It seems that usmap is missing Shannon County, SD.

Running this...

usmap::fips(state= "SD", county = "Shannon")

...yields this error

Error in fips(state = "SD", county = "Shannon") : 
  Shannon is not a valid county in SD.

Expected behavior
Meanwhile, it works just fine for others:

usmap::fips(state= "SD", county = "Meade")

[1] "46093"

Environment:

• Platform: Linux
• version output:

platform       x86_64-pc-linux-gnu         
arch           x86_64                      
os             linux-gnu                   
system         x86_64, linux-gnu           
status                                     
major          4                           
minor          1.3                         
year           2022                        
month          03                          
day            10                          
svn rev        81868                       
language       R                           
version.string R version 4.1.3 (2022-03-10)
nickname       One Push-Up  

• sessionInfo() output:

R version 4.1.3 (2022-03-10)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Rocky Linux 8.6 (Green Obsidian)

Matrix products: default
BLAS/LAPACK: /usr/lib64/libopenblasp-r0.3.15.so

locale:
[1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8       
[4] LC_COLLATE=en_US.UTF-8     LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
[7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
[10] LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] googlesheets4_1.0.0 ranger_0.13.1       showtext_0.9-5      showtextdb_3.0     
[5] sysfonts_0.8.8      doParallel_1.0.17   iterators_1.0.14    vip_0.3.2          
[9] usemodels_0.2.0     lubridate_1.8.0     foreach_1.5.2       yardstick_1.0.0    
[13] workflowsets_0.2.1  workflows_0.2.6     tune_0.2.0          rsample_0.1.1      
[17] recipes_0.2.0       parsnip_0.2.1       modeldata_0.1.1     infer_1.0.0        
[21] dials_0.1.1         scales_1.2.0        tidymodels_0.2.0    RSocrata_1.7.11-2  
[25] usmap_0.6.0         magrittr_2.0.3      broom_0.8.0         MASS_7.3-57        
[29] patchwork_1.1.1     arrow_8.0.0         forcats_0.5.1       stringr_1.4.0      
[33] dplyr_1.0.9         purrr_0.3.4         readr_2.1.2         tidyr_1.2.0        
[37] tibble_3.1.7        ggplot2_3.3.6       tidyverse_1.3.1    

loaded via a namespace (and not attached):
[1] googledrive_2.0.0  colorspace_2.0-3   ellipsis_0.3.2     class_7.3-20      
[5] readtext_0.81      fs_1.5.2           rstudioapi_0.13    farver_2.1.0      
[9] listenv_0.8.0      furrr_0.3.0        bit64_4.0.5        prodlim_2019.11.13
[13] fansi_1.0.3        xml2_1.3.3         codetools_0.2-18   splines_4.1.3     
[17] cachem_1.0.6       jsonlite_1.8.0     dbplyr_2.2.0       compiler_4.1.3    
[21] httr_1.4.3         backports_1.4.1    fastmap_1.1.0      assertthat_0.2.1  
[25] Matrix_1.4-1       gargle_1.2.0       cli_3.3.0          tools_4.1.3       
[29] gtable_0.3.0       glue_1.6.2         rappdirs_0.3.3     Rcpp_1.0.8.3      
[33] cellranger_1.1.0   DiceDesign_1.9     vctrs_0.4.1        conflicted_1.1.0  
[37] timeDate_3043.102  gower_1.0.0        globals_0.15.0     rvest_1.0.2       
[41] mime_0.12          lifecycle_1.0.1    future_1.26.1      ipred_0.9-13      
[45] vroom_1.5.7        hms_1.1.1          rematch2_2.1.2     curl_4.3.2        
[49] memoise_2.0.1      gridExtra_2.3      rpart_4.1.16       stringi_1.7.6     
[53] lhs_1.1.5          hardhat_1.0.0      lava_1.6.10        rlang_1.0.2       
[57] pkgconfig_2.0.3    lattice_0.20-45    labeling_0.4.2     bit_4.0.4         
[61] tidyselect_1.1.2   usmapdata_0.1.0    parallelly_1.32.0  plyr_1.8.7        
[65] R6_2.5.1           generics_0.1.2     DBI_1.1.2          pillar_1.7.0      
[69] haven_2.5.0        withr_2.5.0        survival_3.3-1     nnet_7.3-17       
[73] future.apply_1.9.0 modelr_0.1.8       crayon_1.5.1       utf8_1.2.2        
[77] tzdb_0.3.0         grid_4.1.3         readxl_1.4.0       data.table_1.14.2 
[81] reprex_2.0.1       digest_0.6.29      openssl_2.0.2      GPfit_1.0-8       
[85] munsell_0.5.0      viridisLite_0.4.0  askpass_1.1  

There should be an option in plot_usmap to display state abbreviation labels (AK, AL, AR, etc.) in the center of each state polygon.

