Question

我有一个由sf对象组成的列表列的小标题。这些sf对象中的每一个都有一个或多个多边形行。类似于我取消嵌套小节列表的方式，我希望取消sf对象的嵌套。

数据看起来像这样

library(sf)
library(tibble)
poly <- st_polygon(list(cbind(c(0,3,3,0,0),c(0,0,3,3,0))))

a = st_sf(st_sfc(poly,poly))
b = st_sf(st_sfc(poly,poly))

df <- tibble(name=c('a','b'),poly=list(a,b))

我希望结果是一个如下所示的sf对象：

|name | geometry |
|-----|----------|
|a    | POLYGON  |
|a    | POLYGON  |
|b    | POLYGON  |
|b    | POLYGON  |

但是，当我使用unnest时，我收到一个错误，并且由于某种原因找不到unnest.sf。

编辑：我正在使用SF版本0.7-4

谢谢！

本

Answer 1

使用新术语，unchop()比unnest()多。这是一种方法：

library(tidyverse)
library(sf)
#> Linking to GEOS 3.6.1, GDAL 2.2.3, PROJ 4.9.3

poly <- st_polygon(list(cbind(c(0,3,3,0,0),c(0,0,3,3,0))))

a = st_sf(st_sfc(poly,poly))
b = st_sf(st_sfc(poly,poly))

df <- tibble(name=c('a','b'),poly=list(a,b))

unchop_poly <- function(data, col) {
  mutate(data, {{col}} := map({{col}}, ~split(., seq_len(nrow(.))))) %>%
    unchop({{col}})
}
df2 <- df %>% 
  unchop_poly(poly)

df2
#> # A tibble: 4 x 2
#>   name  poly            
#>   <chr> <list>          
#> 1 a     <df[,1] [1 x 1]>
#> 2 a     <df[,1] [1 x 1]>
#> 3 b     <df[,1] [1 x 1]>
#> 4 b     <df[,1] [1 x 1]>

df2$poly
#> [[1]]
#> Simple feature collection with 1 feature and 0 fields
#> geometry type:  POLYGON
#> dimension:      XY
#> bbox:           xmin: 0 ymin: 0 xmax: 3 ymax: 3
#> epsg (SRID):    NA
#> proj4string:    NA
#>               st_sfc.poly..poly.
#> 1 POLYGON ((0 0, 3 0, 3 3, 0 ...
#> 
#> [[2]]
#> Simple feature collection with 1 feature and 0 fields
#> geometry type:  POLYGON
#> dimension:      XY
#> bbox:           xmin: 0 ymin: 0 xmax: 3 ymax: 3
#> epsg (SRID):    NA
#> proj4string:    NA
#>               st_sfc.poly..poly.
#> 1 POLYGON ((0 0, 3 0, 3 3, 0 ...
#> 
#> [[3]]
#> Simple feature collection with 1 feature and 0 fields
#> geometry type:  POLYGON
#> dimension:      XY
#> bbox:           xmin: 0 ymin: 0 xmax: 3 ymax: 3
#> epsg (SRID):    NA
#> proj4string:    NA
#>               st_sfc.poly..poly.
#> 1 POLYGON ((0 0, 3 0, 3 3, 0 ...
#> 
#> [[4]]
#> Simple feature collection with 1 feature and 0 fields
#> geometry type:  POLYGON
#> dimension:      XY
#> bbox:           xmin: 0 ymin: 0 xmax: 3 ymax: 3
#> epsg (SRID):    NA
#> proj4string:    NA
#>               st_sfc.poly..poly.
#> 1 POLYGON ((0 0, 3 0, 3 3, 0 ...

^{由reprex package（v0.3.0）于2019-09-19创建}

Answer 2

您可以使用data.table将列表按组绑定在一起，然后再转换为sf对象

library(sf)
library(tibble)
poly <- st_polygon(list(cbind(c(0,3,3,0,0),c(0,0,3,3,0))))
poly2 <- st_polygon(list(cbind(c(0,1,1,0,0),c(0,0,1,1,0))))

a = st_sf(st_sfc(poly,poly))
b = st_sf(st_sfc(poly2,poly2))

df <- tibble(name=c('a','b'),poly=list(a,b))

library(data.table)

setDT( df )[, rbindlist( poly ), by = name ] %>% 
  sf::st_as_sf()


# Simple feature collection with 4 features and 1 field
# geometry type:  POLYGON
# dimension:      XY
# bbox:           xmin: 0 ymin: 0 xmax: 3 ymax: 3
# epsg (SRID):    NA
# proj4string:    NA
# name             st_sfc.poly..poly.
# 1    a POLYGON ((0 0, 3 0, 3 3, 0 ...
# 2    a POLYGON ((0 0, 3 0, 3 3, 0 ...
# 3    b POLYGON ((0 0, 1 0, 1 1, 0 ...
# 4    b POLYGON ((0 0, 1 0, 1 1, 0 ...

如何在sf对象的列表列上使用unnest？

2 个答案: