labels do not appear in 0.9.5

[choiceMaster-ai]

Summary

Behaviour difference between identical GGPLOT code on Mac and Azure platform
Linux Ubuntu. The labels are produced on Mac, but not on Ubuntu.

Minimal code example

Here is the minimum amount of code needed to demonstrate the issue:

p <- p + ggrepel::geom_text_repel(aes(y=cumulative_percent, label = paste0(cumulative_percent,"%") ),
                                     size = 3.0,
                                     box.padding   = 0.0,
                                     point.padding = 0.0,
                                     segment.color = 'grey50',
                                     nudge_y = 10,
                                     direction = "both")

Here is an image of the output produced by the code:

Suggestions

Sorry, I have no idea how to resolve this

Version information

Version 0.9.4 on Mac , but 0.9.5 on Ubuntu

Here is the output from sessionInfo() in my R session:

R version 4.3.2 (2023-10-31)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 22.04.3 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0

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

time zone: Etc/UTC
tzcode source: system (glibc)

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

other attached packages:
 [1] plumber_1.2.1     janitor_2.2.0     redux_1.1.4       stringr_1.5.1     lubridate_1.9.3   htmlwidgets_1.6.4
 [7] reactable_0.4.4   ggrepel_0.9.5     ggtext_0.1.2      ggplot2_3.4.4     formattable_0.2.1 knitr_1.45       
[13] uuid_1.2-0        dplyr_1.1.4       glue_1.7.0        pool_1.0.2        DBI_1.2.1        

loaded via a namespace (and not attached):
 [1] gtable_0.3.4      xfun_0.41         pingr_2.0.3       remotes_2.4.2.1   websocket_1.4.1   processx_3.8.3   
 [7] swagger_3.33.1    vctrs_0.6.5       tools_4.3.2       ps_1.7.6          generics_0.1.3    tibble_3.2.1     
[13] fansi_1.0.6       RSQLite_2.3.5     blob_1.2.4        pkgconfig_2.0.3   lifecycle_1.0.4   farver_2.1.1     
[19] compiler_4.3.2    textshaping_0.3.7 munsell_0.5.0     chromote_0.1.2    snakecase_0.11.1  httpuv_1.6.14    
[25] htmltools_0.5.7   later_1.3.2       pillar_1.9.0      ellipsis_0.3.2    cachem_1.0.8      webshot2_0.1.1   
[31] commonmark_1.9.1  tidyselect_1.2.0  digest_0.6.34     stringi_1.8.3     labeling_0.4.3    fastmap_1.1.1    
[37] grid_4.3.2        colorspace_2.1-0  cli_3.6.2         magrittr_2.0.3    base64enc_0.1-3   utf8_1.2.4       
[43] withr_3.0.0       scales_1.3.0      promises_1.2.1    bit64_4.0.5       timechange_0.3.0  rmarkdown_2.25   
[49] bit_4.0.5         ragg_1.2.7        webutils_1.2.0    memoise_2.0.1     evaluate_0.23     markdown_1.12    
[55] rlang_1.1.3       gridtext_0.1.5    Rcpp_1.0.12       xml2_1.3.6        rstudioapi_0.15.0 jsonlite_1.8.8   
[61] R6_2.5.1          systemfonts_1.0.5
Browse[1]> 

[slowkow]

Version 0.9.4 on Mac , but 0.9.5 on Ubuntu

If you are comparing two versions of ggrepel, then you should not necessarily expect the same output.

Next time, please also consider sharing a complete, minimal reproducible example. In your code, p is left undefined.

[choiceMaster-ai]

Thank you for your response. I have now compared two versions of ggrepel::geom_label_repel on Mac to address the observed discrepancy in functionality. Specifically, version 9.5 does not function correctly on the Mac or Ubuntu, whereas version 9.4 on the Mac does, as evidenced by the accompanying images. Version 9.4 is not available on the Linux Ubuntu platform.

Regarding the comment about p being left undefined in the code, I'm uncertain of its significance or relevance to the reported bug. However, the main concern is the evidential inconsistency in functionality between the two versions.

The evidence presented below clearly illustrates the missing labels between versions 9.5 and 9.4 on the Mac.

I am unable to provide any additional evidence beyond what has already been provided. I encourage consideration of a regression test to further investigate and address the observed discrepancy.

Thank you for your attention to this matter.

Image shows version 9.5 output:

Image shows version 9.4 output:

[choiceMaster-ai]

Thank you for your response. I have now compared two versions of ggrepel::geom_label_repel to address the observed discrepancy in functionality between the versions. Specifically, version 9.5 does not function correctly on the Mac or Ubuntu, whereas version 9.4 on the Mac does, as evidenced by the accompanying images. Version 9.4 is not available on the Linux Ubuntu platform. Regarding the comment about p being left undefined in the code, I'm uncertain of its significance or relevance to the reported bug. However, the main concern is the evidential inconsistency in functionality between the two versions. The evidence presented below clearly illustrates the missing labels between versions 9.5 and 9.4 I am unable to provide any additional evidence beyond what has already been provided. I encourage consideration of a regression test to further investigate and address the observed discrepancy. Thank you for your attention to this matter. Image shows version 9.5 output:  Image shows version 9.4 output:  David F George Mobile: +44 (0)7468 858638  The smart choice starts with choiceMaster <http://www.choicemaster.org/> e-Innovation Centre, University of Wolverhampton, Priorslee, Telford, Shropshire TF2 9FT

…

On 15 Feb 2024, at 01:39, Kamil Slowikowski ***@***.***> wrote: Version 0.9.4 on Mac , but 0.9.5 on Ubuntu If you are comparing two versions of ggrepel, then you should not necessarily expect the same output. Next time, please also consider sharing a complete, minimal reproducible example. In your code, p is left undefined. — Reply to this email directly, view it on GitHub <#257 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATZIJKL3STLLC7ESQ7MLTOLYTVRNBAVCNFSM6AAAAABDJB6YLOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNBVGIZDQMJTGM>. You are receiving this because you authored the thread.

[aphalo]

@choiceMaster-ai It is difficult to help you without code that we can actually run. A reprex does not need to use your confidential data. You can generate a similar plot with other data. How can we know before spending our unpaid free time dealing with your report that the problem is not related to the data you used? How can we be sure that there was nothing else different between the two R installations than the version of 'ggrepel'? For example, are you using exactly the same graphics device in both systems? Your report gives evidence that there is a problem in a specific system to which we do not have access. Without additional information, we are blindfolded in the search for a solution. We need additional help from you in the form of a reprex for dealing efficiently with your bug report. Hopefully, you will be able provide it. Thanks!

p.s.: I see in the first pair of plots that the labels are missing in a plot with a larger font size and truncated title. Have you tried zooming in and out? Does changing the size aesthetic affect the plotting of the labels? (With a reprex we could test this...)

[slowkow]

@choiceMaster-ai I have an inkling that your issue might be the same one as these issues:

• y = Inf does not work in current version of geom_label_repel #252
• Nudge_x argument not working properly #255

Package developers always want complete, minimal, reproducible examples, because they want to run your code on their own machine to reproduce the problem.

My guess is that you might be able to work around this issue in a few different ways:

• Use ggrepel 0.9.4 (or earlier versions that function the way you like).
• Try the development version with devtools::install_github("slowkow/ggrepel") to get 0.9.5.9999, which might solve your issue as it did for the folks in the issues I mentioned above. In the future, when 0.9.6 is released, it should have the same functionality as 0.9.5.9999.

Each time I update ggrepel, I manually inspect the examples in the documentation and confirm that they look good. Since multiple users have reported the same issue with 0.9.5, I will consider adding one ore two examples to the documentation where we use Inf to position labels.

[choiceMaster-ai]

Hi Kamal. Thank you for the follow-up which I appreciate. I do use 0.9.4 on Mac. I will certainly try your suggestion and let you know. Cheers, and all the best. David F George Mobile: +44 (0)7468 858638  The smart choice starts with choiceMaster <http://www.choicemaster.org/> e-Innovation Centre, University of Wolverhampton, Priorslee, Telford, Shropshire TF2 9FT

…

On 15 Feb 2024, at 17:09, Kamil Slowikowski ***@***.***> wrote: @choiceMaster-ai <https://github.com/choiceMaster-ai> I have an inkling that your issue might be the same one as these issues: #252 <#252> #255 <#255> Package developers always want complete, minimal, reproducible examples, because they want to run your code on their own machine to reproduce the problem. My guess is that you might be able to work around this issue in a few different ways: Use ggrepel 0.9.4 (or earlier versions that function the way you like). Try the development version with devtools::install_github("slowkow/ggrepel") to get 0.9.5.9999, which might solve your issue as it did for the folks in the issues I mentioned above. In the future, when 0.9.6 is released, it should have the same functionality as 0.9.5.9999. Each time I update ggrepel, I manually inspect the examples in the documentation and confirm that they look good. Since multiple users have reported the same issue with 0.9.5, I will consider adding one ore two examples to the documentation where we use Inf to position labels. — Reply to this email directly, view it on GitHub <#257 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATZIJKMXQGWM6FGJSIOIF4TYTY6N3AVCNFSM6AAAAABDJB6YLOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNBWGYZDCOBXGQ>. You are receiving this because you were mentioned.

