> And all those extra member functions could just as well be separate
> functions.
>
> It seems that even the std::string is now considered too fat. Most of
> what it does is possible to implement outside of the class.
>
> So I am not argueing against functions that operate on strings. I am
> argue
Kuba Ober <[EMAIL PROTECTED]> writes:
| On czwartek 09 grudzieŠ2004 10:42 am, Lars Gullik Bjønnes wrote:
>> Kuba Ober <[EMAIL PROTECTED]> writes:
>> >> Are you sure of this. I know that the std::string in libstdc++ has
>> >> gotten quite a bit of performance tweaks lately.
>> >> (gcc 3.4.x and
On czwartek 09 grudzieÅ 2004 10:42 am, Lars Gullik BjÃnnes wrote:
> Kuba Ober <[EMAIL PROTECTED]> writes:
> >> Are you sure of this. I know that the std::string in libstdc++ has
> >> gotten quite a bit of performance tweaks lately.
> >> (gcc 3.4.x and 4.x)
> |
> | So they finally play catch up with
Kuba Ober <[EMAIL PROTECTED]> writes:
>> Are you sure of this. I know that the std::string in libstdc++ has
>> gotten quite a bit of performance tweaks lately.
>> (gcc 3.4.x and 4.x)
>
| So they finally play catch up with Qt. Nice :)
>
| Even then, it's essentially just a dumb container. You'd exp
> And I certainly don't think the Trolltech implementation will outperform
> either gcc's or VC++'s implementation. Performance is not exactly a
> highly visible target in Qt development...
I don't think it will outperform, but it should be on par with. It's really
not rocket science, at least th
> Are you sure of this. I know that the std::string in libstdc++ has
> gotten quite a bit of performance tweaks lately.
> (gcc 3.4.x and 4.x)
So they finally play catch up with Qt. Nice :)
Even then, it's essentially just a dumb container. You'd expect something more
of a useful string implement
On Wed, Nov 24, 2004 at 05:05:00PM -0500, Kuba Ober wrote:
> > Right now they are re-inventing STL. It's called 'Tulip'. The nice
> > thing that I can see so far is that one does not have to use it.
>
> I don't think this is such a bad idea. Even if you think about performance
> only. E.g. if you
Kuba Ober <[EMAIL PROTECTED]> writes:
| On poniedziaÅek 22 listopad 2004 05:58 am, Andre Poenitz wrote:
>> On Mon, Nov 22, 2004 at 10:12:05AM +0100, Lars Gullik Bjønnes wrote:
>> > >> | + v->setBottom(glen);
>> > >> | + return v;
>> > >> | +}
>> > >>
>> > >> I really dont' like these returning p
On poniedziaÅek 22 listopad 2004 05:58 am, Andre Poenitz wrote:
> On Mon, Nov 22, 2004 at 10:12:05AM +0100, Lars Gullik BjÃnnes wrote:
> > >> | + v->setBottom(glen);
> > >> | + return v;
> > >> | +}
> > >>
> > >> I really dont' like these returning pointers.
> > |
> > | This is 'best Qt practice'.
On Mon, Nov 22, 2004 at 10:12:05AM +0100, Lars Gullik Bjønnes wrote:
> >> | +v->setBottom(glen);
> >> | +return v;
> >> | +}
> >>
> >> I really dont' like these returning pointers.
> >
> | This is 'best Qt practice'. The pointees life time is taken care of by
> | its parent, and th
On Mon, Nov 22, 2004 at 10:32:18AM +0100, Juergen Spitzmueller wrote:
> > LyXLength len;
>
> Shouldn't len be (explicitely) initialized?
It's default constructed.
If the result is different from
LyXLength len = LyXLength()
the copy constructor has extremely unusual behaviour.
Andre'
Lars Gullik Bjønnes wrote:
> I can live with this.
committed.
Jürgen
Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
| Partially changed patch attached.
I can live with this.
--
Lgb
Lars Gullik Bjønnes wrote:
> | Index: QExternalDialog.C
> | ===
> | RCS file:
> | /usr/local/lyx/cvsroot/lyx-devel/src/frontends/qt2/QExternalDialog.C,v
> | retrieving revision 1.22
> | diff -p -u -r1.22 QExternalDialog.C
> | --- QExte
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Mon, Nov 22, 2004 at 09:24:41AM +0100, Lars Gullik Bjønnes wrote:
>> | - v->setBottom(LyXLength());
>> | + LyXLength len = LyXLength();
>>
>> LyXLength len;
>
| v->setBottom(LyXLength()) is just fine.
>
>> | + v->setBottom(glen);
>> | + return v;
On Mon, Nov 22, 2004 at 08:50:52AM +0100, Juergen Spitzmueller wrote:
> Andre Poenitz wrote:
> > I think I saw some code similar to
> >
> > void foo(LyXLength len);
> >
> > This should be
> >
> > void foo(LyXLength const & len);
>
> Corrected in the attached patch. I will commit this if I he
Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
| Andre Poenitz wrote:
>> I think I saw some code similar to
>>
>> void foo(LyXLength len);
>>
>> This should be
>>
>> void foo(LyXLength const & len);
>
| Corrected in the attached patch. I will commit this if I hear no more
| objections.
>
Andre Poenitz wrote:
> I think I saw some code similar to
>
> void foo(LyXLength len);
>
> This should be
>
> void foo(LyXLength const & len);
Corrected in the attached patch. I will commit this if I hear no more
objections.
Jürgen
Index: QExternalDialog.C
=
On Sun, Nov 21, 2004 at 01:23:54PM +0100, Juergen Spitzmueller wrote:
> Andre Poenitz wrote:
> > > Comments?
> >
> > LyX[Glue]Length arguments should be passed by reference.
>
> I don't understand.
I think I saw some code similar to
void foo(LyXLength len);
This should be
void foo(LyXLen
Alfredo Braunstein wrote:
> I think that he means just to replace
> boolÂLengthValidator::setBottom(LyXGlueLengthÂg)
> by
> boolÂLengthValidator::setBottom(LyXGlueLengthÂconst & g)
> and so on, whenever possible.
I see. OK, I'll change this.
Regards,
JÃrgen
> Regards, Alfredo
Juergen Spitzmueller wrote:
>> LyX[Glue]Length arguments should be passed by reference.
>
> I don't understand.
Hi Juergen,
I think that he means just to replace
boolÂLengthValidator::setBottom(LyXGlueLengthÂg)
by
boolÂLengthValidator::setBottom(LyXGlueLengthÂconst & g)
and so on, whenever
Andre Poenitz wrote:
> > Comments?
>
> LyX[Glue]Length arguments should be passed by reference.
I don't understand.
Jürgen
On Fri, Nov 19, 2004 at 03:09:34PM +0100, Juergen Spitzmueller wrote:
> The attached patch adds a glue length validator to qt/lengthvalidator
> and uses it in QVSpace (this is related to bug . More dialogs to
> come.
>
> Comments?
LyX[Glue]Length arguments should be passed by reference.
Andre'
The attached patch adds a glue length validator to qt/lengthvalidator and uses
it in QVSpace (this is related to bug . More dialogs to come.
Comments?
Jürgen
Index: QVSpace.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/fronte
24 matches
Mail list logo