This may look something like this:

plot_usmap(data, ..., labels = true)

labels would be the new parameter and accept true or false values (default being false) indicating whether or not to display labels.

Another new option could be the ability to set label color.

_{Sent with GitHawk}

The Northwestern Hawaiian Islands are erroneously appearing on the map. A fix for this is being worked on.

Currently the data frame which contains coordinates to draw the US state and county boundaries is stored within the usmap package. This data should be extracted into a separate package that can be updated independently of usmap. This will greatly reduce the file size of usmap and will make it easier to develop for with the smaller footprint.

It will also prevent the following NOTE that currently appears during R CMD check (notice extdata size compared to installed size):

> checking installed package size ... NOTE
    installed size is  8.4Mb
    sub-directories of 1Mb or more:
      doc       1.6Mb
      extdata   6.4Mb

To do

• determine name for new package (usmap_data, usmapd, usmapdata, ?) (#39 (comment))
• create repository for new package (#39 (comment))
• extract data and tests related to the map data frame (i.e. the us_map function) (#39 (comment))
• extract data generation scripts (located in data-raw folder) (#39 (comment))
• submit new package to CRAN (#39 (comment))
• once uploaded and functional, update usmap to make use of new package instead of internal data (#49)

Extra

• consider creating more robust script to download shape files from census.gov and automatically create map data frames This would require automated updating of the county-fips or state-fips files and currently there is no easily parseable document providing this information.

usmap has grown substantially in feature set since it was first released. The vignettes should be updated and added to such that all features are clearly explained and made accessible for new users.

_{Sent with GitHawk}

For example:

plot_usmap("counties", size = 0.1)

For example:

plot_usmap(include = .new_england, exclude = "ME", labels = TRUE)

The label for Maine ("ME") will be displayed, despite the state being excluded.

US territories are frequently excluded from US maps. In the spirit of this package, which is to include often forgotten Alaska & Hawaii, it may be a good idea to allow the option to present the inhabited US territories on the map as well.

The territories being considered are:

• American Samoa
• Guam
• Northern Mariana Islands
• Puerto Rico
• US Virgin Islands

These territories all have FIPS codes and official abbreviations so incorporating them into the data set should be seamless.

The difficult part will be moving the coordinates of these places so that they appear on the map in a suitable and aesthetically adequate part of the US map.

Update r-lib based workflows to v2 versions.

These include setup-r and setup-r-dependencies used in test-coverage.yaml and R-CMD-check.yaml.

The following documents need to be added to provide a way for new contributors to understand how to contribute and what is expected of them:

• Code of conduct
• Contributing guidelines
• Issue templates
• Pull request template

This is also to make the repository more complete. As part of this effort, the Wiki should be revamped as well.

The bug

plot_usmap is confused when a column is named "values"

Assume a data.frame as follows...

str(myDF)
$ fips : Factor w/ 30 levels ...
$ values : int ...

plot_usmap(data = myDF)
Error in [.data.frame(map_df, , values) : undefined columns selected

this also fails...

plot_usmap(data = myDF, values = "values" )
Error in [.data.frame(map_df, , values) : undefined columns selected

BUT...this works...

str(myDF)
$ fips : Factor w/ 30 levels ...
$ population : int ...

plot_usmap(data = myDF, values = "population" )

which produces the expected plot

Describe the bug
This used to work in version 0.2.1, now an error is returned.

To Reproduce
Steps to reproduce the behavior:

library(usmap)
library(dplyr)
data(USArrests)
USArrests$full <- rownames(USArrests)
USA50 <- inner_join(us_map(), USArrests)
plot_usmap(data=USA50, values="Murder")
Error in [.data.frame(result, , c(setdiff(names(result), names(data)), :
undefined columns selected

Expected behavior
A map appearing in the plot window

Screenshots
If applicable, add screenshots to help explain your problem.

