[R-sig-Geo] Problems with rbind(list(), makeUniqueIDs=T)

[Bacou, Melanie]

Robert,
Great, thanks for pointing `raster::bind()` out as well. Guess I should 
re-read the vignette periodically!
--Mel.


On 7/23/2016 5:04 PM, Robert J. Hijmans wrote:
> The raster way would be:
>
> library(raster)
> m <- lapply(c("TZA", "UGA", "GHA"), function(x) getData("GADM",
> country=x, level=1))
> m <- do.call(bind, m)
>
> Robert
>
> On Sat, Jul 23, 2016 at 4:10 AM, Bacou, Melanie <mel at mbacou.com> wrote:
>> Edzer, Rolf,
>> Many thanks for the clarification!
>> Just to confirm that Rolf's `do.call()` example works for me using the
>> latest GitHub commit.
>>
>>   > m <- lapply(c("TZA", "ETH", "GHA"), function(x) getData("GADM",
>> country=x, level=1))
>> # with sp_1.2-3
>>   > m <- do.call(rbind, c(m, list(makeUniqueIDs=TRUE)))
>> Error in SpatialPolygonsDataFrame(pl, df) :
>>     row.names of data and Polygons IDs do not match
>> # with sp_1.2-4
>>   > m <- do.call(rbind, c(m, list(makeUniqueIDs=TRUE)))
>>   > class(m)
>> [1] "SpatialPolygonsDataFrame"
>>
>> --Mel.
>>
>>
>> On 7/23/2016 5:28 AM, Edzer Pebesma wrote:
>>> On 23/07/16 05:00, Bacou, Melanie wrote:
>>>> Thanks! I wonder why the original syntax passing a list used to work (I
>>>> believe).
>>> It shouldn't have worked:
>>>
>>>> rbind(list(1,2,3))
>>>        [,1] [,2] [,3]
>>> [1,] 1    2    3
>>>> rbind(1,2,3)
>>>        [,1]
>>> [1,]    1
>>> [2,]    2
>>> [3,]    3
>>>
>>>> For a much longer (and unknown) list of SpatialPolygonsDataFrames could
>>>> an approach using do.call() work instead? I tried but:
>>>>
>>>>    > m <- do.call(rbind, m, makeUniqueIDs=T)
>>>> Error in do.call(rbind, m, makeUniqueIDs = T) :
>>>>      unused argument (makeUniqueIDs = T)
>>> See Rolf's email for the correct way of using do.call; with sp from CRAN
>>> it still doesn't work. As an alternative to installing sp from github
>>> directly, you could source the relevant rbind methods from github in
>>> your script:
>>>
>>> rbind.SpatialPolygons = function(..., makeUniqueIDs = FALSE) {
>>>        dots = list(...)
>>>        names(dots) <- NULL
>>>        stopifnot(identicalCRS(dots))
>>>        # checkIDSclash(dots)
>>>        pl = do.call(c, lapply(dots, function(x) slot(x, "polygons")))
>>>        if (makeUniqueIDs)
>>>                pl = makeUniqueIDs(pl)
>>>        SpatialPolygons(pl, proj4string = CRS(proj4string(dots[[1]])))
>>> }
>>> rbind.SpatialPolygonsDataFrame <- function(..., makeUniqueIDs = TRUE) {
>>>        dots = list(...)
>>>        names(dots) <- NULL # bugfix Clement Calenge 100417
>>>        lst = lapply(dots, function(x) as(x, "SpatialPolygons"))
>>>        lst$makeUniqueIDs = makeUniqueIDs
>>>        pl = do.call(rbind.SpatialPolygons, lst)
>>>        df = do.call(rbind, lapply(dots, function(x) x at data))
>>>        SpatialPolygonsDataFrame(pl, df, match.ID = FALSE)
>>> }
>>>
>>>> # This works though but takes an extra step
>>>>    > spChFIDs(m[[1]]) <- paste0("A", seq(length(m[[1]])))
