[GRASS-dev] v.net.allpairs seems broken

[Markus Metz]

On Wed, Oct 17, 2012 at 10:52 AM, Moritz Lennert
<mlennert at club.worldonline.be> wrote:
> On 16/10/12 15:34, Michael Barton wrote:
>>
>> Fixing this makes the module usable. The flaw is near to a bug.
>
>
> Well, you can use the module as-is. You just have to make sure that
> nodes and arcs have different cat values, e.g.:
>
> #check cat values in arcs layer (streets_wake here):
> v.category streets_waks op=report
>
> #take the existing cat values of the points in schools_wake and add 50000
> (because the cat values in streets_wake go up to just under 50000)
>
> v.category schools_wake out=schools op=sum cat=50000
>
> #now link the nodes with their new categories to the arcs
>
> v.net streets_wake points=schools op=connect thresh=200 out=network nlayer=1
>
> #then launch your analysis indicating the cat values of the points
> v.net.allpairs network cats=50000-50167 out=allpairs

I would prefer to update the documentation for GRASS 6. I started
backporting the changes to the v.net.* modules coming from the GSoC
project to my local copy of devbr6, but the changes are too
fundamental. IOW I did not yet find the time to do proper backporting
to GRASS 6, which must obviously include proper testing. Therefore I
would suggest to update the manuals for GRASS 6 with your example and
equivalent for the other affected modules.

In this case, if someone (probably me) manages to properly backport
the changes in trunk I would support the exception from the rule to
break script compatibility. These modules have been added only
recently anyway, and do not conform to GRASS vector network analysis
standards.
>
> So it is possible, but should be documented.
>
> I do agree though that it's a real PINA compared to the simple two-layer
> system in other modules.

I guess you mean PIT* (your choice).

Markus M

>
> Moritz
>
>>
>> +1
>>
>> Michael ____________________ C. Michael Barton Director, Center for
>> Social Dynamics&  Complexity Professor of Anthropology, School of
>> Human Evolution&  Social Change Arizona State University
>>
>>
>> voice:  480-965-6262 (SHESC), 480-727-9746 (CSDC) fax:
>> 480-965-7671 (SHESC),  480-727-0709 (CSDC) www:
>> http://www.public.asu.edu/~cmbarton, http://csdc.asu.edu
>>
>>
>>
>> On Oct 16, 2012, at 4:12 AM, Moritz
>> Lennert<mlennert at club.worldonline.be> wrote:
>>
>>> Or we decide to break the rule to not change options and backport
>>> the alayer/nlayer patches from grass 7 to grass 6, as it is such an
>>> obvious design flaw compared to the general logic of the v.net.*
>>> modules.
>>>
>>> What do you think ?
>>
>>
>