[R-sig-Geo] dissolve internal borders of polygons using st_union and group_by

[Roger Bivand]

On Thu, 17 Oct 2019, Marta Rufino wrote:

> Dear Roger,
>
> Thank you very much for your quick and dedicated response.
>
> Please state all versions:
>> sessionInfo()
>> sf_extSoftVersion()
>
> See bellow. I though it was a trivial issue, thus sorry for not reporting
> it earlier.
>
>
>> With an updated system, most of your code just does not work for me.
>
> Just updated everything, r included, and it still runs in my machine -
> perhaps because I failed to update to proj6 and I am still using proj4 - I
> will have to dedicate more time to overcome it. Is there sites that explain
> win-dummies installation?

You don't need to update yet, but modern PROJ will upset many workflows.

>
>
>> You
>> are looking for sf::aggregate():
>> kk <- aggregate(world_map, list(world_map$continent), head, n=1)
>> plot(st_geometry(kk))
>>
>
> This will be then a fourth option:
> # Answer from ROGER:
> kk4 <- aggregate(world_map, list(world_map$continent), head, n=1)
>
> object.size(kk4)
> kk4 %>%
>  ggplot()+
>  geom_sf()
>
> The sp_transform was simply to check if it was workable the produced file
> (in other cases, the code functioned, but then I could not work with the
> files produced) - so ok to remove it or change it.
>
> You example using rgeos is really nice and I am grateful for it, but I
> really wanted to understand how to do it overall so I can apply in other
> situations I am working with.
> I don't understand the difference between all those options, that
> apparently are giving the same result, but not so much in fact.

My attempts to use st_precision() to try to find a precision level 
that removed the artefacts ended in segmentation faults, so the current 
best workaround is as shown to use rgeos, until the correct sf incantation 
is forthcoming.

Roger

>
> Any further ideas on the subject?
>
> Thank you once again,
> Best wishes,
> M.
>
>
>> sessionInfo()
> R version 3.6.1 (2019-07-05)
> Platform: x86_64-w64-mingw32/x64 (64-bit)
> Running under: Windows >= 8 x64 (build 9200)
>
> Matrix products: default
>
> locale:
> [1] LC_COLLATE=English_United Kingdom.1252
> [2] LC_CTYPE=English_United Kingdom.1252
> [3] LC_MONETARY=English_United Kingdom.1252
> [4] LC_NUMERIC=C
> [5] LC_TIME=English_United Kingdom.1252
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods
> [7] base
>
> other attached packages:
> [1] rgeos_0.5-2         ggthemes_4.2.0      ggrepel_0.8.1
> [4] ggplot2_3.2.1       tidyr_1.0.0         scales_1.0.0
> [7] dplyr_0.8.3         reshape2_1.4.3      viridis_0.5.1
> [10] viridisLite_0.3.0   units_0.6-5         rnaturalearth_0.1.0
> [13] tmap_2.3-1          mapview_2.7.0       plotKML_0.5-9
> [16] lubridate_1.7.4     zoo_1.8-6           rasterVis_0.46
> [19] latticeExtra_0.6-28 RColorBrewer_1.1-2  lattice_0.20-38
> [22] raster_3.0-7        maptools_0.9-8      sf_0.8-0
> [25] gstat_2.0-3         rgdal_1.4-6         sp_1.3-1
>
> loaded via a namespace (and not attached):
> [1] colorspace_1.4-1   class_7.3-15       colorRamps_2.3
> [4] leaflet_2.0.2      htmlTable_1.13.2   satellite_1.0.1
> [7] base64enc_0.1-3    dichromat_2.0-0    rstudioapi_0.10
> [10] hexbin_1.27.3      fansi_0.4.0        codetools_0.2-16
> [13] splines_3.6.1      knitr_1.25         zeallot_0.1.0
> [16] Formula_1.2-3      tmaptools_2.0-2    cluster_2.1.0
> [19] png_0.1-7          shiny_1.4.0        compiler_3.6.1
> [22] backports_1.1.5    assertthat_0.2.1   Matrix_1.2-17
> [25] fastmap_1.0.1      lazyeval_0.2.2     cli_1.1.0
> [28] later_1.0.0        acepack_1.4.1      htmltools_0.4.0
> [31] tools_3.6.1        gtable_0.3.0       glue_1.3.1
> [34] Rcpp_1.0.2         vctrs_0.2.0        leafsync_0.1.0
> [37] crosstalk_1.0.0    lwgeom_0.1-7       xfun_0.10
> [40] stringr_1.4.0      mime_0.7           lifecycle_0.1.0
> [43] XML_3.98-1.20      MASS_7.3-51.4      promises_1.1.0
> [46] parallel_3.6.1     yaml_2.2.0         gridExtra_2.3
> [49] aqp_1.17           rpart_4.1-15       reshape_0.8.8
> [52] stringi_1.4.3      plotrix_3.7-6      e1071_1.7-2
> [55] checkmate_1.9.4    intervals_0.15.1   rlang_0.4.0
> [58] pkgconfig_2.0.3    pixmap_0.4-11      RSAGA_1.3.0
> [61] purrr_0.3.2        htmlwidgets_1.5.1  tidyselect_0.2.5
> [64] plyr_1.8.4         magrittr_1.5       R6_2.4.0
> [67] Hmisc_4.2-0        DBI_1.0.0          pillar_1.4.2
> [70] foreign_0.8-72     withr_2.1.2        shapefiles_0.7
> [73] xts_0.11-2         survival_2.44-1.1  nnet_7.3-12
> [76] tibble_2.1.3       spacetime_1.2-2    crayon_1.3.4
> [79] utf8_1.1.4         KernSmooth_2.23-15 grid_3.6.1
> [82] data.table_1.12.4  FNN_1.1.3          dismo_1.1-4
> [85] digest_0.6.21      classInt_0.4-1     webshot_0.5.1
> [88] xtable_1.8-4       httpuv_1.5.2       stats4_3.6.1
> [91] munsell_0.5.0
>> sf_extSoftVersion()
>          GEOS           GDAL         proj.4 GDAL_with_GEOS
>       "3.6.1"        "2.2.3"        "4.9.3"         "true"
>    USE_PROJ_H
>       "false"
>>
>
>>
>>
>

-- 
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: Roger.Bivand using nhh.no
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en