error in tableformula or bug

error in tableformula or bug

[Martin Halder]

Hi all,

found some strange behavior which was working before as far as I remember.. I am on commit 078c01b.

this is working correctly:
| 10 |
| 10 |
| 20 |
|----|
| 40 |
#+TBLFM: @>$1=vsum(@1..@3)

this one with relative indexing (@>-1) not: (value is changing every time the formula is applied: 37, 73, 109 (+36 every time)
| 10 |
| 10 |
| 20 |
|----|
| 37 |
#+TBLFM: @>$1=vsum(@1..@>-1)

Thanks for help,
Martin

error in tableformula or bug

[Martin Halder]

Hi again,

have traced it down to the following commit which introduces the problem (thanks to bisect):
[3dd474575205d3808390fc6ea2d5feccdb3d4305] Tables:  Make @< and $< point to row/column 1 in a stable way

Was there a change in the format or is it really a bug ?

Thanks for help,
Martin

> found some strange behavior which was working before as far as I remember.. I am on commit 078c01b.
> 
> this is working correctly:
> | 10 |
> | 10 |
> | 20 |
> |----|
> | 40 |
> #+TBLFM: @>$1=vsum(@1..@3)
> 
> this one with relative indexing (@>-1) not: (value is changing every time the formula is applied: 37, 73, 109 (+36 every time)
> | 10 |
> | 10 |
> | 20 |
> |----|
> | 37 |
> #+TBLFM: @>$1=vsum(@1..@>-1)

Re: error in tableformula or bug

[Carsten Dominik]

On 24.3.2011, at 10:46, Martin Halder wrote:

> Hi again,
> 
> have traced it down to the following commit which introduces the problem (thanks to bisect):
> [3dd474575205d3808390fc6ea2d5feccdb3d4305] Tables:  Make @< and $< point to row/column 1 in a stable way
> 
> Was there a change in the format or is it really a bug ?
> 
> Thanks for help,
> Martin
> 
>> found some strange behavior which was working before as far as I remember.. I am on commit 078c01b.
>> 
>> this is working correctly:
>> | 10 |
>> | 10 |
>> | 20 |
>> |----|
>> | 40 |
>> #+TBLFM: @>$1=vsum(@1..@3)
>> 
>> this one with relative indexing (@>-1) not: (value is changing every time the formula is applied: 37, 73, 109 (+36 every time)
>> | 10 |
>> | 10 |
>> | 20 |
>> |----|
>> | 37 |
>> #+TBLFM: @>$1=vsum(@1..@>-1)

@>-1 was allowed only for a short time, this format is flawed.
Please use @>> instead to mean the second to last row.

- Carsten

> 
> 
> 

Re: error in tableformula or bug

[Nick Dokos]

Carsten Dominik <carsten.dominik@gmail.com> wrote:

> 
> On 24.3.2011, at 10:46, Martin Halder wrote:
> 
> > Hi again,
> > 
> > have traced it down to the following commit which introduces the problem (thanks to bisect):
> > [3dd474575205d3808390fc6ea2d5feccdb3d4305] Tables:  Make @< and $< point to row/column 1 in a stable way
> > 
> > Was there a change in the format or is it really a bug ?
> > 
> > Thanks for help,
> > Martin
> > 
> >> found some strange behavior which was working before as far as I remember.. I am on commit 078c01b.
> >> 
> >> this is working correctly:
> >> | 10 |
> >> | 10 |
> >> | 20 |
> >> |----|
> >> | 40 |
> >> #+TBLFM: @>$1=vsum(@1..@3)
> >> 
> >> this one with relative indexing (@>-1) not: (value is changing every time the formula is applied: 37, 73, 109 (+36 every time)
> >> | 10 |
> >> | 10 |
> >> | 20 |
> >> |----|
> >> | 37 |
> >> #+TBLFM: @>$1=vsum(@1..@>-1)
> 
> @>-1 was allowed only for a short time, this format is flawed.
> Please use @>> instead to mean the second to last row.
> 

I think he wants the penultimate line, not the second line.

Nick

Re: error in tableformula or bug

[Nick Dokos]

Nick Dokos <nicholas.dokos@hp.com> wrote:

> > @>-1 was allowed only for a short time, this format is flawed.
> > Please use @>> instead to mean the second to last row.
> > 
> 
> I think he wants the penultimate line, not the second line.
> 

... and I see that's what you said and I misread it: sorry for the noise.

Nick