[R-sig-Geo] colocated cokriging with empty neighbourhood causes crash

[Tom Gottfried]

Wow! Thanks for the quick reaction.
Tom

Am 26.06.2011 21:54, schrieb Edzer Pebesma:
> Well spotted! I submitted an updated gstat version (0.9-82) that fixes
> this bug to CRAN.
>
> Best regards,
>
> On 06/24/2011 05:24 PM, Tom Gottfried wrote:
>> Dear list,
>>
>> while doing local colocated cokriging it seems that, if some points to
>> be predicted have no value of the primary variable in the neighbourhood,
>> this causes R to crash with the messages shown below (script with
>> modified code from ASDAR book to generate the crash attached).
>> Should it be expected that NA is predicted at the respective points (as
>> I understood, this is what ordinary kriging and cokriging do when the
>> neighbourhood is empty)?
>>
>> regards,
>> Tom
>>
>> Output at crash-time:
>>
>> Intrinsic Correlation found. Good.
>> [using ordinary cokriging]
>>
>>   *** caught segfault ***
>> address 0x50, cause 'memory not mapped'
>>
>> Traceback:
>>   1: .Call("gstat_predict", as.integer(nrow(as.matrix(new.X))),
>> as.double(as.vector(raw$locations)),     as.vector(new.X),
>> as.integer(block.cols), as.vector(block),     as.vector(bl_weights),
>> as.integer(nsim), as.integer(BLUE))
>>   2: predict.gstat(g.cc, meuse.grid)
>>   3: predict(g.cc, meuse.grid)
>>
>> Possible actions:
>> 1: abort (with core dump, if enabled)
>> 2: normal R exit
>> 3: exit R without saving workspace
>> 4: exit R saving workspace
>> Selection:
>>
>>
>>
>> _______________________________________________
>> R-sig-Geo mailing list
>> R-sig-Geo at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

-- 
Technische Universität München
Department für Pflanzenwissenschaften
Lehrstuhl für Grünlandlehre
Alte Akademie 12
85350 Freising / Germany
Phone: ++49 (0)8161 715324
Fax:   ++49 (0)8161 713243
email: tom.gottfried at wzw.tum.de
http://www.wzw.tum.de/gruenland