Re: [GRASS-user] Re: nc_spm_08 - 'cat' double

[hamish-2]

Martin wrote:
> in GRASS 7 we can support as key column also non-integer
> columns,


"just because we can, doesn't mean we should" ...



Hamish




_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev

[hamish-2]

ps- if cleaning up NC data please fix UID.GID in tarball and
remove cruft from mapset dirs (.tmp, etc.)


thanks
Hamish




_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev

[Martin Landa]

2009/10/18 Hamish <[hidden email]>:
> Martin wrote:
>> in GRASS 7 we can support as key column also non-integer
>> columns,
>
>
> "just because we can, doesn't mean we should" ...

why key column should be restricted only for integers? If you use
unique strings as keys I don't see any problem (just indexing is
slower).

Martin

--
Martin Landa <landa.martin gmail.com> * http://gama.fsv.cvut.cz/~landa
_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev

[Martin Landa]

Hi,

2009/10/18 Martin Landa <[hidden email]>:
> why key column should be restricted only for integers? If you use
> unique strings as keys I don't see any problem (just indexing is
> slower).

or better to say if db driver supports creating index on non-integer
column why to restrict it on GRASS level?

Martin

--
Martin Landa <landa.martin gmail.com> * http://gama.fsv.cvut.cz/~landa
_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev

[Markus Metz-2]

Hamish wrote:
> Martin wrote:
>  
>> in GRASS 7 we can support as key column also non-integer
>> columns,
>>    
>
>
> "just because we can, doesn't mean we should" ...
>  
+1

The key column is used to link attributes to vector features through the
category number assigned to vector features. Category numbers are stored
both in the coor and in the cidx file as integer, changing that would
break forward/backward compatibility completely. Unless there are huge
advantages of cat values being double instead of integer, I would
suggest not to rewrite substantial parts of the vector libs and many
vector modules in order to support cat values as double (or char* or
whatever).

Markus M

_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev

[Markus Neteler]

On Sun, Oct 18, 2009 at 12:30 PM, Hamish <[hidden email]> wrote:
>
> ps- if cleaning up NC data please fix UID.GID in tarball and
> remove cruft from mapset dirs (.tmp, etc.)

If this is a change request please tell us what the problem is :)
- .tmp - I see some spurious 0 byte entries which should be removed.
- but UID.GID?

Markus
_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev

[Helena Mitasova]

I can do an update of the data set to the extent that it does not break
any instruction material (adding metadata is one of items
on my TODO list) so let me know any other things that need fixing.

I have an additional mapset for this location with time series of
coastal lidar data which I can post if it would be useful
(great for showing what you can do with r.series)

I too have no idea what UID.GIS is,

Helena


On Oct 22, 2009, at 6:01 PM, Markus Neteler wrote:

> On Sun, Oct 18, 2009 at 12:30 PM, Hamish <[hidden email]> wrote:
>>
>> ps- if cleaning up NC data please fix UID.GID in tarball and
>> remove cruft from mapset dirs (.tmp, etc.)
>
> If this is a change request please tell us what the problem is :)
> - .tmp - I see some spurious 0 byte entries which should be removed.
> - but UID.GID?
>
> Markus
> _______________________________________________
> grass-dev mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-dev

_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev

[Martin Landa]

Hi,

2009/10/23 Helena Mitasova <[hidden email]>:
> I have an additional mapset for this location with time series of
> coastal lidar data which I can post if it would be useful
> (great for showing what you can do with r.series)

+1

Martin

--
Martin Landa <landa.martin gmail.com> * http://gama.fsv.cvut.cz/~landa
_______________________________________________
grass-dev mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-dev