geom_sf(): "polygon edge not found" intermittent error

[johngoldin]

I am using geom_sf to do some county maps based on tigris data. I successfully followed some examples, but when I try to do my local county (New Haven, CT), I got the following error while doing geom_sf():

Error in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
polygon edge not found

I spent some time to create a reproducible error and even learned to use the reprex package. But I didn't actually post the issue. Then I opened my lap top this morning and ran the reprex example again. No error. I couldn't think of any reason why it started working, but it worked so I went on about my business. After several hours of work I hit the same error again, under somewhat different circumstances. Given the earlier inconsistency, I tried running the same call multiple times and did not get the error when the same call was repeated.

I think there is a problem, but I have only a nebulous idea of what the problem is. So I will put my non-reproduced reproducible error code below and maybe I won't be the only person who gets an intermittent "polygon edge not found" error. I realize this is not a terribly helpful issue report.

First, here is a copy of my console showing several variations of a geom_sf() call. I was varying the size parameter because I thought that triggered the error in this case. But if you look closely you can see that I make the identical call to geom_sf() and get an error in one instance and no error in another )with the same parameters (size = 0.1).

ggplot() +

• geom_sf(data = p2) +
• geom_sf(data = newhaven_necta, fill = NA, color = "red")

ggplot() +

• geom_sf(data = p2, size = 1) +
• geom_sf(data = newhaven_necta, fill = NA, color = "red")
  Warning message:
  In getNamespace("grDevices") : reached elapsed time limit

ggplot() +

• geom_sf(data = p2, size = 0.1) +
• geom_sf(data = newhaven_necta, fill = NA, color = "red")
  Error in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
  polygon edge not found

ggplot() +

• geom_sf(data = p2, size = 0.3) +
• geom_sf(data = newhaven_necta, fill = NA, color = "red")

ggplot() +

• geom_sf(data = p2, size = 0.1) +
• geom_sf(data = newhaven_necta, fill = NA, color = "red")

OK, next is the description what I thought was going to be a reproducible error:

I am using the tracts function of the tigris package to get shapefiles from the census. I thought the key factor in whether geom_sf gave me an error was whether or not the tracts functions specified cb = FALSE.

library(tigris)
#> To enable 
#> caching of data, set `options(tigris_use_cache = TRUE)` in your R script or .Rprofile.
#> 
#> Attaching package: 'tigris'
#> The following object is masked from 'package:graphics':
#> 
#>     plot
library(sf)
#> Linking to GEOS 3.6.1, GDAL 2.1.3, proj.4 4.9.3
library(tidyverse)
#> Loading tidyverse: ggplot2
#> Loading tidyverse: tibble
#> Loading tidyverse: tidyr
#> Loading tidyverse: readr
#> Loading tidyverse: purrr
#> Loading tidyverse: dplyr
#> Conflicts with tidy packages ----------------------------------------------
#> filter(): dplyr, stats
#> lag():    dplyr, stats

options(tigris_class = "sf")

# this interior Connecticut county works OK:
hartford_tigris <- tracts("CT", county = "Hartford", cb = TRUE)
#> 
  
ggplot(hartford_tigris) + geom_sf(data = hartford_tigris)

# New Haven works OK if I specifiy cb = FALSE
nh_tigris <- tracts("CT", county = "New Haven", cb = FALSE)
#> 
  
ggplot(nh_tigris) + geom_sf()

# New Haven fails if cb = TRUE.
# Error message I get is:
# Error in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y,  : 
#                      polygon edge not found

nh_tigris_cb <- tracts("CT", county = "New Haven", cb = TRUE)
ggplot(nh_tigris_cb) + geom_sf()

# From the successful example when cb = FALSE, one can see that New Haven county shows with
# a boundary far out into Long Sound. I'm guessing this is related to the polygon edge
# that is causing problems. I tried this for some counties in Rhode Island.
# Washington county fails with "polygon edge not found" when cb = TRUE.
# Newport county does not fail.

# One can examine some other coastal counties. Some have this problem and some don't.
ri_washington_cb <- tracts("RI", county = "Washington", cb = TRUE)
#> 
  
ri_newport_cb <- tracts("RI", county = "Newport", cb = TRUE)
ri_newport <- tracts("RI", county = "Newport")

ggplot(ri_newport_cb) + geom_sf()

ggplot(ri_newport) + geom_sf()

ggplot(ri_washington_cb) + geom_sf()

ggplot(ri_washington) + geom_sf()

Session info

devtools::session_info()
#> Session info -------------------------------------------------------------
#>  setting  value                       
#>  version  R version 3.4.1 (2017-06-30)
#>  system   x86_64, darwin15.6.0        
#>  ui       X11                         
#>  language (EN)                        
#>  collate  en_US.UTF-8                 
#>  tz       America/New_York            
#>  date     2017-08-26
#> Packages -----------------------------------------------------------------
#>  package    * version    date       source                            
#>  assertthat   0.2.0      2017-04-11 CRAN (R 3.4.0)                    
#>  backports    1.1.0      2017-05-22 CRAN (R 3.4.0)                    
#>  base       * 3.4.1      2017-07-07 local                             
#>  bindr        0.1        2016-11-13 CRAN (R 3.4.0)                    
#>  bindrcpp     0.2        2017-06-17 CRAN (R 3.4.0)                    
#>  bitops       1.0-6      2013-08-17 CRAN (R 3.4.0)                    
#>  broom        0.4.2      2017-02-13 CRAN (R 3.4.0)                    
#>  cellranger   1.1.0      2016-07-27 CRAN (R 3.4.0)                    
#>  colorspace   1.3-2      2016-12-14 CRAN (R 3.4.0)                    
#>  compiler     3.4.1      2017-07-07 local                             
#>  curl         2.8.1      2017-07-21 CRAN (R 3.4.1)                    
#>  datasets   * 3.4.1      2017-07-07 local                             
#>  DBI          0.7        2017-06-18 CRAN (R 3.4.0)                    
#>  devtools     1.13.3     2017-08-02 CRAN (R 3.4.1)                    
#>  digest       0.6.12     2017-01-27 CRAN (R 3.4.0)                    
#>  dplyr      * 0.7.2      2017-07-20 CRAN (R 3.4.1)                    
#>  evaluate     0.10.1     2017-06-24 CRAN (R 3.4.1)                    
#>  forcats      0.2.0      2017-01-23 CRAN (R 3.4.0)                    
#>  foreign      0.8-69     2017-06-22 CRAN (R 3.4.1)                    
#>  ggplot2    * 2.2.1.9000 2017-08-24 Github (tidyverse/ggplot2@41f154f)
#>  glue         1.1.1      2017-06-21 CRAN (R 3.4.1)                    
#>  graphics   * 3.4.1      2017-07-07 local                             
#>  grDevices  * 3.4.1      2017-07-07 local                             
#>  grid         3.4.1      2017-07-07 local                             
#>  gtable       0.2.0      2016-02-26 CRAN (R 3.4.0)                    
#>  haven        1.1.0      2017-07-09 CRAN (R 3.4.1)                    
#>  hms          0.3        2016-11-22 CRAN (R 3.4.0)                    
#>  htmltools    0.3.6      2017-04-28 CRAN (R 3.4.0)                    
#>  httr         1.3.1      2017-08-20 cran (@1.3.1)                     
#>  jsonlite     1.5        2017-06-01 CRAN (R 3.4.0)                    
#>  knitr        1.17       2017-08-10 CRAN (R 3.4.1)                    
#>  lattice      0.20-35    2017-03-25 CRAN (R 3.4.1)                    
#>  lazyeval     0.2.0      2016-06-12 CRAN (R 3.4.0)                    
#>  lubridate    1.6.0      2016-09-13 CRAN (R 3.4.0)                    
#>  magrittr     1.5        2014-11-22 CRAN (R 3.4.0)                    
#>  maptools     0.9-2      2017-03-25 CRAN (R 3.4.0)                    
#>  memoise      1.1.0      2017-04-21 CRAN (R 3.4.0)                    
#>  methods    * 3.4.1      2017-07-07 local                             
#>  mnormt       1.5-5      2016-10-15 CRAN (R 3.4.0)                    
#>  modelr       0.1.1      2017-07-24 CRAN (R 3.4.1)                    
#>  munsell      0.4.3      2016-02-13 CRAN (R 3.4.0)                    
#>  nlme         3.1-131    2017-02-06 CRAN (R 3.4.1)                    
#>  parallel     3.4.1      2017-07-07 local                             
#>  pkgconfig    2.0.1      2017-03-21 CRAN (R 3.4.0)                    
#>  plyr         1.8.4      2016-06-08 CRAN (R 3.4.0)                    
#>  psych        1.7.5      2017-05-03 CRAN (R 3.4.1)                    
#>  purrr      * 0.2.3      2017-08-02 CRAN (R 3.4.1)                    
#>  R6           2.2.2      2017-06-17 CRAN (R 3.4.0)                    
#>  rappdirs     0.3.1      2016-03-28 CRAN (R 3.4.0)                    
#>  Rcpp         0.12.12    2017-07-15 CRAN (R 3.4.1)                    
#>  RCurl        1.95-4.8   2016-03-01 CRAN (R 3.4.0)                    
#>  readr      * 1.1.1      2017-05-16 CRAN (R 3.4.0)                    
#>  readxl       1.0.0      2017-04-18 CRAN (R 3.4.0)                    
#>  reshape2     1.4.2      2016-10-22 CRAN (R 3.4.0)                    
#>  rgdal        1.2-8      2017-07-01 CRAN (R 3.4.1)                    
#>  rgeos        0.3-23     2017-04-06 CRAN (R 3.4.0)                    
#>  rlang        0.1.2      2017-08-09 CRAN (R 3.4.1)                    
#>  rmarkdown    1.6        2017-06-15 CRAN (R 3.4.0)                    
#>  rprojroot    1.2        2017-01-16 CRAN (R 3.4.0)                    
#>  rvest        0.3.2      2016-06-17 CRAN (R 3.4.0)                    
#>  scales       0.5.0      2017-08-24 Github (hadley/scales@21bf4c0)    
#>  sf         * 0.5-3      2017-07-30 CRAN (R 3.4.1)                    
#>  sp           1.2-5      2017-06-29 CRAN (R 3.4.1)                    
#>  stats      * 3.4.1      2017-07-07 local                             
#>  stringi      1.1.5      2017-04-07 CRAN (R 3.4.0)                    
#>  stringr      1.2.0      2017-02-18 CRAN (R 3.4.0)                    
#>  tibble     * 1.3.3      2017-05-28 CRAN (R 3.4.0)                    
#>  tidyr      * 0.7.0      2017-08-16 CRAN (R 3.4.1)                    
#>  tidyverse  * 1.1.1      2017-01-27 CRAN (R 3.4.0)                    
#>  tigris     * 0.5.9      2017-08-25 Github (walkerke/tigris@4d6a38a)  
#>  tools        3.4.1      2017-07-07 local                             
#>  udunits2     0.13       2016-11-17 CRAN (R 3.4.0)                    
#>  units        0.4-5      2017-06-15 CRAN (R 3.4.0)                    
#>  utils      * 3.4.1      2017-07-07 local                             
#>  uuid         0.1-2      2015-07-28 CRAN (R 3.4.0)                    
#>  withr        2.0.0      2017-08-19 Github (jimhester/withr@0c4a86d)  
#>  XML          3.98-1.9   2017-06-19 CRAN (R 3.4.1)                    
#>  xml2         1.1.1      2017-01-24 CRAN (R 3.4.0)                    
#>  yaml         2.1.14     2016-11-12 CRAN (R 3.4.0)

[johngoldin]

I have even less idea what triggers the "polygon edge not found" error. In the project that I am working on with tigris maps I have made the same call to geom_sf multiple times with no problem and then suddenly have it switch to throwing this error and no longer able to run. As far as I can tell the calling parameters are identical.

[johngoldin]

Sigh, one more problem with my attempt to create a repeatable example: to run the tigris and tidycensus calls you have to have a Census API key.

[hadley]

This appears to be some grid bug. Unfortunately I haven't been able to reliably reproduce it either.

[chris-prener]

hi @hadley and @johngoldin - I came across this error (and subsequently this issue) while getting ready for a lecture on sf and geom_sf() on Monday in my GIS course.

I have been able to reproduce the error with a set of data that I use for teaching. The data are packaged up and available on GitHub:

devtools::install_github("chris-prener/stlData")

Here is a reprex that I have been using to trigger the error:

library(ggplot2)
library(stlData)

ggplot() +
  geom_sf(data = stlBoundary, fill = "#5d5d5d", color = "#5d5d5d") +
  geom_sf(data = stlHistoric, fill = "#d48a72", color = "#d48a72") +
  geom_sf(data = stlHydro, fill = "#72bcd4", color = "#72bcd4")

The error does not happen every time I re-create the plot. This afternoon, I was able to trigger the error every other time or every third time I created the plot:

> library(ggplot2)
> library(stlData)
> ggplot() +
+   geom_sf(data = stlBoundary, fill = "#5d5d5d", color = "#5d5d5d") +
+   geom_sf(data = stlHistoric, fill = "#d48a72", color = "#d48a72") +
+   geom_sf(data = stlHydro, fill = "#72bcd4", color = "#72bcd4")
> ggplot() +
+   geom_sf(data = stlBoundary, fill = "#5d5d5d", color = "#5d5d5d") +
+   geom_sf(data = stlHistoric, fill = "#d48a72", color = "#d48a72") +
+   geom_sf(data = stlHydro, fill = "#72bcd4", color = "#72bcd4")
Error in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y,  : 
  polygon edge not found

I hope this is helpful for you @hadley in terms of getting to the bottom of what is happening! Thanks for all of your hard work on this package (and others) - they are a joy to use and to teach with.

[tbrambor]

I ran into the same error on this tutorial for making a map in ggplot: https://github.com/grssnbchr/thematic-maps-ggplot2

Since all data and code is on Github, perhaps that is easily replicable. Specifically, the error appears in this code chunk:

p <- ggplot() +
    # municipality polygons
    geom_polygon(data = map_data, aes(fill = avg_age_15, 
                                      x = long, 
                                      y = lat, 
                                      group = group)) +
    # municipality outline
    geom_path(data = map_data, aes(x = long, 
                                   y = lat, 
                                   group = group), 
              color = "white", size = 0.1) +
    coord_equal() +
    # add the previously defined basic theme
    theme_map() +
    labs(x = NULL, 
         y = NULL,         
         title = "Switzerland's regional demographics",
         subtitle = "Average age in Swiss municipalities, 2015", 
         caption = "Geometries: ThemaKart, BFS; Data: BFS, 2016")

The addition of the labs argument triggers the error.

[hadley]

@tbrambor does it happen every time? If so, can you please make a small reprex that I can run by copying and pasting?

[tbrambor]

@hadley . The following reprex shows the error. Oddly, it appears to be related to using a font that is not installed on my system in defining the theme of the plot. If the title and labels of the plot axes are surpressed, the error disappears.

library(ggplot2)
library(mapdata)
#> Loading required package: maps
usa <- map_data("usa")

# Define a theme
theme_map <- function(...) {
  theme_minimal() + theme(text = element_text(family = "Sans"))
}

# Map
ggplot() + geom_polygon(data = usa, aes(x = long, y = lat)) + theme_map()
#> Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
#> no font could be found for family "Sans"

#> Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
#> no font could be found for family "Sans"

#> Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
#> no font could be found for family "Sans"

#> Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
#> no font could be found for family "Sans"

#> Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
#> no font could be found for family "Sans"
#> Error in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : polygon edge not found

[chris-prener]

@tbrambor does this error also occur when you do not use the theme_map() you've defined?

[tbrambor]

No. If the (incorrect) theme mapping is excluded, the plot works fine. It appears the plotting function tries to use the non-installed font which results in the error.

[hadley]

@tbrambor I suspect that's a different problem which generates the same error

[tbrambor]

@hadley. Quite possible. I was just looking for a solution and came across this filed issue here. Feel free to exclude -- I will just adjust the theme to rely on installed fonts and the error is resolved for me. Thank you.

[chris-prener]

@hadley I have a little more info on how to replicate this error consistently. I am able to trigger the Error in grid.Call every time when using colorblindr in the console or in a script:

> library(colorblindr)
> library(ggplot2)
> library(stlData)
> map <- ggplot() +
+         geom_sf(data = stlBoundary, fill = "#5d5d5d", color = "#5d5d5d") +
+         geom_sf(data = stlHistoric, fill = "#d48a72", color = "#d48a72") +
+         geom_sf(data = stlHydro, fill = "#72bcd4", color = "#72bcd4")
> cvd_grid(map)
Error in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y,  : 
  polygon edge not found

What is a little nuts is that it works fine in code chunks in .Rmd files and with reprex (see below). Perhaps this indicates where the issue is?

library(colorblindr)
#> Loading required package: colorspace
#> Loading required package: ggplot2
library(ggplot2)
library(stlData)

map <- ggplot() +
        geom_sf(data = stlBoundary, fill = "#5d5d5d", color = "#5d5d5d") +
        geom_sf(data = stlHistoric, fill = "#d48a72", color = "#d48a72") +
        geom_sf(data = stlHydro, fill = "#72bcd4", color = "#72bcd4")

cvd_grid(map)

Created on 2018-02-18 by the reprex package (v0.2.0).

[chris-prener]

Due to some updates I'm working on to the package I've been using for these reprexes, anyone who wants to investigate this error should install a specific release:

devtools::install_github("chris-prener/stlData", ref = "f51fe10")