Smartphone (please complete the following information):

• R version: [4.0.0]
• usmap version: [0.5.0]
• Platform: [Windows]

Additional context
Add any other context about the problem here.

The function would take a data frame as input that contains two columns, the first representing longitude and the second representing latitude (column names don't matter).

The output will be a data frame containing the same data frame/coordinates except with the canonical usmap projection applied to it:

CRS("+proj=laea +lat_0=45 +lon_0=-100 +x_0=0 +y_0=0 +a=6370997 +b=6370997 +units=m +no_defs")

Function name: usmap_proj

Related issue: #13

Version 0.6.0 is currently failing several tests on certain platforms: https://cran.r-project.org/web/checks/check_results_usmap.html

Example:

Version: 0.6.0 
Check: tests 
Result: ERROR 
     Running 'testthat.R' [10s/14s]
    Running the tests in 'tests/testthat.R' failed.
    Complete output:
     > library(testthat)
     > library(usmap)
     > 
     > test_check("usmap")
     [ FAIL 2 | WARN 3 | SKIP 0 | PASS 163 ]
     
     == Failed tests ================================================================
     -- Failure ('test-plot.R:138'): singular regions can be used -------------------
     plot_usmap(regions = "states")$layers not equal to plot_usmap(regions = "state")$layers.
     Component 1: Component 12: Component 5: Component 21: Component 11: 1 string mismatch
     Component 1: Component 12: Component 5: Component 21: Component 12: 1 string mismatch
     Component 2: Component 12: Component 5: Component 21: Component 11: 1 string mismatch
     Component 2: Component 12: Component 5: Component 21: Component 12: 1 string mismatch
     -- Failure ('test-plot.R:140'): singular regions can be used -------------------
     plot_usmap(regions = "counties")$layers not equal to plot_usmap(regions = "county")$layers.
     Component 1: Component 12: Component 5: Component 21: Component 11: 1 string mismatch
     Component 1: Component 12: Component 5: Component 21: Component 12: 1 string mismatch
     Component 2: Component 12: Component 5: Component 21: Component 11: 1 string mismatch
     Component 2: Component 12: Component 5: Component 21: Component 12: 1 string mismatch
     
     [ FAIL 2 | WARN 3 | SKIP 0 | PASS 163 ]
     Error: Test failures
     Execution halted 
Flavor: [r-devel-linux-x86_64-debian-clang](https://www.r-project.org/nosvn/R.check/r-devel-linux-x86_64-debian-clang/usmap-00check.html)

This must be fixed by November 21, 2022 or package will be removed from CRAN. Fix will be included in version 0.6.1.

I get the error listed in the title of this issue when running a simple program:

library("usmap")
library("ggplot2")

fipsCodes <- c("54001","54003")
values <- c(10,20)
mapData <- data.frame (fipsCodes,values,stringsAsFactors=FALSE)
names(mapData) <- c("fips","value")
meanValue <- mean(values)
                  
usmap::plot_usmap(data = mapData, values="value", lines="red",
                  regions="counties", 
                  include = c("WV"),
                  scale_fill_continuous(low="#559999",mid="grey90",high="#886508",midpoint=meanValue),
                  theme(legend.position = "right")) 

Error in if (labels) { : argument is not interpretable as logical

If I remove the line containing the parameter "scale_fill_continuous(...", it works fine.

This is an important piece of projection info that will let others use other dataframes, for instance, having unprojected geographic data (i.e., lat/long coordinates) transform to the right spatial projection. I would include this in the intro/vignette part where you describe which projection is used (i.e., page 17 of https://cran.r-project.org/web/packages/usmap/usmap.pdf).

The county map should be able to display only counties in certain states using the include parameter.

For example,

plot_usmap("counties", include = c("NJ", "PA", "NY"))

This can be fixed at this location in usmap.R:

if (length(include) > 0) {
  if (regions_ %in% c("counties", "county")) {
    df <- df[df$fips %in% include, ]
  } else {
    df <- df[df$full %in% include | df$abbr %in% include | df$fips %in% include, ]
  }
}

The body of the if statement should be changed to:

if (length(include) > 0) {
  df <- df[df$full %in% include | df$abbr %in% include | df$fips %in% include, ]
}

This should work because both the county and state map files have columns fips, full, and abbr.

