monday.
I've integrated it, you don't have to worry about closing this out. Pity we
haven't heard from Patrick though, I hope he's OK.
Which Patrick? At least the Spendrin one is OK. I've seen him today :)
Cheers
--
Milian Wolff
m...@milianw.de
http://milianw.de
?
There are tons more cleanups required in these interfaces.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
signature.asc
Description: This is a digitally signed message part.
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
get a truly verbose flag, which actually outputs
the output from whatever tool is currently running? For me as a developer, its
really annoying having to tail -f on some random log files to get my hands on
the make output... How are other developers handling this?
Bye
--
Milian Wolff
m
/jpegloader.cpp:430:29: warning: cast
from 'uchar *' (aka 'unsigned char *') to 'QRgb *' (aka 'unsigned int *')
increases required alignment from 1 to 4 [-Wcast-align]
QRgb *out = (QRgb *)scanline;
^~~~
Bye
--
Milian Wolff
m...@milianw.de
http
-options -DCMAKE_BUILD_TYPE:STRING=debug
end module-set
Considering that all other people should hit the same issue - how did you
resolve this? Can we get the above into the default configuration somehow
please?
Thanks
--
Milian Wolff
m...@milianw.de
http://milianw.de
On Saturday 01 March 2014 16:42:31 David Faure wrote:
On Wednesday 26 February 2014 22:25:01 Milian Wolff wrote:
module-set
repository kde-projects
branch qt5
use-modules polkit-qt-1
cmake-options -DCMAKE_BUILD_TYPE:STRING=debug
end module-set
Considering
?
Thanks
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
114101d525309e423f50c5164d83afee8c634413
Diff: https://git.reviewboard.kde.org/r/116544/diff/
Testing
---
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/116542/#review51851
---
On March 2, 2014, 8:20 p.m., Milian Wolff wrote
,
^
static_castint()
Diffs
-
src/kjs/create_hash_table 94f3e4358a6d78fc7c658369d65b0e75ca131bc8
Diff: https://git.reviewboard.kde.org/r/116545/diff/
Testing
---
Thanks,
Milian Wolff
___
Kde-frameworks
? Is it really worth it here?
Diffs
-
src/xml/dom_stringimpl.h 114101d525309e423f50c5164d83afee8c634413
Diff: https://git.reviewboard.kde.org/r/116544/diff/
Testing
---
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks
};
^~
static_castunsigned int( )
Diffs
-
src/misc/AtomicString.cpp 4c35662ddd47759ed7e70524ededdd5223c14c09
Diff: https://git.reviewboard.kde.org/r/116541/diff/
Testing
---
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde
On March 19, 2014, 2:56 a.m., Michael Pyne wrote:
src/lib/util/kuser_unix.cpp, line 349
https://git.reviewboard.kde.org/r/116883/diff/1/?file=255233#file255233line349
Does this need to be a template, or would std::function be sufficient?
Templates have a poor reputation amongst
. And if this becomes the
bottleneck for an application, it should not use KIO in the first place
probably but rather QDir directly. KIO's biggest feature is the network
transparency after all. And if you add network latency in the mix, the CPU
time becomes negligible.
Bye
--
Milian Wolff
m
On June 6, 2014, 10:34 p.m., David Faure wrote:
Interestingly, I had a benchmark to compare a number of data structures for
UDSEntry (which made me turn the Qt3 QList into the Qt4 QHash). That
benchmark is the commented-out newApiPerformance() method in jobtest.cpp.
I just extended
to modify something is odd. Rather, I'd expect that a edit operation would
invalidate/update the cache as needed. So in the best case, one would just have
lists/maps of const KFileItem - so pure value semantics without any pointers.
- Milian Wolff
On June 16, 2014, 7:58 a.m., Frank Reininghaus wrote
a kf5 app tried to load a Qt4
plugin/library, which lead to _very_ strange crashes. You can make sure that's
not the case by doing something like ldd on any exe/lib involved. Furthermore,
try info shared in gdb on crash.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
/x86_64/start.S:122
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
On Monday 30 June 2014 00:05:10 šumski wrote:
On Thursday 26 of June 2014 12:14:49 Milian Wolff wrote:
Hey,
did you run it through valgrind?
Here's what valgrind says:
Sounds like a bug in Qt to me, I have to say. Looking at the code,
QDBusConnectionPrivate::objectDestroyed looks
On July 4, 2014, 7:22 p.m., David Faure wrote:
recursive mutexes are more expensive. Can you try this patch instead?
http://www.davidfaure.fr/2014/kprotocolmanager.cpp.diff
David Faure wrote:
Actually, nothing beats a unittest :) Added, fix pushed.
]lock.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119242/#review62402
---
Ship it!
Ship It!
- Milian Wolff
On July 12, 2014, 9:43
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119241/#review62401
---
Ship it!
Ship It!
- Milian Wolff
On July 12, 2014, 9:41
/#comment22055
This is missing const for the workingDirectory parameter.
(Sorry, forgot to publish that review - I hate that on ReviewBoard :-/)
- Milian Wolff
On March 18, 2013, 7:51 p.m., Martin Tobias Holmedahl Sandsmark wrote
://git.reviewboard.kde.org/r/111689/#comment26951
why not directly file.write(jdoc.toJson())?
staging/kservice/tools/desktoptojson/kconfigtojson.cpp
http://git.reviewboard.kde.org/r/111689/#comment26952
will be done automatically when the function's scope is left.
- Milian Wolff
On July 25, 2013, 4:10
://git.reviewboard.kde.org/r/111688/#comment26956
static const QString metaData = QStringLiteral(MetaData);
Also, all the strings below should use QStringLiteral.
- Milian Wolff
On July 25, 2013, 4:10 p.m., Sebastian Kügler wrote
://git.reviewboard.kde.org/r/111688/#comment27169
I'd leave these unrelated changes out and do it in a separate cleanup
commit. just new code should not need these cleanups afterwards.
- Milian Wolff
On July 30, 2013, 2:41 a.m., Sebastian Kügler wrote
://git.reviewboard.kde.org/r/111689/#comment27215
I'm not sure about the style guide in KDELibs, but aren't the
KLocalizedString kind of includes preferred? Same for the Qt files.
- Milian Wolff
On July 30, 2013, 7:45 p.m., Sebastian Kügler wrote
();
if (!valid(var)) {
return false;
}
do_stuff();
}
staging/kservice/tools/desktoptojson/kconfigtojson.cpp
http://git.reviewboard.kde.org/r/111689/#comment27246
merge with above to else if?
- Milian Wolff
On Aug. 1, 2013, 6:03 p.m., Sebastian Kügler wrote
On Aug. 1, 2013, 6:14 p.m., Milian Wolff wrote:
staging/kservice/tools/desktoptojson/kconfigtojson.cpp, line 60
http://git.reviewboard.kde.org/r/111689/diff/6/?file=175868#file175868line60
just do return convert(...);
Nicolás Alvarez wrote:
No, that would give the opposite
---
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/111689/#review36947
---
Ship it!
awesome, I like it now :)
- Milian Wolff
On Aug
://git.reviewboard.kde.org/r/111689/#comment27345
why members? that's imo a verschlimmbesserung. If at all, mark them as
static const.
- Milian Wolff
On Aug. 3, 2013, 10:07 p.m., Sebastian Kügler wrote:
---
This is an automatically
On Aug. 2, 2013, 9:08 a.m., David Faure wrote:
staging/kservice/tools/desktoptojson/kconfigtojson.h, line 31
http://git.reviewboard.kde.org/r/111689/diff/8/?file=175878#file175878line31
I don't like file-static QStrings (global objects), and even less in
headers (it creates an
://git.reviewboard.kde.org/r/111689/#comment27507
that should be const,no?
- Milian Wolff
On Aug. 5, 2013, 11:29 p.m., Sebastian Kügler wrote:
---
This is an automatically generated e-mail. To reply, visit:
http
accidentally be misused.
