On 06/13/2012 11:07 PM, Till Oliver Knoll wrote:
> Am 13.06.12 22:03, schrieb Harri Pasanen:
>> I started working on a bit more substantial QML application for symbian
>> and ran into some issues. One of these is the startup time with 10+ QML
>> non trivial files, and I'm not finished adding them.
On 06/14/2012 12:32 AM, Thiago Macieira wrote:
> On quarta-feira, 13 de junho de 2012 22.50.29, Harri Pasanen wrote:
>> Can the V8 state be dumped on disk? If yes, it could do like python,
>> compile the first run on target and dump the result for subsequent
>> runs. (Although python bytecode is
Am 14.06.2012 um 01:52 schrieb Josiah Bryan :
> ...
> I'll double-check globals, but a stack trace is really what's needed
> here anyway - debugger just says "Application exited with exit code
> exit-code="03"" (Yes, I googled that - no, the only three results I
> found did not help.)
What comp
On 6/13/12, André Pönitz wrote:
> On Wed, Jun 13, 2012 at 04:25:59PM -0400, Josiah Bryan wrote:
>> I'm attempting to debug an odd problem wherein my program (http://
>> code.google.com/p/dviz/) crashes when displaying certain UTF-8
>> strings. That's not really why I'm emailing - the problem I'd l
You are missing that the baseline is free access to _the source code_
_and_ a well-defined, provably working, reasonably fast way to get
_your_ changes in, and influence the direction the whole thing goes with
_your_ ideas.
Points two and three are _quite_ different from what we had at Trolltech
On 13.06.2012 23:35, Jan Kundrát wrote:
> On 06/13/12 22:50, Jochen Becher wrote:
>> Of course it is not a bad idea to access siblings in a destructor. Why
>> should it? If destruction follows in a known and reliable way there is
>> no reason to avoid it. I do it to check some post-conditions using
On quarta-feira, 13 de junho de 2012 22.50.29, Harri Pasanen wrote:
> Can the V8 state be dumped on disk? If yes, it could do like python,
> compile the first run on target and dump the result for subsequent
> runs. (Although python bytecode is machine independent).
>
> I'm not familiar with QML
On quarta-feira, 13 de junho de 2012 23.56.46, André Pönitz wrote:
> If deterministic order of destruction is needed one can always do it
> "manually" in the destructor body of the QObject-derived parent class
> instead of waiting for the parent's base class desctructor with
> it's "non-determinist
On 12 June 2012 15:26, Jan Kundrát wrote:
> So, the questions:
>
> - Is there a publicly available way of doing what I want to do, ie. am I
> missing something?
Almost certainly not. In KDE we implemented our own KDateTime and
KTimeZone classes to handle it properly, although QDateTime is used
On Wed, Jun 13, 2012 at 12:33:02PM -0700, David Ching wrote:
> Hello,
>
> A couple days ago Turunen Tuukka from Digia responded to my question
> here about getting an SDK with Qt 4.8.2 in it by saying
>
> > There is no Qt Project SDK for 4.8 - just the Nokia SDK and Qt
> > Commercial SDK.
>
> Do
On Wed, Jun 13, 2012 at 04:25:59PM -0400, Josiah Bryan wrote:
> I'm attempting to debug an odd problem wherein my program (http://
> code.google.com/p/dviz/) crashes when displaying certain UTF-8
> strings. That's not really why I'm emailing - the problem I'd like
> your help with is that I'm unabl
On Wed, Jun 13, 2012 at 11:35:16PM +0200, Jan Kundrát wrote:
> Hi Jochen,
>
> On 06/13/12 22:50, Jochen Becher wrote:
> > Of course it is not a bad idea to access siblings in a destructor. Why
> > should it? If destruction follows in a known and reliable way there is
> > no reason to avoid it. I d
Hello List!
I am using a QSqlTableModel to display the contents of a database table
in a QTableView. (I am building a lightweight application to edit some
small database tables.)
I now want to determine whether a given data item exists in the table: E.g.,
is the result of
select count(*) fr
On 6/13/2012 11:51 AM, Konrad Rosenbaum wrote:
Hi,
On Wednesday 13 June 2012 17:27:20 Carl Schumann wrote:
Where MccModel subclassed from QAbstractTableModel.This did not
work, specifically it displayed the whole long rich text string in
black. How should I be doing multicolored table cel
Hi Jochen,
On 06/13/12 22:50, Jochen Becher wrote:
> Of course it is not a bad idea to access siblings in a destructor. Why
> should it? If destruction follows in a known and reliable way there is
> no reason to avoid it. I do it to check some post-conditions using
> Q_ASSERT() to ensure all resou
Am 13.06.12 22:03, schrieb Harri Pasanen:
> I started working on a bit more substantial QML application for symbian
> and ran into some issues. One of these is the startup time with 10+ QML
> non trivial files, and I'm not finished adding them.