>>>>    > spChFIDs(m[[2]]) <- paste0("B", seq(length(m[[2]])))
>>>>    > spChFIDs(m[[3]]) <- paste0("C", seq(length(m[[3]])))
>>>>    > m <- do.call(rbind, m)
>>>>    > class(m)
>>>> [1] "SpatialPolygonsDataFrame"
>>>> attr(,"package")
>>>> [1] "sp"
>>>>
>>>> --Mel.
>>>>
>>>> On 7/22/2016 5:53 PM, Edzer Pebesma wrote:
>>>>> The correct call to rbind would be
>>>>>
>>>>> mm = rbind(m[[1]], m[[2]], m[[3]], makeUniqueIDs=T)
>>>>>
>>>>> with sp on CRAN this doesn't work; with the version on github it does.
>>>>>
>>>>> An alternative with sp from CRAN is to make the IDs unique by hand:
>>>>>
>>>>> spChFIDs(m[[1]]) <- paste0("A", seq(length(m[[1]])))
>>>>> spChFIDs(m[[2]]) <- paste0("B", seq(length(m[[2]])))
>>>>> spChFIDs(m[[3]]) <- paste0("C", seq(length(m[[3]])))
>>>>> mm = rbind(m[[1]], m[[2]], m[[3]])
>>>>>
>>>>>
>>>>> On 22/07/16 22:55, Bacou, Melanie wrote:
>>>>>> Hi,
>>>>>> I'm getting weird results trying to rbind a list of
>>>>>> SpatialPolygonsDataFrames with R 3.2.1 and raster 2.5.8. I believe the
>>>>>> code below used to merge all 3 country boundaries, but instead I now get
>>>>>> a list with 6 elements (incl. 3 logical TRUE). Am I doing something wrong?
>>>>>>
>>>>>> Thx, --Mel.
>>>>>>
>>>>>>> library(raster)
>>>>>>> m <- lapply(c("TZA", "UGA", "GHA"), function(x) getData("GADM",
>>>>>> country=x, level=1))
>>>>>>> m <- rbind(m, makeUniqueIDs=T)
>>>>>>> sapply(m, class)
>>>>>> [1] "SpatialPolygonsDataFrame" "logical" "SpatialPolygonsDataFrame"
>>>>>> [4] "logical"                  "SpatialPolygonsDataFrame" "logical"
>>>>>>
>>>>>>> sessionInfo()
>>>>>> R version 3.2.1 (2015-06-18)
>>>>>> Platform: x86_64-w64-mingw32/x64 (64-bit)
>>>>>> Running under: Windows 7 x64 (build 7601) Service Pack 1
>>>>>>
>>>>>> locale:
>>>>>> [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United
>>>>>> States.1252
>>>>>> [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
>>>>>> [5] LC_TIME=English_United States.1252
>>>>>>
>>>>>> attached base packages:
>>>>>> [1] stats     graphics  grDevices utils     datasets  methods base
>>>>>>
>>>>>> other attached packages:
>>>>>> [1] raster_2.5-8 sp_1.2-3     rj_2.0.5-2
>>>>>>
>>>>>> loaded via a namespace (and not attached):
>>>>>> [1] rj.gd_2.0.0-1   Rcpp_0.12.6     grid_3.2.1 lattice_0.20-33
>>>>>>
>>>>> _______________________________________________
>>>>> R-sig-Geo mailing list
>>>>> R-sig-Geo at r-project.org
>>>>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>>>       [[alternative HTML version deleted]]
>>>>
>>>> _______________________________________________
>>>> R-sig-Geo mailing list
>>>> R-sig-Geo at r-project.org
>>>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>>>
>>>
>>> _______________________________________________
>>> R-sig-Geo mailing list
>>> R-sig-Geo at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>
>>          [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-Geo mailing list
>> R-sig-Geo at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo