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 QtXmlPatterns 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 <QHash>
|
---|
43 |
|
---|
44 | #include "qpatternistlocale_p.h"
|
---|
45 |
|
---|
46 | #include "qpatternplatform_p.h"
|
---|
47 |
|
---|
48 | QT_BEGIN_NAMESPACE
|
---|
49 |
|
---|
50 | using namespace QPatternist;
|
---|
51 |
|
---|
52 | namespace QPatternist
|
---|
53 | {
|
---|
54 | /**
|
---|
55 | * @short Used internally by PatternPlatform and describes
|
---|
56 | * a flag that affects how a pattern is treated.
|
---|
57 | *
|
---|
58 | * The member variables aren't declared @c const, in order
|
---|
59 | * to make the synthesized assignment operator and copy constructor work.
|
---|
60 | *
|
---|
61 | * @ingroup Patternist_utils
|
---|
62 | * @author Frans Englich <[email protected]>
|
---|
63 | */
|
---|
64 | class PatternFlag
|
---|
65 | {
|
---|
66 | public:
|
---|
67 | typedef QHash<QChar, PatternFlag> Hash;
|
---|
68 |
|
---|
69 | inline PatternFlag() : flag(PatternPlatform::NoFlags)
|
---|
70 | {
|
---|
71 | }
|
---|
72 |
|
---|
73 | inline PatternFlag(const PatternPlatform::Flag opt,
|
---|
74 | const QString &descr) : flag(opt),
|
---|
75 | description(descr)
|
---|
76 | {
|
---|
77 | }
|
---|
78 |
|
---|
79 | PatternPlatform::Flag flag;
|
---|
80 | QString description;
|
---|
81 |
|
---|
82 | static inline Hash flagDescriptions();
|
---|
83 | };
|
---|
84 | }
|
---|
85 |
|
---|
86 | static inline PatternFlag::Hash flagDescriptions()
|
---|
87 | {
|
---|
88 | PatternFlag::Hash retval;
|
---|
89 |
|
---|
90 | retval.insert(QChar(QLatin1Char('s')),
|
---|
91 | PatternFlag(PatternPlatform::DotAllMode,
|
---|
92 | QtXmlPatterns::tr("%1 matches newline characters").arg(formatKeyword(QLatin1Char('.')))));
|
---|
93 |
|
---|
94 | retval.insert(QChar(QLatin1Char('m')),
|
---|
95 | PatternFlag(PatternPlatform::MultiLineMode,
|
---|
96 | QtXmlPatterns::tr("%1 and %2 match the start and end of a line.")
|
---|
97 | .arg(formatKeyword(QLatin1Char('^')))
|
---|
98 | .arg(formatKeyword(QLatin1Char('$')))));
|
---|
99 |
|
---|
100 | retval.insert(QChar(QLatin1Char('i')),
|
---|
101 | PatternFlag(PatternPlatform::CaseInsensitive,
|
---|
102 | QtXmlPatterns::tr("Matches are case insensitive")));
|
---|
103 |
|
---|
104 | retval.insert(QChar(QLatin1Char('x')),
|
---|
105 | PatternFlag(PatternPlatform::SimplifyWhitespace,
|
---|
106 | QtXmlPatterns::tr("Whitespace characters are removed, except when they appear "
|
---|
107 | "in character classes")));
|
---|
108 |
|
---|
109 | return retval;
|
---|
110 | }
|
---|
111 |
|
---|
112 | PatternPlatform::PatternPlatform(const qint8 flagsPosition) : m_compiledParts(NoPart),
|
---|
113 | m_flags(NoFlags),
|
---|
114 | m_flagsPosition(flagsPosition)
|
---|
115 | {
|
---|
116 | }
|
---|
117 |
|
---|
118 | const QRegExp PatternPlatform::pattern(const DynamicContext::Ptr &context) const
|
---|
119 | {
|
---|
120 | if(m_compiledParts == FlagsAndPattern) /* This is the most common case. */
|
---|
121 | {
|
---|
122 | Q_ASSERT(m_pattern.isValid());
|
---|
123 | return m_pattern;
|
---|
124 | }
|
---|
125 |
|
---|
126 | QRegExp retvalPattern;
|
---|
127 | Flags flags;
|
---|
128 |
|
---|
129 | /* Compile the flags, if necessary. */
|
---|
130 | if(m_compiledParts.testFlag(FlagsPrecompiled))
|
---|
131 | flags = m_flags;
|
---|
132 | else
|
---|
133 | {
|
---|
134 | const Expression::Ptr flagsOp(m_operands.value(m_flagsPosition));
|
---|
135 |
|
---|
136 | if(flagsOp)
|
---|
137 | flags = parseFlags(flagsOp->evaluateSingleton(context).stringValue(), context);
|
---|
138 | else
|
---|
139 | flags = NoFlags;
|
---|
140 | }
|
---|
141 |
|
---|
142 | /* Compile the pattern, if necessary. */
|
---|
143 | if(m_compiledParts.testFlag(PatternPrecompiled))
|
---|
144 | retvalPattern = m_pattern;
|
---|
145 | else
|
---|
146 | {
|
---|
147 | retvalPattern = parsePattern(m_operands.at(1)->evaluateSingleton(context).stringValue(),
|
---|
148 | context);
|
---|
149 |
|
---|
150 | }
|
---|
151 |
|
---|
152 | applyFlags(flags, retvalPattern);
|
---|
153 |
|
---|
154 | Q_ASSERT(m_pattern.isValid());
|
---|
155 | return retvalPattern;
|
---|
156 | }
|
---|
157 |
|
---|
158 | void PatternPlatform::applyFlags(const Flags flags, QRegExp &patternP)
|
---|
159 | {
|
---|
160 | Q_ASSERT(patternP.isValid());
|
---|
161 | if(flags == NoFlags)
|
---|
162 | return;
|
---|
163 |
|
---|
164 | if(flags & CaseInsensitive)
|
---|
165 | {
|
---|
166 | patternP.setCaseSensitivity(Qt::CaseInsensitive);
|
---|
167 | }
|
---|
168 | // TODO Apply the other flags, like 'x'.
|
---|
169 | }
|
---|
170 |
|
---|
171 | QRegExp PatternPlatform::parsePattern(const QString &patternP,
|
---|
172 | const DynamicContext::Ptr &context) const
|
---|
173 | {
|
---|
174 | if(patternP == QLatin1String("(.)\\3") ||
|
---|
175 | patternP == QLatin1String("\\3") ||
|
---|
176 | patternP == QLatin1String("(.)\\2"))
|
---|
177 | {
|
---|
178 | context->error(QLatin1String("We don't want to hang infinitely on K2-MatchesFunc-9, "
|
---|
179 | "10 and 11. See Trolltech task 148505."),
|
---|
180 | ReportContext::FOER0000, this);
|
---|
181 | return QRegExp();
|
---|
182 | }
|
---|
183 |
|
---|
184 | QString rewrittenPattern(patternP);
|
---|
185 |
|
---|
186 | /* We rewrite some well known patterns to QRegExp style here. Note that
|
---|
187 | * these character classes only works in the ASCII range, and fail for
|
---|
188 | * others. This support needs to be in QRegExp, since it's about checking
|
---|
189 | * QChar::category(). */
|
---|
190 | rewrittenPattern.replace(QLatin1String("[\\i-[:]]"), QLatin1String("[a-zA-Z_]"));
|
---|
191 | rewrittenPattern.replace(QLatin1String("[\\c-[:]]"), QLatin1String("[a-zA-Z0-9_\\-\\.]"));
|
---|
192 | rewrittenPattern.replace(QLatin1String("\\i"), QLatin1String("[a-zA-Z:_]"));
|
---|
193 | rewrittenPattern.replace(QLatin1String("\\c"), QLatin1String("[a-zA-Z0-9:_\\-\\.]"));
|
---|
194 | rewrittenPattern.replace(QLatin1String("\\p{L}"), QLatin1String("[a-zA-Z]"));
|
---|
195 | rewrittenPattern.replace(QLatin1String("\\p{Lu}"), QLatin1String("[A-Z]"));
|
---|
196 | rewrittenPattern.replace(QLatin1String("\\p{Ll}"), QLatin1String("[a-z]"));
|
---|
197 | rewrittenPattern.replace(QLatin1String("\\p{Nd}"), QLatin1String("[0-9]"));
|
---|
198 |
|
---|
199 | QRegExp retval(rewrittenPattern);
|
---|
200 |
|
---|
201 | if(retval.isValid())
|
---|
202 | return retval;
|
---|
203 | else
|
---|
204 | {
|
---|
205 | context->error(QtXmlPatterns::tr("%1 is an invalid regular expression pattern: %2")
|
---|
206 | .arg(formatExpression(patternP), retval.errorString()),
|
---|
207 | ReportContext::FORX0002, this);
|
---|
