---|
| 82 | if (cp <= begin->max)
|
|---|
| 83 | return cp >= begin->min;
|
|---|
| 84 |
|
|---|
| 85 | while (begin != end) {
|
|---|
| 86 | int delta = (end - begin) / 2;
|
|---|
| 87 | RangeIter mid = begin + delta;
|
|---|
| 88 |
|
|---|
| 89 | if (mid->min > cp)
|
|---|
| 90 | end = mid;
|
|---|
| 91 | else if (mid->max < cp)
|
|---|
| 92 | begin = mid;
|
|---|
| 93 | else
|
|---|
| 94 | return true;
|
|---|
| 95 |
|
|---|
| 96 | if (delta == 0)
|
|---|
| 97 | break;
|
|---|
| 98 | }
|
|---|
| 99 |
|
|---|
| 100 | return false;
|
|---|
| 101 | }
|
|---|
| 102 |
|
|---|
| 103 | // [85] BaseChar ::= ...
|
|---|
| 104 |
|
|---|
| 105 | static const QXmlCharRange g_base_begin[] =
|
|---|
| 106 | {
|
|---|
| 107 | {0x0041, 0x005A}, {0x0061, 0x007A}, {0x00C0, 0x00D6}, {0x00D8, 0x00F6}, {0x00F8, 0x00FF},
|
|---|
| 108 | {0x0100, 0x0131}, {0x0134, 0x013E}, {0x0141, 0x0148}, {0x014A, 0x017E}, {0x0180, 0x01C3},
|
|---|
| 109 | {0x01CD, 0x01F0}, {0x01F4, 0x01F5}, {0x01FA, 0x0217}, {0x0250, 0x02A8}, {0x02BB, 0x02C1},
|
|---|
| 110 | {0x0386, 0x0386}, {0x0388, 0x038A}, {0x038C, 0x038C}, {0x038E, 0x03A1}, {0x03A3, 0x03CE},
|
|---|
| 111 | {0x03D0, 0x03D6}, {0x03DA, 0x03DA}, {0x03DC, 0x03DC}, {0x03DE, 0x03DE}, {0x03E0, 0x03E0},
|
|---|
| 112 | {0x03E2, 0x03F3}, {0x0401, 0x040C}, {0x040E, 0x044F}, {0x0451, 0x045C}, {0x045E, 0x0481},
|
|---|
| 113 | {0x0490, 0x04C4}, {0x04C7, 0x04C8}, {0x04CB, 0x04CC}, {0x04D0, 0x04EB}, {0x04EE, 0x04F5},
|
|---|
| 114 | {0x04F8, 0x04F9}, {0x0531, 0x0556}, {0x0559, 0x0559}, {0x0561, 0x0586}, {0x05D0, 0x05EA},
|
|---|
| 115 | {0x05F0, 0x05F2}, {0x0621, 0x063A}, {0x0641, 0x064A}, {0x0671, 0x06B7}, {0x06BA, 0x06BE},
|
|---|
| 116 | {0x06C0, 0x06CE}, {0x06D0, 0x06D3}, {0x06D5, 0x06D5}, {0x06E5, 0x06E6}, {0x0905, 0x0939},
|
|---|
| 117 | {0x093D, 0x093D}, {0x0958, 0x0961}, {0x0985, 0x098C}, {0x098F, 0x0990}, {0x0993, 0x09A8},
|
|---|
| 118 | {0x09AA, 0x09B0}, {0x09B2, 0x09B2}, {0x09B6, 0x09B9}, {0x09DC, 0x09DD}, {0x09DF, 0x09E1},
|
|---|
| 119 | {0x09F0, 0x09F1}, {0x0A05, 0x0A0A}, {0x0A0F, 0x0A10}, {0x0A13, 0x0A28}, {0x0A2A, 0x0A30},