Updating this function also has the added side-effect of allowing users to specify any combination of states and counties to include using either FIPS code, state abbreviation, state name, or a mixture of the three.

aes_string

aes_string is soft-deprecated in ggplot2 according to https://ggplot2.tidyverse.org/reference/aes_.html, see warnings below.

══ Warnings ════════════════════════════════════════════════════════════════════
── Warning ('test-plot.R:10'): (code run outside of `test_that()`) ─────────────
Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
ℹ Please use `linewidth` instead.
Backtrace:
 1. usmap::plot_usmap("counties", fill = "red")
      at test-plot.R:10:0
 3. ggplot2 (local) `<fn>`(fill = "red", color = "black", size = 0.4, mapping = `<uneval>`)
 4. ggplot2::layer(...)
 5. ggplot2:::deprecate_soft0(...)
── Warning ('test-plot.R:11'): (code run outside of `test_that()`) ─────────────
`aes_string()` was deprecated in ggplot2 3.0.0.
ℹ Please use tidy evaluation ideoms with `aes()`
Backtrace:
 1. usmap::plot_usmap(data = statepop, values = "pop_2015", color = "blue")
      at test-plot.R:11:0
 2. ggplot2::aes_string(x = "x", y = "y", group = "group", fill = values)
 3. ggplot2:::deprecate_soft0("3.0.0", "aes_string()", details = "Please use tidy evaluation ideoms with `aes()`")

size

size is deprecated in ggplot2 3.4.0 and has been replaced with linewidth

Plan

Fixes will be included in v0.6.1 release.

Expected behavior

Passing in a vector (e.g. a list of state abbreviations) to fips should result in a vector of the same length containing the respective FIPS codes.

Actual behavior

A warning is presented, due to the non-vectorized nature of the internal code of fips.

Steps to reproduce behavior

Attempt to pass a vector to fips:

fips(c("AK", "AL", "AR"))
#> Warning message:
#> In if (!(state_ %in% abbr) & !(state_ %in% full)) { :
#>   the condition has length > 1 and only the first element will be used

How to fix

Rework the if-statements to check whether the input is valid. This can be done in conjunction with issue #7 and made into a much more robust system of checking whether the input is states, counties, abbreviations, names, etc.

The x and y in the US map look like they are UTM.

Is it possible to get lat long coordinates instead?

No shapes for this county are contained in the us_counties.csv file.

See https://r-spatial.org/r/2022/04/12/evolution.html. From pkgapi analysis https://github.com/r-lib/pkgapi, this package uses elide() methods, which is used in R/transform.R. The maptools::elide() methods could be moved to sp, but I don't know whether this is worth doing given the restricted use in this package and no detected use elsewhere.

How could I plotting both state and county boundaries on the same map?

the following codes could run smoothly in the past,

counties_df <- us_map(regions = "counties", exclude = c("AK", "HI"))
states_df <- us_map(regions = "states", exclude = c("AK", "HI"))

ggplot() +
  geom_polygon(data = counties_df,
               aes(x = x, y = y, group = group),
               fill = NA, color = "black") +
  geom_polygon(data = states_df,
               aes(x = x, y = y, group = group),
               fill = NA, color = "red") +
  coord_equal()

But now the two do not overlap. Is there anyway to fix this? Thank you so much!

Is your feature request related to a problem? Please describe.
There should be a way to plot both states and counties simultaneously. For example, plotting Alaska as a state with several Alaskan counties also outlined within them (possibly with different aesthetic features).

Describe the solution you'd like
This needs more thought but perhaps a boolean parameter to plot_usmap indicating that the include/exclude parameters that are states should be plotted as states and counties should be plotted as counties (instead of just following what is defined in the regions parameter).

Describe alternatives you've considered
Currently, the only way to achieve something like this is to create two separate plot_usmap objects (one for the states and one for counties) and then combine them by extracting the $layers[[1]] object from the plot_usmap object which is both not ideal and could be unreliable if ggplot2 behavior were to be changed in a future update.

For example:

library(usmap)
library(ggplot2)

states <- plot_usmap("states", include = "CA")
counties <- plot_usmap("counties", include = "06073", color = "red")

ggplot() +
  states$layers[[1]] +
  counties$layers[[1]] +
  counties$theme +
  coord_equal()

Actual Behavior

Currently, the following does not work:

fips("NJ", county = c("Bergen", "Hudson"))

As it outputs the following:

#> [1] "34003"
#> Warning messages:
#> 1: In if (county_ == "") { :
#>   the condition has length > 1 and only the first element will be used
#> 2: In name == county_ :
#>   longer object length is not a multiple of shorter object length
#> 3: In name == paste(county_, "county") :
#>   longer object length is not a multiple of shorter object length

It just results in the FIPS code of the first county entered in the vector, followed by a warning.

Expected Behavior

fips("NJ", county = c("Bergen", "Hudson"))
#> [1] "34003" "34017"

This would allow the ability to quickly convert a column of county names to a FIPS code in a data frame.

Things to Consider

• How to handle multiple states and multiple counties at the same time?
  For example:

fips(state = c("NJ", "NY"), county = c("Bergen", "Hudson", "Westchester")

This would not work unless there was a direct mapping between the state and county parameters. In this case it would be easier to throw an error.

Currently, plotting and merging with the map data requires a data frame containing a fips column. The data should be able to be plotted with a column of state abbreviations or names as well.

This can be easily achieved by converting to FIPS codes with the fips function internally if abbreviations/names are detected.

_{Sent with GitHawk}

Since a transform is applied to the original shape file, the column headers lat and long for the map generated by us_map() are not appropriate. Renaming them to x and y would make more sense.

User should be able to add other shape file layers superimposed on the US map (e.g. roads, rivers, temperature data, etc.).

One way this could work is as follows:

roads <- # load shape file...

plot_usmap("counties") +
  add_layer(roads) +
  add_layer(...)
# etc.

This could work by applying the same spTransform that was applied to the original US map data to the layers passed through to plot_usmap.

Something in the merging process breaks the appearance of the map when plotted. Polygon points are mismatched causing extraneous lines to appear all over the map (especially in county map).

I was surprised that fips_info would sort and remove duplicates from the argument.

Current:

fips_info(c("21","20","20"))

  abbr fips     full
1   KS   20   Kansas
2   KY   21 Kentucky

Expected:

  abbr fips     full
1   KY   21 Kentucky
2   KS   20   Kansas
3   KS   20   Kansas

In other words

 new_fips_info <- function(fips) if (length(fips)!=0)
     merge(list(fips=fips), fips_info(), all.x=TRUE )

This comes up when I want the state name from the fips stored in df$fips. I expected this to work within(df, state <- fips_info(fips)$full). Instead merge(df, fips_info(), all.x=T) %>% rename(state='full') will do.

Please add the proj and CRS to the README, and possibly also to each vignette.

The mapping vignettes (Mapping the US and Advanced Mapping) should include some examples of what is considered valid data and how data frames should be structured so that they can be used with plot_usmap.

Input to usmap_transoform has many duplicate coords (but different other columns). The output of the function removes all the duplicate lat lon rows.

The R CMD check and test coverage workflows cause the following deprecation warning:

The set-env command is deprecated and will be disabled soon. Please upgrade to using Environment Files. For more information see: https://github.blog/changelog/2020-10-01-github-actions-deprecating-set-env-and-add-path-commands/

More info: https://github.blog/changelog/2020-10-01-github-actions-deprecating-set-env-and-add-path-commands/

Describe the bug
When plotting maps I get the following warnings:

Use of `map_df$x` is discouraged. Use `x` instead.
Use of `map_df$y` is discouraged. Use `y` instead.
Use of `map_df$group` is discouraged. Use `group` instead.`

To Reproduce
This is running in RStudio (R version 3.5.3) ggplot 3.3.0, and usmap 0.5.0

Additional context
Package still works as expected, would like to update code to avoid the warnings.

County map should be able to display labels just like state maps.

However, this require further consideration on how to display the labels since they would be much longer than state abbreviations (should "county" be omitted?, sizing, geom_text vs geom_label etc.).

Describe the bug
As I understand it, currently the only way to have county data and state data in the same map is to create the maps separately, extract their layers and add it to a single ggplot object.
However, this method throws an error.

Simplest code to reproduce

ggplot() +
  plot_usmap()$layers[[1]]

Error

Error in FUN(X[[i]], ...) : object 'x' not found

Environment:

• Windows 10 x64
• R v4.0.3
• ggplot2_3.3.2
• usmap_0.5.1

Additional info
This method is also the only way to add multiple levels of data on the same map. So this should be looked into regardless of #26

For example:

# Plot New England region without Maine
plot_usmap("states", include = .new_england, exclude = "ME")

# Plot US map without Alaska and Hawaii (not cool)
plot_usmap("states", exclude = c("AK", "HI"))

The following states may cause issues when plotting with data because the leading zero may be omitted if the input data is numeric rather than a character (e.g. "01" becomes 1 and "01001" becomes 1001). This affects any counties within the following states as well.

Abbr  FIPS  Name
====  ====  ====
AK    02    ALASKA
AL    01    ALABAMA
AR    05    ARKANSAS
AZ    04    ARIZONA
CA    06    CALIFORNIA
CO    08    COLORADO
CT    09    CONNECTICUT

The solution will be to cast the input fips column to characters and make sure they are padded correctly ("00" for states and "00000" for counties).

Is your feature request related to a problem? Please describe.
Users have requested the ability to define visual properties of the defined US Census Bureau regions such as borders and fill colors.

Describe the solution you'd like
A new shape file would probably have to be constructed where each state within the region is merged into a single entity. New parameters would have to be added to plot_usmap to define which regions' visual properties should be modified and how.

Describe alternatives you've considered
None, this idea will require some refinement and planning.

Additional context
Related StackOverflow questions:

• https://stackoverflow.com/questions/72089015/how-to-fill-region-by-different-color-on-us-map-in-r
• https://stackoverflow.com/questions/72543606/how-to-make-census-region-border-thicker-on-usmap-packake-r

Please correct me if I am wrong here, but the only spatial features I see in the maps are to obtain the CRS, and transform the object to a data frame (but not an SPDF). I see there is an option to snap other data frames having the same CRS to the usmap (gg).

Is there a simple way to back-transform the gg object to an SPDF or other spatial format outside gg universe?

Great package, thanks!

I request you add one or both of the following:
a) a data object (retrievable via utils::data()) which returns all FIPs info
b) fips_info() and/or fips() returns all FIPS info when blank, or specifying states="all", etc.

Perhaps this feature is buried elsewhere?

Love the package -- great for quick/easy mapping.

Would it be straightforward to add support for Public Use Microdata Area (PUMA) geography?

That would allow for mapping of PUMA-level values derived from American Community Survey microdata files.

Auxiliary files for census shape data missing and need to be included in data folder.

There are now a lot more functions within usmap than when the project was started. There should be some thought given to the naming of these functions going forward (before we reach a stable 1.0 release).

If the functions are to be changed, they will be renamed by v0.7.0 and the old function names will remain but be deprecated until v1.0.0 (where they will be removed).

When looking up the FIPS code of a county, a state should be provided for ambiguous county names.

e.g.
Orange County, CA
Orange County, FL
Orange County, IN
Orange County, NY
Orange County, NC
Orange County, TX
Orange County, VT
Orange County, VA

Describe the bug
Tests fail on certain r-devel environments due to changes in R discussed here: https://stat.ethz.ch/pipermail/r-devel/2020-December/080172.html

To Reproduce
See logs here: https://cran.r-project.org/web/checks/check_results_usmap.html

Expected behavior
All tests should run properly and successfully on all environments.

Environments:
usmap 0.5.1

r-devel-linux-x86_64-debian-clang
r-devel-linux-x86_64-debian-gcc
r-devel-linux-x86_64-fedora-clang
r-devel-linux-x86_64-fedora-gcc	
r-devel-windows-ix86+x86_64

Additional context
Possible solution here: Ryo-N7/tvthemes#15

Must be resolved by January 8, 2021 to prevent removal from CRAN.

Running usmap::usmap_crs() results in the following warning:

Warning message:
In showSRID(uprojargs, format = "PROJ", multiline = "NO") :
Discarded datum unknown in CRS definition

The associated tests produce the following warnings:

test-transform.R:16: warning: data frame with AK and HI points is transformed
Discarded datum unknown in CRS definition

test-transform.R:16: warning: data frame with AK and HI points is transformed
CRS object has comment, which is lost in output

test-transform.R:16: warning: data frame with AK and HI points is transformed
Discarded datum Unknown based on Normal Sphere (r=6370997) ellipsoid in CRS definition

test-transform.R:16: warning: data frame with AK and HI points is transformed
CRS object has comment, which is lost in output

test-transform.R:16: warning: data frame with AK and HI points is transformed
Discarded datum Unknown based on Normal Sphere (r=6370997) ellipsoid in CRS definition

test-transform.R:32: warning: data frame with AK points is transformed
Discarded datum unknown in CRS definition

test-transform.R:32: warning: data frame with AK points is transformed
CRS object has comment, which is lost in output

test-transform.R:32: warning: data frame with AK points is transformed
Discarded datum Unknown based on Normal Sphere (r=6370997) ellipsoid in CRS definition

test-transform.R:48: warning: data frame with HI points is transformed
Discarded datum unknown in CRS definition

test-transform.R:48: warning: data frame with HI points is transformed
CRS object has comment, which is lost in output

test-transform.R:48: warning: data frame with HI points is transformed
Discarded datum Unknown based on Normal Sphere (r=6370997) ellipsoid in CRS definition

test-transform.R:64: warning: data frame with no AK or HI points is transformed
Discarded datum unknown in CRS definition

Environment

macOS 10.15.7

> version
               _                           
platform       x86_64-apple-darwin17.0     
arch           x86_64                      
os             darwin17.0                  
system         x86_64, darwin17.0          
status                                     
major          4                           
minor          0.2                         
year           2020                        
month          06                          
day            22                          
svn rev        78730                       
language       R                           
version.string R version 4.0.2 (2020-06-22)
nickname       Taking Off Again

> sessionInfo()
R version 4.0.2 (2020-06-22)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Catalina 10.15.7

Matrix products: default
BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

loaded via a namespace (and not attached):
 [1] usmap_0.5.1       rstudioapi_0.11   knitr_1.30        magrittr_1.5      maptools_1.0-2   
 [6] munsell_0.5.0     colorspace_1.4-1  lattice_0.20-41   R6_2.4.1          rlang_0.4.7      
[11] tools_4.0.2       grid_4.0.2        gtable_0.3.0      xfun_0.18         tinytex_0.26     
[16] htmltools_0.5.0   ellipsis_0.3.1    yaml_2.2.1        digest_0.6.25     tibble_3.0.3     
[21] lifecycle_0.2.0   crayon_1.3.4.9000 ggplot2_3.3.2     vctrs_0.3.4       glue_1.4.2       
[26] evaluate_0.14     rmarkdown_2.4     sp_1.4-2          pillar_1.4.6      compiler_4.0.2   
[31] scales_1.1.1      foreign_0.8-80    pkgconfig_2.0.3 

Improve data input flexibility

Currently the data frame that is passed to usmap_transform must have longitude in the first column and latitude in the second column. It might be more natural to allow columns in any order that have the names lon/long/longitude and lat/latitude.

There can also be two optional parameters to specify the longitude/latitude column names.

For example:
If a data.frame contains columns "x" and "y" instead,
usmap_transform(data, longitude_name = "x", latitude_name = "y")

Improve data output

Currently the output data frame contains 4 columns: the two input columns followed by the two output columns. The output columns have the same name as the input columns but with a .1 appended to the end. This does not allow for intuitive use of the function as someone would have to inspect and notice the differences manually.

It might be better include an optional parameter that allows replacement of the input columns (i.e. input data frame has 2 columns named "lon" and "lat", and the output data frame has the same 2 columns, with data transformed). If all 4 columns are to be included in the output, a better convention might be to append _tf or prefix transformed_, etc. (TBD)

Other Fixes

• #32

Please provide metadata or information in the README (or elsewhere) which provides information on the frequency (preferably the times) at which the boundaries are updated.

Would also be really cool to have historical boundaries available (might increase probability of people using this package for time series visualizations of political boundary shifts, eg gerrymandering)