Singletons are an anti-pattern.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
some result out
of that should be made thread safe.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
On Wednesday 06 August 2014 17:29:40 Vishesh Handa wrote:
On Wednesday, August 06, 2014 04:09:19 PM Milian Wolff wrote:
Hmm, how would we do async? I thought people could just run it in
another
thread / process if they want. The only thing I can think of changing is
that the plugins
) do you share the git history, i.e. did you do a proper fork, or did you
reimport the sources and started from scratch (I hope not).
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119714/#review64270
---
Ship it!
Ship It!
- Milian Wolff
On Aug. 11, 2014, 1:24
for Hugo is doing and has done for KDE. But
since I have no clue about styles (I even said so!) I wanted to ask what the
advantage of a native style over a QtCurve config file are. Hugo answered
that. So whats the deal Martin?
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
Hugo's commits on a fork of
QtCurve? Having the history in another repository means it does not exist.
People won't bother finding the file (esp. if it eventually gets renamed or
moved) in another repository just to do a git blame on it in the future.
Bye
--
Milian Wolff
m...@milianw.de
http
On Tuesday 12 August 2014 23:34:40 Àlex Fiestas wrote:
On Monday 11 August 2014 12:51:21 Milian Wolff wrote:
a) what's the advantage of having a native widget style, compared to using
QtCurve settings? Are there things missing / not implementable in QtCurve?
We did some rought tests
://git.reviewboard.kde.org/r/119970/#comment45705
ret.reserve(list.size());
src/services/kplugininfo.cpp
https://git.reviewboard.kde.org/r/119970/#comment45706
ret.reserve(list.size());
- Milian Wolff
On Aug. 28, 2014, 11:13 a.m., Alexander Richardson wrote
On Aug. 28, 2014, 12:07 p.m., Milian Wolff wrote:
src/services/kplugininfo.cpp, line 523
https://git.reviewboard.kde.org/r/119970/diff/1/?file=307971#file307971line523
don't create the string literals multiple times, i.e. Name at least is
twice here which increases the lib size
of that framework so someone else
will have to OK it as well, I guess.
- Milian Wolff
On Aug. 28, 2014, 12:56 p.m., Alexander Richardson wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r
/#comment45852
personally, I'd prefer the C++ way of casting over unsafe C-style casts
every time.
Also, what is the magic number 8? That should probably be sizeof(quint64),
no?
- Milian Wolff
On Aug. 31, 2014, 2:46 p.m., Albert Astals Cid wrote
/#comment45855
Ok, I'm not affiliated to this project just got interested by the ASAN in
the review title ;-)
Ignore this at will. Though the maintainer of this code I'd urge to rewrite
it to use sizeof(quint64) and c++ casts instead.
- Milian Wolff
On Aug. 31, 2014, 2:46 p.m., Albert
procedure takes longer just because we need to compress
every icon, then trading size against time is not the best compromise ;)
You'd only need to compress it when the file changed. Otherwise incremental
builds will just install the compressed files in your build folder.
Bye
--
Milian Wolff
m
/Scalable_Vector_Graphics#Compression
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
/k7zip.cpp
https://git.reviewboard.kde.org/r/120096/#comment46119
this would be obsoleted if you use a QByteArray above directly
- Milian Wolff
On Sept. 7, 2014, 3:20 p.m., Patrick Spendrin wrote:
---
This is an automatically
a look at it as well I guess
- Milian Wolff
On Sept. 9, 2014, 9:14 a.m., Patrick Spendrin wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120096
= ...;
if (size encodedBuffer.size()) {
encodedBuffer.resize(size);
}
...
}
:)
- Milian Wolff
On Sept. 9, 2014, 10:19 a.m., Patrick Spendrin wrote
good enough nowadays. Otherwise I agree that we'll need some information
on the accuracy of the match.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman
same as above
- Milian Wolff
On Sept. 12, 2014, 12:16 p.m., Fabio D'Urso wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120160
wrap these things in
anonymous namespaces to give them file-local linkage?
- Milian Wolff
On Sept. 13, 2014, 4:02 p.m., Volker Krause wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120187/#review66429
---
Ship it!
yay
- Milian Wolff
On Sept. 13, 2014, 5:19 p.m
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120200/#review66473
---
Ship it!
Ouch, nice catch!