|
|---|
| 120 | {0x0A32, 0x0A33}, {0x0A35, 0x0A36}, {0x0A38, 0x0A39}, {0x0A59, 0x0A5C}, {0x0A5E, 0x0A5E},
|
|---|
| 121 | {0x0A72, 0x0A74}, {0x0A85, 0x0A8B}, {0x0A8D, 0x0A8D}, {0x0A8F, 0x0A91}, {0x0A93, 0x0AA8},
|
|---|
| 122 | {0x0AAA, 0x0AB0}, {0x0AB2, 0x0AB3}, {0x0AB5, 0x0AB9}, {0x0ABD, 0x0ABD}, {0x0AE0, 0x0AE0},
|
|---|
| 123 | {0x0B05, 0x0B0C}, {0x0B0F, 0x0B10}, {0x0B13, 0x0B28}, {0x0B2A, 0x0B30}, {0x0B32, 0x0B33},
|
|---|
| 124 | {0x0B36, 0x0B39}, {0x0B3D, 0x0B3D}, {0x0B5C, 0x0B5D}, {0x0B5F, 0x0B61}, {0x0B85, 0x0B8A},
|
|---|
| 125 | {0x0B8E, 0x0B90}, {0x0B92, 0x0B95}, {0x0B99, 0x0B9A}, {0x0B9C, 0x0B9C}, {0x0B9E, 0x0B9F},
|
|---|
| 126 | {0x0BA3, 0x0BA4}, {0x0BA8, 0x0BAA}, {0x0BAE, 0x0BB5}, {0x0BB7, 0x0BB9}, {0x0C05, 0x0C0C},
|
|---|
| 127 | {0x0C0E, 0x0C10}, {0x0C12, 0x0C28}, {0x0C2A, 0x0C33}, {0x0C35, 0x0C39}, {0x0C60, 0x0C61},
|
|---|
| 128 | {0x0C85, 0x0C8C}, {0x0C8E, 0x0C90}, {0x0C92, 0x0CA8}, {0x0CAA, 0x0CB3}, {0x0CB5, 0x0CB9},
|
|---|
| 129 | {0x0CDE, 0x0CDE}, {0x0CE0, 0x0CE1}, {0x0D05, 0x0D0C}, {0x0D0E, 0x0D10}, {0x0D12, 0x0D28},
|
|---|
| 130 | {0x0D2A, 0x0D39}, {0x0D60, 0x0D61}, {0x0E01, 0x0E2E}, {0x0E30, 0x0E30}, {0x0E32, 0x0E33},
|
|---|
| 131 | {0x0E40, 0x0E45}, {0x0E81, 0x0E82}, {0x0E84, 0x0E84}, {0x0E87, 0x0E88}, {0x0E8A, 0x0E8A},
|
|---|
| 132 | {0x0E8D, 0x0E8D}, {0x0E94, 0x0E97}, {0x0E99, 0x0E9F}, {0x0EA1, 0x0EA3}, {0x0EA5, 0x0EA5},
|
|---|
| 133 | {0x0EA7, 0x0EA7}, {0x0EAA, 0x0EAB}, {0x0EAD, 0x0EAE}, {0x0EB0, 0x0EB0}, {0x0EB2, 0x0EB3},
|
|---|
| 134 | {0x0EBD, 0x0EBD}, {0x0EC0, 0x0EC4}, {0x0F40, 0x0F47}, {0x0F49, 0x0F69}, {0x10A0, 0x10C5},
|
|---|
| 135 | {0x10D0, 0x10F6}, {0x1100, 0x1100}, {0x1102, 0x1103}, {0x1105, 0x1107}, {0x1109, 0x1109},
|
|---|
| 136 | {0x110B, 0x110C}, {0x110E, 0x1112}, {0x113C, 0x113C}, {0x113E, 0x113E}, {0x1140, 0x1140},
|
|---|
| 137 | {0x114C, 0x114C}, {0x114E, 0x114E}, {0x1150, 0x1150}, {0x1154, 0x1155}, {0x1159, 0x1159},
|
|---|
| 138 | {0x115F, 0x1161}, {0x1163, 0x1163}, {0x1165, 0x1165}, {0x1167, 0x1167}, {0x1169, 0x1169},
|
|---|
| 139 | {0x116D, 0x116E}, {0x1172, 0x1173}, {0x1175, 0x1175}, {0x119E, 0x119E}, {0x11A8, 0x11A8},
|
|---|
| 140 | {0x11AB, 0x11AB}, {0x11AE, 0x11AF}, {0x11B7, 0x11B8}, {0x11BA, 0x11BA}, {0x11BC, 0x11C2},
|
|---|
| 141 | {0x11EB, 0x11EB}, {0x11F0, 0x11F0}, {0x11F9, 0x11F9}, {0x1E00, 0x1E9B}, {0x1EA0, 0x1EF9},
|
|---|
| 142 | {0x1F00, 0x1F15}, {0x1F18, 0x1F1D}, {0x1F20, 0x1F45}, {0x1F48, 0x1F4D}, {0x1F50, 0x1F57},
|
|---|
| 143 | {0x1F59, 0x1F59}, {0x1F5B, 0x1F5B}, {0x1F5D, 0x1F5D}, {0x1F5F, 0x1F7D}, {0x1F80, 0x1FB4},
|
|---|
| 144 | {0x1FB6, 0x1FBC}, {0x1FBE, 0x1FBE}, {0x1FC2, 0x1FC4}, {0x1FC6, 0x1FCC}, {0x1FD0, 0x1FD3},
|
|---|
| 145 | {0x1FD6, 0x1FDB}, {0x1FE0, 0x1FEC}, {0x1FF2, 0x1FF4}, {0x1FF6, 0x1FFC}, {0x2126, 0x2126},
|
|---|
| 146 | {0x212A, 0x212B}, {0x212E, 0x212E}, {0x2180, 0x2182}, {0x3041, 0x3094}, {0x30A1, 0x30FA},
|
|---|
| 147 | {0x3105, 0x312C}, {0xAC00, 0xD7A3}
|
|---|
| 148 | };
|
|---|
| 149 | static const RangeIter g_base_end = g_base_begin + sizeof(g_base_begin) / sizeof(QXmlCharRange);
|
|---|
| 150 |
|
|---|
| 151 | static const QXmlCharRange g_ideographic_begin[] =
|
|---|
| 152 | {
|
|---|
| 153 | {0x3007, 0x3007}, {0x3021, 0x3029}, {0x4E00, 0x9FA5}
|
|---|
| 154 | };
|
|---|
| 155 | static const RangeIter g_ideographic_end = g_ideographic_begin + sizeof(g_ideographic_begin) / sizeof(QXmlCharRange);
|
|---|
| 156 |
|
|---|
| 157 | bool QXmlUtils::isIdeographic(const QChar c)
|
|---|
| 158 | {
|
|---|
| 159 | return rangeContains(g_ideographic_begin, g_ideographic_end, c);
|
|---|
| 160 | }
|
|---|
| 161 |
|
|---|
| 162 | static const QXmlCharRange g_combining_begin[] =
|
|---|
| 163 | {
|
|---|
| 164 | {0x0300, 0x0345}, {0x0360, 0x0361}, {0x0483, 0x0486}, {0x0591, 0x05A1}, {0x05A3, 0x05B9},
|
|---|
| 165 | {0x05BB, 0x05BD}, {0x05BF, 0x05BF}, {0x05C1, 0x05C2}, {0x05C4, 0x05C4}, {0x064B, 0x0652},
|
|---|
| 166 | {0x0670, 0x0670}, {0x06D6, 0x06DC}, {0x06DD, 0x06DF}, {0x06E0, 0x06E4}, {0x06E7, 0x06E8},
|
|---|
| 167 | {0x06EA, 0x06ED}, {0x0901, 0x0903}, {0x093C, 0x093C}, {0x093E, 0x094C}, {0x094D, 0x094D},
|
|---|
| 168 | {0x0951, 0x0954}, {0x0962, 0x0963}, {0x0981, 0x0983}, {0x09BC, 0x09BC}, {0x09BE, 0x09BE},
|
|---|
| 169 | {0x09BF, 0x09BF}, {0x09C0, 0x09C4}, {0x09C7, 0x09C8}, {0x09CB, 0x09CD}, {0x09D7, 0x09D7},
|
|---|
| 170 | {0x09E2, 0x09E3}, {0x0A02, 0x0A02}, {0x0A3C, 0x0A3C}, {0x0A3E, 0x0A3E}, {0x0A3F, 0x0A3F},
|
|---|
| 171 | {0x0A40, 0x0A42}, {0x0A47, 0x0A48}, {0x0A4B, 0x0A4D}, {0x0A70, 0x0A71}, {0x0A81, 0x0A83},
|
|---|
| 172 | {0x0ABC, 0x0ABC}, {0x0ABE, 0x0AC5}, {0x0AC7, 0x0AC9}, {0x0ACB, 0x0ACD}, {0x0B01, 0x0B03},
|
|---|
| 173 | {0x0B3C, 0x0B3C}, {0x0B3E, 0x0B43}, {0x0B47, 0x0B48}, {0x0B4B, 0x0B4D}, {0x0B56, 0x0B57},
|
|---|
| 174 | {0x0B82, 0x0B83}, {0x0BBE, 0x0BC2}, {0x0BC6, 0x0BC8}, {0x0BCA, 0x0BCD}, {0x0BD7, 0x0BD7},
|
|---|
| 175 | {0x0C01, 0x0C03}, {0x0C3E, 0x0C44}, {0x0C46, 0x0C48}, {0x0C4A, 0x0C4D}, {0x0C55, 0x0C56},
|
|---|
| 176 | {0x0C82, 0x0C83}, {0x0CBE, 0x0CC4}, {0x0CC6, 0x0CC8}, {0x0CCA, 0x0CCD}, {0x0CD5, 0x0CD6},
|
|---|
| 177 | {0x0D02, 0x0D03}, {0x0D3E, 0x0D43}, {0x0D46, 0x0D48}, {0x0D4A, 0x0D4D}, {0x0D57, 0x0D57},
|
|---|
| 178 | {0x0E31, 0x0E31}, {0x0E34, 0x0E3A}, {0x0E47, 0x0E4E}, {0x0EB1, 0x0EB1}, {0x0EB4, 0x0EB9},
|
|---|
| 179 | {0x0EBB, 0x0EBC}, {0x0EC8, 0x0ECD}, {0x0F18, 0x0F19}, {0x0F35, 0x0F35}, {0x0F37, 0x0F37},
|
|---|
| 180 | {0x0F39, 0x0F39}, {0x0F3E, 0x0F3E}, {0x0F3F, 0x0F3F}, {0x0F71, 0x0F84}, {0x0F86, 0x0F8B},
|
|---|
| 181 | {0x0F90, 0x0F95}, {0x0F97, 0x0F97}, {0x0F99, 0x0FAD}, {0x0FB1, 0x0FB7}, {0x0FB9, 0x0FB9},
|
|---|
| 182 | {0x20D0, 0x20DC}, {0x20E1, 0x20E1}, {0x302A, 0x302F}, {0x3099, 0x3099}, {0x309A, 0x309A}
|
|---|
| 183 | };
|
|---|
| 184 | static const RangeIter g_combining_end = g_combining_begin + sizeof(g_combining_begin) / sizeof(QXmlCharRange);
|
|---|
| 185 |
|
|---|
| 186 | bool QXmlUtils::isCombiningChar(const QChar c)
|
|---|
| 187 | {
|
|---|
| 188 | return rangeContains(g_combining_begin, g_combining_end, c);
|
|---|
| 189 | }
|
|---|
| 190 |
|
|---|
| 191 | // [88] Digit ::= ...
|
|---|
| 192 | static const QXmlCharRange g_digit_begin[] =
|
|---|
| 193 | {
|
|---|
| 194 | {0x0030, 0x0039}, {0x0660, 0x0669}, {0x06F0, 0x06F9}, {0x0966, 0x096F}, {0x09E6, 0x09EF},
|
|---|
| 195 | {0x0A66, 0x0A6F}, {0x0AE6, 0x0AEF}, {0x0B66, 0x0B6F}, {0x0BE7, 0x0BEF}, {0x0C66, 0x0C6F},
|
|---|
| 196 | {0x0CE6, 0x0CEF}, {0x0D66, 0x0D6F}, {0x0E50, 0x0E59}, {0x0ED0, 0x0ED9}, {0x0F20, 0x0F29}
|
|---|
| 197 | };
|
|---|
| 198 | static const RangeIter g_digit_end = g_digit_begin + sizeof(g_digit_begin) / sizeof(QXmlCharRange);
|
|---|
| 199 |
|
|---|
| 200 | bool QXmlUtils::isDigit(const QChar c)
|
|---|
| 201 | {
|
|---|
| 202 | return rangeContains(g_digit_begin, g_digit_end, c);
|
|---|
| 203 | }
|
|---|
| 204 |
|
|---|
| 205 | // [89] Extender ::= ...
|
|---|
| 206 | static const QXmlCharRange g_extender_begin[] =
|
|---|
| 207 | {
|
|---|
| 208 | {0x00B7, 0x00B7}, {0x02D0, 0x02D0}, {0x02D1, 0x02D1}, {0x0387, 0x0387}, {0x0640, 0x0640},
|
|---|
| 209 | {0x0E46, 0x0E46}, {0x0EC6, 0x0EC6}, {0x3005, 0x3005}, {0x3031, 0x3035}, {0x309D, 0x309E},
|
|---|
| 210 | {0x30FC, 0x30FE}
|
|---|
| 211 | };
|
|---|
| 212 | static const RangeIter g_extender_end = g_extender_begin + sizeof(g_extender_begin) / sizeof(QXmlCharRange);
|
|---|
| 213 |
|
|---|
| 214 | bool QXmlUtils::isExtender(const QChar c)
|
|---|
| 215 | {
|
|---|
| 216 | return rangeContains(g_extender_begin, g_extender_end, c);
|
|---|
| 217 | }
|
|---|
| 218 |
|
|---|
| 219 | bool QXmlUtils::isBaseChar(const QChar c)
|
|---|
| 220 | {
|
|---|
| 221 | return rangeContains(g_base_begin, g_base_end, c);
|
|---|
| 222 | }
|
|---|
| 223 |
|
|---|
| 224 | /*!
|
|---|
| 225 | \internal
|
|---|
| 226 |
|
|---|
| 227 | Determines whether \a encName is a valid instance of production [81]EncName in the XML 1.0
|
|---|
| 228 | specification. If it is, true is returned, otherwise false.
|
|---|
| 229 |
|
|---|
| 230 | \sa \l {http://www.w3.org/TR/REC-xml/#NT-EncName}
|
|---|
| 231 | {Extensible Markup Language (XML) 1.0 (Fourth Edition), [81] EncName}
|
|---|
| 232 | */
|
|---|
| 233 | bool QXmlUtils::isEncName(const QString &encName)
|
|---|
| 234 | {
|
|---|
| 235 | /* Right, we here have a dependency on QRegExp. Writing a manual parser to
|
|---|
| 236 | * replace that regexp is probably a 70 lines so I prioritize this to when
|
|---|
| 237 | * the dependency is considered alarming, or when the rest of the bugs
|
|---|
| 238 | * are fixed. */
|
|---|
| 239 | const QRegExp encNameRegExp(QLatin1String("[A-Za-z][A-Za-z0-9._\\-]*"));
|
|---|
| 240 | Q_ASSERT(encNameRegExp.isValid());
|
|---|
| 241 |
|
|---|
| 242 | return encNameRegExp.exactMatch(encName);
|
|---|
| 243 | }
|
|---|
| 244 |
|
|---|
| 245 | /*!