Wait a sec... what is a QML file, what does it con
On 13 June 2012 21:27, Konrad Rosenbaum wrote:
> Hi,
>
> On Wednesday 13 June 2012 11:29:36 Christian Gagneraud wrote:
>> Will Qt5 have better timezone support?
>
> 5.0.x - no, it has about the same features in QDateTime as Qt4.x
>
> 5.1.x - maybe, depends on whether anyone actually has time to in
Can the V8 state be dumped on disk? If yes, it could do like python,
compile the first run on target and dump the result for subsequent
runs. (Although python bytecode is machine independent).
I'm not familiar with QML inner workings though, especially where does
the split with C++ and javasc
Of course it is not a bad idea to access siblings in a destructor. Why
should it? If destruction follows in a known and reliable way there is
no reason to avoid it. I do it to check some post-conditions using
Q_ASSERT() to ensure all resources were freed and that involves some
checks in siblings.
I thought he V8 engine (which is used in Qt5) would have that feature already,
since it gets a lot of performance by whole-method compiling the JS. However my
understanding is the output of V8 is machine code, and therefore, not
portable.And therefore, you'd need an intermediate layer, and how m
Hi,
On Wednesday 13 June 2012 11:29:36 Christian Gagneraud wrote:
> Will Qt5 have better timezone support?
5.0.x - no, it has about the same features in QDateTime as Qt4.x
5.1.x - maybe, depends on whether anyone actually has time to integrate it
into Qt
We talked about it at the Contributor S
I'm attempting to debug an odd problem wherein my program (http://
code.google.com/p/dviz/) crashes when displaying certain UTF-8 strings.
That's not really why I'm emailing - the problem I'd like your help with is
that I'm unable to run the program in "Debug" mode in Qt Creator on Windows
7.
I'v
On quarta-feira, 13 de junho de 2012 22.03.22, Harri Pasanen wrote:
> I started working on a bit more substantial QML application for symbian
> and ran into some issues. One of these is the startup time with 10+ QML
> non trivial files, and I'm not finished adding them.
>
> Has there ever been tho
Hi,
On Wednesday 13 June 2012 17:27:20 Carl Schumann wrote:
> Where MccModel subclassed from QAbstractTableModel.This did not
> work, specifically it displayed the whole long rich text string in
> black. How should I be doing multicolored table cells please?
Use a custom delegate instead: d
I started working on a bit more substantial QML application for symbian
and ran into some issues. One of these is the startup time with 10+ QML
non trivial files, and I'm not finished adding them.
Has there ever been thought for a bytecode compiler, that would store a
pre-parsed representation
Hello,
A couple days ago Turunen Tuukka from Digia responded to my question here
about getting an SDK with Qt 4.8.2 in it by saying
> There is no Qt Project SDK for 4.8 -
> just the Nokia SDK and Qt Commercial SDK.
Does this mean there are two official SDK's now, and the commercial one is
diff
Hello List!
I have a single-column database table containing a small number of items
with which I populate a QComboBox. (I am using this table to emulate an
enum in sql.)
Looking at various examples, I came up with a scheme using QSqlTable
model that seems to be a "Qt-approved" way of doing this
Hi,
Use role Qt::ForegroundRole and Qt::BackgroundRole for implement this, ex:
QVariant MccModel::data(const QModelIndex &index, int role) const
{
if ( role == Qt::BackgroundRole ){
return QVariant(QColor(200,200,200));
}
> Date: Wed, 13 Jun 2012 10:27:20 -0500
> From: schum...@fna
Am 12.06.2012 um 21:36 schrieb Jochen Becher :
> In my case childB is not a child of childA but both are children of
> parent (widgets in the same group/layout). But childB has a reference to
> childA which might be accessed in destructor.
Smells like bad gasoline: usually widgets ("Views") shoul
Hi,
On Wed, Jun 13, 2012 at 5:27 PM, Carl Schumann wrote:
> Qt community,
>
> I would like to have multicolored text in a single QTableView cell.
> Since labels can support multicolored text with a rich text string I
> tried the following:
> > QVariant MccModel::data(const QModelIndex &index, in
Qt community,
I would like to have multicolored text in a single QTableView cell.
Since labels can support multicolored text with a rich text string I
tried the following:
> QVariant MccModel::data(const QModelIndex &index, int role) const
> {
> if( role == Qt::DisplayRole ){
> ret
Ah, and so we have to ask the question - what does an additional connection
mean in terms of your application?
I think it best to !assume! that if you get another connection you should drop
the first. This is because the client software may have restarted. Alternate
scenarios are that you want
On 12/06/12 19:05, Thiago Macieira wrote:
> On terça-feira, 12 de junho de 2012 17.48.14, Jan Kundrát wrote:
>>> You're entirely on your own if you touch the internals. Behaviour might
>>> change in any release.
>>
>> Fair enough. The better question is then "any chances of making these
>> two fun
33 matches
Mail list logo