- Milian Wolff
On Sept. 14
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120208/#review66497
---
Ship it!
Ship It!
- Milian Wolff
On Sept. 14, 2014, 9:23
upstream. 5.3 is probably a good choice for
now, until 5.4 is released.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
/kmessagewidgetautotest.cpp
https://git.reviewboard.kde.org/r/120160/#comment46609
similar to above
- Milian Wolff
On Sept. 18, 2014, 10:40 a.m., Fabio D'Urso wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
need to override
it at all when you just delegate to the base class anyways?
tests/krcctest.cpp
https://git.reviewboard.kde.org/r/120075/#comment47570
I knew it, you are a time traveler! :P
- Milian Wolff
On Oct. 10, 2014, 8:46 p.m., David Faure wrote
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120075/#review68270
---
Ship it!
Ship It!
- Milian Wolff
On Oct. 11, 2014, 3:28
On Oct. 11, 2014, 2:38 p.m., Milian Wolff wrote:
src/krcc.h, line 93
https://git.reviewboard.kde.org/r/120075/diff/1/?file=310011#file310011line93
scoped pointer?
David Faure wrote:
Hmm. Not done anywhere else in karchive or frameworks in general. Not
much point, now
.
snip
- with bugzilla integration (additionally to the previous ones):
KF5::WebKit
Could this maybe be replaced by using QtWebKit directly?
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks
. For these two, it is OK to add the / as default path.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
to
have a useful subset of C++11 (case in point ktexteditor which prompted this
thread builds with it).
Yes, let's go for 2012 everywhere!
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel
:
https://bugzilla.redhat.com/show_bug.cgi?id=1056769
Diffs
-
src/solid/devices/backends/upower/upowermanager.cpp
53c858093a1c439f0faca0c956a51199f4e882e4
Diff: https://git.reviewboard.kde.org/r/121007/diff/
Testing
---
warning gone!
Thanks,
Milian Wolff
Hey all,
what do I do to get middle-click-closes-tab in Qt 5 without KTabBar?
Previously, we used KTabBar and its mouseMiddleClick signal.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks
of maintaining it) is worth it, but it would be best to first
determine if the current code path can be optimized.
True.
Bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https
/services/kplugininfo.cpp
https://git.reviewboard.kde.org/r/120198/#comment48924
inline this as above, use QString and take the service smart ptr by const
- Milian Wolff
On Sept. 14, 2014, 2:05 p.m., Alexander Richardson wrote
(s_serviceTypesKey).toArray().contains(serviceType);
- Milian Wolff
On Sept. 14, 2014, 2:06 p.m., Alexander Richardson wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120199
, KDevelop and
other projects already battle for that resource. Adding one more watcher won't
simplify that situation. Generally, there will be occasions where adding a
watcher failed. You cannot rely on that feature to work reliably.
bye
--
Milian Wolff
m...@milianw.de
http://milianw.de
/r/120198/#comment48935
ugh. why not an anon namespace? I'd still use `const foo QString =
QStringLiteral(...)` personally...
- Milian Wolff
On Nov. 6, 2014, 5:50 p.m., Alexander Richardson wrote:
---
This is an automatically
/r/120199/#comment48936
here and below: QStringLiteral
- Milian Wolff
On Nov. 6, 2014, 5:49 p.m., Alexander Richardson wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120199
---
warning gone!
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
://git.reviewboard.kde.org/r/121007/diff/
Testing
---
warning gone!
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
/upowermanager.cpp
53c858093a1c439f0faca0c956a51199f4e882e4
Diff: https://git.reviewboard.kde.org/r/121007/diff/
Testing
---
warning gone!
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman
---
On Nov. 7, 2014, 1:33 p.m., Milian Wolff wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121007
that
indexOfSignal(DeviceAdded(QDBusObjectPath)) works?
(simpler than downgrading UPower).
Milian Wolff wrote:
Yes, I did check that. You can also iterate over the QMetaObject's
QMetaProperties and print their names. I did see
`DeviceAdded(QDBusObjectPath)` and friends in there. I assume
.