|
|---|
| 246 | \internal
|
|---|
| 247 |
|
|---|
| 248 | Determines whether \a c is a valid instance of production [84]Letter in the XML 1.0
|
|---|
| 249 | specification. If it is, true is returned, otherwise false.
|
|---|
| 250 |
|
|---|
| 251 | \sa \l {http://www.w3.org/TR/REC-xml/#NT-Letter}
|
|---|
| 252 | {Extensible Markup Language (XML) 1.0 (Fourth Edition), [84] Letter}
|
|---|
| 253 | */
|
|---|
| 254 | bool QXmlUtils::isLetter(const QChar c)
|
|---|
| 255 | {
|
|---|
| 256 | return isBaseChar(c) || isIdeographic(c);
|
|---|
| 257 | }
|
|---|
| 258 |
|
|---|
| 259 | /*!
|
|---|
| 260 | \internal
|
|---|
| 261 |
|
|---|
| 262 | Determines whether \a c is a valid instance of production [2]Char in the XML 1.0
|
|---|
| 263 | specification. If it is, true is returned, otherwise false.
|
|---|
| 264 |
|
|---|
| 265 | \sa \l {http://www.w3.org/TR/REC-xml/#NT-Char}
|
|---|
| 266 | {Extensible Markup Language (XML) 1.0 (Fourth Edition), [2] Char}
|
|---|
| 267 | */
|
|---|
| 268 | bool QXmlUtils::isChar(const QChar c)
|
|---|
| 269 | {
|
|---|
| 270 | return (c.unicode() >= 0x0020 && c.unicode() <= 0xD7FF)
|
|---|
| 271 | || c.unicode() == 0x0009
|
|---|
| 272 | || c.unicode() == 0x000A
|
|---|
| 273 | || c.unicode() == 0x000D
|
|---|
| 274 | || (c.unicode() >= 0xE000 && c.unicode() <= 0xFFFD);
|
|---|
| 275 | }
|
|---|
| 276 |
|
|---|
| 277 | /*!
|
|---|
| 278 | \internal
|
|---|
| 279 |
|
|---|
| 280 | Determines whether \a c is a valid instance of
|
|---|
| 281 | production [4]NameChar in the XML 1.0 specification. If it
|
|---|
| 282 | is, true is returned, otherwise false.
|
|---|
| 283 |
|
|---|
| 284 | \sa \l {http://www.w3.org/TR/REC-xml/#NT-NameChar}
|
|---|
| 285 | {Extensible Markup Language (XML) 1.0 (Fourth Edition), [4] NameChar}
|
|---|
| 286 | */
|
|---|
| 287 | bool QXmlUtils::isNameChar(const QChar c)
|
|---|
| 288 | {
|
|---|
| 289 | return isBaseChar(c)
|
|---|
| 290 | || isDigit(c)
|
|---|
| 291 | || c.unicode() == '.'
|
|---|
| 292 | || c.unicode() == '-'
|
|---|
| 293 | || c.unicode() == '_'
|
|---|
| 294 | || c.unicode() == ':'
|
|---|
| 295 | || isCombiningChar(c)
|
|---|
| 296 | || isIdeographic(c)
|
|---|
| 297 | || isExtender(c);
|
|---|
| 298 | }
|
|---|
| 299 |
|
|---|
| 300 | /*!
|
|---|
| 301 | \internal
|
|---|
| 302 |
|
|---|
| 303 | Determines whether \a c is a valid instance of
|
|---|
| 304 | production [12] PubidLiteral in the XML 1.0 specification. If it
|
|---|
| 305 | is, true is returned, otherwise false.
