[prev in list] [next in list] [prev in thread] [next in thread] 

List:       koffice
Subject:    RE: Patch for KSPREAD
From:       gerke.kok () tpa-nl ! com
Date:       2000-08-23 6:22:53
[Download RAW message or body]

About series:
should not series be entered in selected cells? This is normaly done in
spreadsheets. And then the user defines how many cells should be filled! You
could even call it quick fill and apend the posibility for more helpfull
cell filling like day or month... Or any other serie of numbers...
wkr,
Gerke

> -----Original Message-----
> From: Laurent montel [mailto:montell@club-internet.fr]
> Sent: Wednesday, August 23, 2000 07:12
> To: koffice@master.kde.org
> Subject: Re: Patch for KSPREAD
> 
> 
> Hi,
> 
> David Faure wrote:
> 
> > At 19:50 22/08/00 +0200, you wrote:
> > >Hi,
> > >
> > >I fix some bug :
> > >-Now undo works again when I changed table name
> > >I replaced name() by tableName() into kspread_undo.cc as 
> David told me
> > >
> > >-I implemented undo when I used insert->series
> > >
> > >
> > >-Sorry but I added two new messages because I found a big 
> bug into series:
> > >Algorithm is :
> > >
> > >for ( int incr=start;incr<=end; )
> > >        {
> > >        if(type==Linear)
> > >                incr=incr+step;
> > >        else if(type==Geometric)
> > >                incr=incr*step;
> > >        }
> > >
> > >
> > >but if step==0 you can stop this boucle
> >
> > You mean "loop", but other than that, you're right :-)
> 
> Ok,
> 
> >
> >
> > >and if step <=1 you can stop it (in geometric mode)
> > >
> > >So I added this two message (in kspread_dlg_serie ) :
> > >KMessageBox::error( this, i18n("Sorry but step should be 
> supperior to zero\n"
> > "superior", one p
> >
> > >                "Otherwise linear series is infinite!") );
> > "Otherwise the linear sery is infinite."
> > Hmm, I'm not sure about 'sery'.
> >
> > >KMessageBox::error( this, i18n("Sorry but step should be 
> supperior to one\n"
> > >                "Otherwise geometric series is infinite!") );
> >
> > That's not exactly true... you can have a geometric sery with
> > a step of 0.5, going from 24 to 1....
> 
> Yes it's right, but step is an integer, because if a people 
> insert a serie :
> start=0; end=100; step =0.01
> there is 10000 cells will create, so I am not sure that it's 
> a good idea.
> 
> >
> >
> > Oh, and you can have a linear sery going from 50 to -15 with a step
> > of -5.
> > The check to do is :
> > (end - begin) / step = integer number
> > (i.e. "(end - begin) can be divided by the step")
> 
> I think that I am going to modify my algo to allow your method.
> Because for the moment start, end, step should be positive. But I
> think that it's important to create a decrease series.
> 
> >
> >
> > David.
> 
> Bye
> 

[prev in list] [next in list] [prev in thread] [next in thread]