Google is full of that warning, and there is also:
https://bugzilla.redhat.com/show_bug.cgi?id=1056769
Diffs (updated)
-
CMakeLists.txt f7c500e
Diff: https://git.reviewboard.kde.org/r/121007/diff/
Testing
---
warning gone!
Thanks,
Milian Wolff
that
indexOfSignal(DeviceAdded(QDBusObjectPath)) works?
(simpler than downgrading UPower).
Milian Wolff wrote:
Yes, I did check that. You can also iterate over the QMetaObject's
QMetaProperties and print their names. I did see
`DeviceAdded(QDBusObjectPath)` and friends in there. I assume
Diff: https://git.reviewboard.kde.org/r/121007/diff/
Testing
---
warning gone!
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
? See also
https://git.reviewboard.kde.org/r/120592/
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
---
On Nov. 10, 2014, 1:50 p.m., Milian Wolff wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121093
was the isseu on linux?
Milian Wolff wrote:
The `_p.h` files do not contain export macros. I showed you some of the
build errors above.
Patrick von Reth wrote:
Ok the problem on windows is that all other files contain exports ^^
So how do we fix this then? Maybe just by defining
://git.reviewboard.kde.org/r/120592/
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
also
https://git.reviewboard.kde.org/r/120592/
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
an issue on Windows or
why did you push this change? See also
https://git.reviewboard.kde.org/r/120592/
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
--
Milian Wolff
m...@milianw.de
http://milianw.de
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
.
Thanks,
Milian Wolff
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
/121218/#comment49682
is the call to `setCheckable` not required anymore? b/c that is otherwise
missing on the new patch now, no?
- Milian Wolff
On Nov. 28, 2014, 8:02 p.m., Alexander Richardson wrote:
---
This is an automatically
On Nov. 30, 2014, 2:29 p.m., Milian Wolff wrote:
src/kstandardaction.cpp, line 153
https://git.reviewboard.kde.org/r/121218/diff/3/?file=330993#file330993line153
is the call to `setCheckable` not required anymore? b/c that is
otherwise missing on the new patch now
---
On Nov. 27, 2014, 1:15 a.m., Milian Wolff wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121263
On Dec. 10, 2014, 7:40 p.m., Mark Gaiser wrote:
src/core/udsentry.cpp, line 99
https://git.reviewboard.kde.org/r/118452/diff/2/?file=332446#file332446line99
micro optimization, but d-fields[index] is - in my experience -
slightly faster.
nope, that is not true.
On Dec. 10,
On Dec. 10, 2014, 7:40 p.m., Frank Reininghaus wrote:
As for all the appends that you do on a QVector, you might be able to speed
them up significantly if you know the size (which you do).
The reason for that potential speedup is append doing some stuff to
determine if it can append or
. But I really don't think its worth it. `at()` and
`operator[]` are equal on const containers. on non-const containers,
`operator[]` could lead to detaching though. I'd be suprised if it is ever
going to be faster than `at()`.
- Milian Wolff
On Dec. 9, 2014, 10:44 p.m., Frank Reininghaus
pull out more.
Milian Wolff wrote:
your benchmark is flawed: your second loop also contains branches, which
the former does not. Both iterate over the same size, no? So you just want to
compare resize + operator[] vs. reserve + append? Yes, the former will be
faster, but afaik only
pull out more.
Milian Wolff wrote:
your benchmark is flawed: your second loop also contains branches, which
the former does not. Both iterate over the same size, no? So you just want to
compare resize + operator[] vs. reserve + append? Yes, the former will be
faster, but afaik only
On Dec. 10, 2014, 11:03 p.m., Milian Wolff wrote:
LGTM, some nitpicks. And of course we need the results of the benchmark in
release mode.
NOTE: I just pushed some fixes to udsentrybenchmark, which was pretty broken in
itself. It contained lots of unrelated stuff in its hotpath. Now
`.toList()`. Since
before it returned `.keys()`, and now you return `.toList()`, I assume that is
not often called anyways, so that would be OK.
- Milian Wolff
On Dec. 9, 2014, 10:44 p.m., Frank Reininghaus wrote
1 - 100 of 489 matches
Mail list logo