geom_text_repel() not working with coord_polar()

[jakeyeung]

Hi slowkow, nice package by the way. I think my code was working before (about one month ago) but recently I got some issues.

I find that geom_text_repel() does not work when I use coord_polar().

Here is an example:

set.seed(0)
test <- data.frame(
  R = runif(n = 15, min = 0.5, max = 1), 
  phi = runif(n = 15, min = 0, max = 24),
  Label = sample(x = letters, 15, replace = FALSE)
)

# works
ggplot(test, aes(x = R, y = phi, label = Label)) +
  coord_polar(theta = "y") +
  geom_text() +
  expand_limits(x = 0) +
  scale_y_continuous(limits = c(0, 24), breaks = seq(6, 24, 6))

# doesnt work:
# Error in if (text_box[1] <= point_pos[1] && point_pos[1] <= text_box[3] &&  : 
# missing value where TRUE/FALSE needed
ggplot(test, aes(x = R, y = phi, label = Label)) +
  coord_polar(theta = "y") + geom_text_repel() +
  expand_limits(x = 0) +
  scale_y_continuous(limits = c(0, 24), breaks = seq(6, 24, 6))

The example code should look like this:

My sessionInfo() is:

> R version 3.3.1 (2016-06-21)
Platform: x86_64-redhat-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)
locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    LC_PAPER=en_US.UTF-8      
 [8] LC_NAME=C                  LC_ADDRESS=C               LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
attached base packages:
[1] grid      stats     graphics  grDevices utils     datasets  methods   base     
other attached packages:
[1] ggrepel_0.6.3  glmnet_2.0-5   foreach_1.4.3  Matrix_1.2-7.1 ggplot2_2.1.0  dplyr_0.5.0    reshape2_1.4.2 hash_2.2.6    
loaded via a namespace (and not attached):
 [1] Rcpp_0.12.7          AnnotationDbi_1.28.2 magrittr_1.5         BiocGenerics_0.12.1  IRanges_2.0.1        munsell_0.4.3        lattice_0.20-34      colorspace_1.2-7     R6_2.2.0            
[10] stringr_1.1.0        GenomeInfoDb_1.2.5   plyr_1.8.4           tools_3.3.1          parallel_3.3.1       Biobase_2.26.0       gtable_0.2.0         DBI_0.5-1            iterators_1.0.8     
[19] lazyeval_0.2.0       assertthat_0.1       tibble_1.2           codetools_0.2-15     S4Vectors_0.4.0      RSQLite_1.0.0        labeling_0.3         stringi_1.1.2        scales_0.4.0        
[28] stats4_3.3.1         XML_3.98-1.4    

[slowkow]

Jake, thanks for the issue report.

Could I ask you to please display the figures in your issue so we can see
how it looks?

I hadn't considered polar coordinates when developing this package.

I would happily review a pull request that implements this feature.

On Nov 7, 2016 5:42 AM, "Jake Yeung" notifications@github.com wrote:

Hi,

I find that geom_text_repel() does not work when I use coord_polar().

Here is an example code:
`
set.seed(0)
test <- data.frame(R = runif(n = 15, min = 0.5, max = 1),
phi = runif(n = 15, min = 0, max = 24),
Label = sample(x = letters, 15, replace = FALSE))
works fine

ggplot(test, aes(x = R, y = phi, label = Label)) + coord_polar(theta =
"y") + geom_text() + expand_limits(x = 0) + scale_y_continuous(limits =
c(0, 24), breaks = seq(6, 24, 6))
doesnt work Error in if (text_box[1] <= point_pos[1] && point_pos[1] <=
text_box[3] && : missing value where TRUE/FALSE needed

ggplot(test, aes(x = R, y = phi, label = Label)) + coord_polar(theta =
"y") + geom_text_repel() + expand_limits(x = 0) + scale_y_continuous(limits
= c(0, 24), breaks = seq(6, 24, 6))
`

My infoSession() is:

R version 3.3.1 (2016-06-21)
Platform: x86_64-redhat-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)

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

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

other attached packages:
[1] ggrepel_0.6.3 glmnet_2.0-5 foreach_1.4.3 Matrix_1.2-7.1 ggplot2_2.1.0
dplyr_0.5.0 reshape2_1.4.2 hash_2.2.6

loaded via a namespace (and not attached):
[1] Rcpp_0.12.7 AnnotationDbi_1.28.2 magrittr_1.5 BiocGenerics_0.12.1
IRanges_2.0.1 munsell_0.4.3 lattice_0.20-34 colorspace_1.2-7 R6_2.2.0
[10] stringr_1.1.0 GenomeInfoDb_1.2.5 plyr_1.8.4 tools_3.3.1
parallel_3.3.1 Biobase_2.26.0 gtable_0.2.0 DBI_0.5-1 iterators_1.0.8
[19] lazyeval_0.2.0 assertthat_0.1 tibble_1.2 codetools_0.2-15
S4Vectors_0.4.0 RSQLite_1.0.0 labeling_0.3 stringi_1.1.2 scales_0.4.0
[28] stats4_3.3.1 XML_3.98-1.4

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#56, or mute the thread
https://github.com/notifications/unsubscribe-auth/AAMzMqa1l8KTwj1dE--Ez5KyiqOShgNeks5q7wCHgaJpZM4KrEO1
.

[jakeyeung]

Hi Slowkow, thanks for the quick response.

The thing is, the package actually worked previously, here is an example of geom_text_repel() working (generated last month):

EDIT:
Actually, an older version of ggrepel solves this issue, specifically version 0.6.1.

Running the plotting script in version 0.6.1 gives the correct output:

I am not familiar with how makeContent.textrepelgrob() is used, but I noticed in version 0.6.3, the variable x$x and x$y outputs NaN whereas in version 0.6.1 it outputs finite numbers.

Jake

[slowkow]

@jakeyeung Could I ask you to please reopen this issue if version 0.6.4 does not work for you?