|
|---|
| 306 |
|
|---|
| 307 | \sa \l {http://www.w3.org/TR/REC-xml/#NT-PubidLiteral}
|
|---|
| 308 | {Extensible Markup Language (XML) 1.0 (Fourth Edition), [12] PubidLiteral}
|
|---|
| 309 | */
|
|---|
| 310 | bool QXmlUtils::isPublicID(const QString &candidate)
|
|---|
| 311 | {
|
|---|
| 312 | const int len = candidate.length();
|
|---|
| 313 |
|
|---|
| 314 | for(int i = 0; i < len; ++i)
|
|---|
| 315 | {
|
|---|
| 316 | const ushort cp = candidate.at(i).unicode();
|
|---|
| 317 |
|
|---|
| 318 | if ((cp >= 'a' && cp <= 'z')
|
|---|
| 319 | || (cp >= 'A' && cp <= 'Z')
|
|---|
| 320 | || (cp >= '0' && cp <= '9'))
|
|---|
| 321 | {
|
|---|
| 322 | continue;
|
|---|
| 323 | }
|
|---|
| 324 |
|
|---|
| 325 | switch (cp)
|
|---|
| 326 | {
|
|---|
| 327 | /* Fallthrough all these. */
|
|---|
| 328 | case 0x20:
|
|---|
| 329 | case 0x0D:
|
|---|
| 330 | case 0x0A:
|
|---|
| 331 | case '-':
|
|---|
| 332 | case '\'':
|
|---|
| 333 | case '(':
|
|---|
| 334 | case ')':
|
|---|
| 335 | case '+':
|
|---|
| 336 | case ',':
|
|---|
| 337 | case '.':
|
|---|
| 338 | case '/':
|
|---|
| 339 | case ':':
|
|---|
| 340 | case '=':
|
|---|
| 341 | case '?':
|
|---|
| 342 | case ';':
|
|---|
| 343 | case '!':
|
|---|
| 344 | case '*':
|
|---|
| 345 | case '#':
|
|---|
| 346 | case '@':
|
|---|
| 347 | case '$':
|
|---|
| 348 | case '_':
|
|---|
| 349 | case '%':
|
|---|
| 350 | continue;
|
|---|
| 351 | default:
|
|---|
| 352 | return false;
|
|---|
| 353 | }
|
|---|
| 354 | }
|
|---|
| 355 |
|
|---|
| 356 | return true;
|
|---|
| 357 | }
|
|---|
| 358 |
|
|---|
| 359 | /*!
|
|---|
| 360 | \internal
|
|---|
| 361 |
|
|---|
| 362 | Determines whether \a c is a valid instance of
|
|---|
| 363 | production [4]NCName in the XML 1.0 Namespaces specification. If it
|
|---|
| 364 | is, true is returned, otherwise false.
|
|---|
| 365 |
|
|---|
| 366 | \sa \l {http://www.w3.org/TR/REC-xml-names/#NT-NCName}
|
|---|
| 367 | {W3CNamespaces in XML 1.0 (Second Edition), [4] NCName}
|
|---|
| 368 | */
|
|---|
| 369 | bool QXmlUtils::isNCName(const QStringRef &ncName)
|
|---|
| 370 | {
|
|---|
| 371 | if(ncName.isEmpty())
|
|---|
| 372 | return false;
|
|---|
| 373 |
|
|---|
| 374 | const QChar first(ncName.at(0));
|
|---|
| 375 |
|
|---|
| 376 | if(!QXmlUtils::isLetter(first) && first.unicode() != '_' && first.unicode() != ':')
|
|---|
| 377 | return false;
|
|---|
| 378 |
|
|---|
| 379 | const int len = ncName.size();
|
|---|
| 380 | for(int i = 0; i < len; ++i)
|
|---|
| 381 | {
|
|---|
| 382 | const QChar &at = ncName.at(i);
|
|---|
| 383 | if(!QXmlUtils::isNameChar(at) || at == QLatin1Char(':'))
|
|---|
| 384 | return false;
|
|---|
| 385 | }
|
|---|
| 386 |
|
|---|
| 387 | return true;
|
|---|
| 388 | }
|
|---|
| 389 |
|
|---|
| 390 | QT_END_NAMESPACE
|
|---|