208 | return QRegExp();
|
---|
209 | }
|
---|
210 | }
|
---|
211 |
|
---|
212 | PatternPlatform::Flags PatternPlatform::parseFlags(const QString &flags,
|
---|
213 | const DynamicContext::Ptr &context) const
|
---|
214 | {
|
---|
215 |
|
---|
216 | if(flags.isEmpty())
|
---|
217 | return NoFlags;
|
---|
218 |
|
---|
219 | const PatternFlag::Hash flagDescrs(flagDescriptions());
|
---|
220 | const int len = flags.length();
|
---|
221 | Flags retval = NoFlags;
|
---|
222 |
|
---|
223 | for(int i = 0; i < len; ++i)
|
---|
224 | {
|
---|
225 | const QChar flag(flags.at(i));
|
---|
226 | const Flag specified = flagDescrs.value(flag).flag;
|
---|
227 |
|
---|
228 | if(specified != NoFlags)
|
---|
229 | {
|
---|
230 | retval |= specified;
|
---|
231 | continue;
|
---|
232 | }
|
---|
233 |
|
---|
234 | /* Generate a nice error message. */
|
---|
235 | QString message(QtXmlPatterns::tr("%1 is an invalid flag for regular expressions. Valid flags are:")
|
---|
236 | .arg(formatKeyword(flag)));
|
---|
237 |
|
---|
238 | /* This is formatting, so don't bother translators with it. */
|
---|
239 | message.append(QLatin1Char('\n'));
|
---|
240 |
|
---|
241 | const PatternFlag::Hash::const_iterator end(flagDescrs.constEnd());
|
---|
242 | PatternFlag::Hash::const_iterator it(flagDescrs.constBegin());
|
---|
243 |
|
---|
244 | for(; it != end;)
|
---|
245 | {
|
---|
246 | // TODO handle bidi correctly
|
---|
247 | // TODO format this with rich text(list/table)
|
---|
248 | message.append(formatKeyword(it.key()));
|
---|
249 | message.append(QLatin1String(" - "));
|
---|
250 | message.append(it.value().description);
|
---|
251 |
|
---|
252 | ++it;
|
---|
253 | if(it != end)
|
---|
254 | message.append(QLatin1Char('\n'));
|
---|
255 | }
|
---|
256 |
|
---|
257 | context->error(message, ReportContext::FORX0001, this);
|
---|
258 | return NoFlags;
|
---|
259 | }
|
---|
260 |
|
---|
261 | return retval;
|
---|
262 | }
|
---|
263 |
|
---|
264 | Expression::Ptr PatternPlatform::compress(const StaticContext::Ptr &context)
|
---|
265 | {
|
---|
266 | const Expression::Ptr me(FunctionCall::compress(context));
|
---|
267 | if(me != this)
|
---|
268 | return me;
|
---|
269 |
|
---|
270 | if(m_operands.at(1)->is(IDStringValue))
|
---|
271 | {
|
---|
272 | const DynamicContext::Ptr dynContext(context->dynamicContext());
|
---|
273 |
|
---|
274 | m_pattern = parsePattern(m_operands.at(1)->evaluateSingleton(dynContext).stringValue(),
|
---|
275 | dynContext);
|
---|
276 | m_compiledParts |= PatternPrecompiled;
|
---|
277 | }
|
---|
278 |
|
---|
279 | const Expression::Ptr flagOperand(m_operands.value(m_flagsPosition));
|
---|
280 |
|
---|
281 | if(!flagOperand)
|
---|
282 | {
|
---|
283 | m_flags = NoFlags;
|
---|
284 | m_compiledParts |= FlagsPrecompiled;
|
---|
285 | }
|
---|
286 | else if(flagOperand->is(IDStringValue))
|
---|
287 | {
|
---|
288 | const DynamicContext::Ptr dynContext(context->dynamicContext());
|
---|
289 | m_flags = parseFlags(flagOperand->evaluateSingleton(dynContext).stringValue(),
|
---|
290 | dynContext);
|
---|
291 | m_compiledParts |= FlagsPrecompiled;
|
---|
292 | }
|
---|
293 |
|
---|
294 | if(m_compiledParts == FlagsAndPattern)
|
---|
295 | applyFlags(m_flags, m_pattern);
|
---|
296 |
|
---|
297 | return me;
|
---|
298 | }
|
---|
299 |
|
---|
300 | QT_END_NAMESPACE
|
---|