| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
After the move to PCRE2, optimizing patterns has been a thorn in the
side due to the fact that PCRE2's JIT compiler modifies the pattern
object itself (instead of returning a new set of data, like PCRE1
did). To make this fit with the existing behavior, a read/write
lock was introduced, with the read part locking when matching and
the write when compiling (or JIT-compiling) the pattern.
This locking strategy however introduced a performance issue,
as we needed:
* to acquire a write lock to compile/optimize the pattern (incl. the
common case where the pattern was already compiled, so bailing out
immediately);
* to acquire a read lock during the actual match, to prevent
some other thread from optimizing the pattern under our nose.
This was due to the "lazy" optimization policy of QRegularExpression
-- optimize a pattern after a certain number of usages. The
excessive amount of locking effectively limited scalability.
Simplify the code, and drop that policy altogether: since JIT
compiling in PCRE2 is faster and pretty much "always recommended",
just always do it for any pattern (unless it gets disabled via
env variables) when compiling it.
This allows to go back to a plain QMutex, and now the actual
matching doesn't require acquiring any locks any longer. Of course,
there is still a mutex acquired just before matching for checking
whether the pattern needs recompiling in the first place; this can
probably be further optimized via double-checked locking (using
atomics), but not doing it right now.
This shift makes a couple of pattern options controlling
optimization useless, and allows to centralize the 3
QRegularExpression tests (which were actually the very same test,
just setting slightly different optimizations strategies).
While at it, install a stress-test for threading, with the idea
of running it under TSAN or helgrind to catch bugs in
QRegularExpression's locking.
[ChangeLog][Important Behavior Changes][QRegularExpression] Regular
expressions are now automatically optimized (including JIT
compiling) on their first usage. The pattern options
OptimizeOnFirstUsageOption and DontAutomaticallyOptimizeOption no
longer have any effect, and will get removed in a future version of
Qt. QRegularExpression::optimize() can be still used to compile and
optimize the regular expression in advance (before any match), if
needed.
Task-number: QTBUG-66781
Change-Id: Ia0e97208ae78255fe811b78029ed01c204e47bd2
Reviewed-by: David Faure <david.faure@kdab.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This method will make QRegularExpression on par with QRegExp and
will allow to replace this class when a wildcard expression can be
set through an API (e.g. QSortFilterProxyModel::setFilterWildcard).
For other use cases, see QTBUG-34052.
[ChangeLog][QRegularExpression] Implemented support for wildcard
patterns.
Warning: QRegularExpression might not give the exact same result
as QRegExp as its implementation follows strictly the glob patterns
definition for the wildcard expressions.
Change-Id: I5ed4617ca679159430c3d46da3449f6b3100e366
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
From Qt 5.7 -> tools & applications are lisenced under GPL v3 with some
exceptions, see
http://blog.qt.io/blog/2016/01/13/new-agreement-with-the-kde-free-qt-foundation/
Updated license headers to use new GPL-EXCEPT header instead of LGPL21 one
(in those files which will be under GPL 3 with exceptions)
Change-Id: I42a473ddc97101492a60b9287d90979d9eb35ae1
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Playing with the offset argument of pcre_exec is not equivalent to
adjusting the pointer to the subject string. In particular, PCRE
can go behind the offset to check for lookbehinds or "transition"
metacharacters (\b, \B, etc.).
This made the code that deals with QStringRefs not matching in behavior
with the corresponding code dealing with QStrings. For instance,
QString subject("Miss");
QRegularExpression re("(?<=M)iss");
re.match(subject.mid(1)); // doesn't match
re.match(subject.midRef(1)); // matches!!!
Instead, actually adjust the pointer to the subject string so that
the behavior is identical. A broken test that relied on the
equivalence is also removed.
Change-Id: If96333241ef59621d7f5a6a170ebd0a186844874
Reviewed-by: Volker Krause <volker.krause@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Qt copyrights are now in The Qt Company, so we could update the source
code headers accordingly. In the same go we should also fix the links to
point to qt.io.
Outdated header.LGPL removed (use header.LGPL21 instead)
Old header.LGPL3 renamed to header.LGPL3-COMM to match actual licensing
combination. New header.LGPL-COMM taken in the use file which were
using old header.LGPL3 (src/plugins/platforms/android/extract.cpp)
Added new header.LGPL3 containing Commercial + LGPLv3 + GPLv2 license
combination
Change-Id: I6f49b819a8a20cc4f88b794a8f6726d975e8ffbe
Reviewed-by: Matti Paaso <matti.paaso@theqtcompany.com>
|
| |
|
|
|
|
|
|
|
| |
- Renamed LICENSE.LGPL to LICENSE.LGPLv21
- Added LICENSE.LGPLv3
- Removed LICENSE.GPL
Change-Id: Iec3406e3eb3f133be549092015cefe33d259a3f2
Reviewed-by: Iikka Eklund <iikka.eklund@digia.com>
|
| |
|
|
|
|
|
|
|
| |
(?J) inside a pattern string can be used to allow or disallow duplicated
capturing group names in the pattern string itself.
Although PCRE supports duplicated names, in Qt we don't yet.
Change-Id: I21cd0c41273cd7ef42870ced3a0fad6ba7035cbc
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It may be useful to know which named capturing groups are defined
in an regular expression, and for each of them, what's the
corresponding index. This commit adds the needed method
to QRegularExpression.
Note that extracting the information doesn't happen while holding
the mutex in the private -- pcre_fullinfo just reads information
from the compiled pattern, so that's thread-safe.
Task-number: QTBUG-29079
Change-Id: I50c00ee860f06427c2e6ea10417d5c0733cc8303
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
| |
|
|
|
|
|
|
| |
This allows to put them in containers, and to enable subsequent
features for QString.
Change-Id: I3b3fe695ffe6930331ed9f670738376722e0fc36
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
| |
|
|
|
|
|
|
| |
Change copyrights and license headers from Nokia to Digia
Change-Id: If1cc974286d29fd01ec6c19dd4719a67f4c3f00e
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Sergio Ahumada <sergio.ahumada@digia.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Although passing a null pointer to pcre16_get_stringnumber for
the compiled pattern should simply make it error out, it's actually
an undocumented behaviour, so let's stay safe and add an explicit
check.
Tests for this codepath are added.
Change-Id: Ifd9c87874f6812ba487104ec1a5bbc83c3b16761
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
PCRE's JIT uses by default 32K on the pcre_exec caller's stack. This
is fine for most situations, but in some cases (esp. patterns with
lot of recursion) more memory is required.
Therefore, if a match execution fails due to exhausting JIT memory,
we let PCRE allocate up to 512KB to be used for the JIT's stack.
The pointer to the allocated memory is put in thread local storage
(so it can be reused from the same thread, if needed, and automatically
goes away when the thread dies).
Change-Id: Ica5fb7d517068befff88ebb198a603a26ec5d8a7
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
Exporting the counter that controls the optimization of a compiled
pattern lets us to forcibly optimize all patterns. Therefore,
two tests are now run: one with default optimization values
and another one which always optimizes the pattern.
The counter itself was renamed with a qt_ prefix and put
inside the Qt compilation namespace
(thanks to rohanpm for pointing it out).
Change-Id: I56602433d37adc127772b2d0d2cdaf2e49d43c71
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