[choiceMaster-ai]

Hi again, Kamil. I did download 0.9.5.9999 on both Linux and Mac platforms but received messages: Linux lazy-load database '/home/dfg/R/x86_64-pc-linux-gnu-library/4.3/ggrepel/R/ggrepel.rdb' is corrupt> Mac

library(ggrepel)

Warning messages: 1: In doTryCatch(return(expr), name, parentenv, handler) : restarting interrupted promise evaluation 2: In get(method, envir = home) : restarting interrupted promise evaluation 3: In get(method, envir = home) : internal error -3 in R_decompress1 I sorry to say I have no idea what they mean technically. Whilst I could run on Mac using 0.9.5.9999, the labels were not produced:  So need to revert to 0.9.4:  I appreciate that a simple example will help you, and I will attempt to produce such. However, whilst this is important to me, at this time I am busy with something urgent for the next week or so. Please bear with me. Thank you. Best regards. David F George Mobile: +44 (0)7468 858638 The smart choice starts with choiceMaster <http://www.choicemaster.org/> e-Innovation Centre, University of Wolverhampton, Priorslee, Telford, Shropshire TF2 9FT

…

On 15 Feb 2024, at 17:09, Kamil Slowikowski ***@***.***> wrote: @choiceMaster-ai <https://github.com/choiceMaster-ai> I have an inkling that your issue might be the same one as these issues: #252 <#252> #255 <#255> Package developers always want complete, minimal, reproducible examples, because they want to run your code on their own machine to reproduce the problem. My guess is that you might be able to work around this issue in a few different ways: Use ggrepel 0.9.4 (or earlier versions that function the way you like). Try the development version with devtools::install_github("slowkow/ggrepel") to get 0.9.5.9999, which might solve your issue as it did for the folks in the issues I mentioned above. In the future, when 0.9.6 is released, it should have the same functionality as 0.9.5.9999. Each time I update ggrepel, I manually inspect the examples in the documentation and confirm that they look good. Since multiple users have reported the same issue with 0.9.5, I will consider adding one ore two examples to the documentation where we use Inf to position labels. — Reply to this email directly, view it on GitHub <#257 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATZIJKMXQGWM6FGJSIOIF4TYTY6N3AVCNFSM6AAAAABDJB6YLOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNBWGYZDCOBXGQ>. You are receiving this because you were mentioned.

[slowkow]

Take your time.

I would recommend starting a new R session before installing new packages. I've seen these kinds of errors before for many different packages, and I always start a new session when I see them.

Also, you can try deleting the rdb file, starting a new R session, and then installing again. That should probably work. Good luck!

[choiceMaster-ai]

Thanks Kamil. Also, thanks for the tips. Cheers, and enjoy the weekend! David. David F George Mobile: +44 (0)7468 858638  The smart choice starts with choiceMaster <http://www.choicemaster.org/> e-Innovation Centre, University of Wolverhampton, Priorslee, Telford, Shropshire TF2 9FT

…

On 16 Feb 2024, at 20:24, Kamil Slowikowski ***@***.***> wrote: Take your time. I would recommend starting a new R session before installing new packages. I've seen these kinds of errors before for many different packages, and I always start a new session when I see them. Also, you can try deleting the rdb file, starting a new R session, and then installing again. That should probably work. Good luck! — Reply to this email directly, view it on GitHub <#257 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATZIJKL2V6YFINPMSXO4UVLYT657FAVCNFSM6AAAAABDJB6YLOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNBZGI4DCMZTGM>. You are receiving this because you were mentioned.

[slowkow]

If you can share a reprex, then I can test if my new patch works for you. Please feel free to re-open this issue when you are ready to share a complete reproducible example for us to run.

[choiceMaster-ai]

Thank you for the email. For my interest, when do you plan to apply the patch and make an updated ggrepel available. All the best. David F George Mobile: +44 (0)7468 858638  The smart choice starts with choiceMaster <http://www.choicemaster.org/> e-Innovation Centre, University of Wolverhampton, Priorslee, Telford, Shropshire TF2 9FT

…

On 16 May 2024, at 16:10, Kamil Slowikowski ***@***.***> wrote: If you can share a reprex, then I can test if my new patch works for you. Please feel free to re-open this issue when you are ready to share a complete reproducible example for us to run. — Reply to this email directly, view it on GitHub <#257 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATZIJKL37XR44MN7SAKEJS3ZCTD6VAVCNFSM6AAAAABDJB6YLOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMJVGUYTGMJUHE>. You are receiving this because you were mentioned.

[slowkow]

Feel free to try the latest code. I have a hunch it will work now.

[choiceMaster-ai]

Thanks - I will download the package and try it. Thanks. David F George Mobile: +44 (0)7468 858638  The smart choice starts with choiceMaster <http://www.choicemaster.org/> e-Innovation Centre, University of Wolverhampton, Priorslee, Telford, Shropshire TF2 9FT

…

On 16 May 2024, at 18:39, Kamil Slowikowski ***@***.***> wrote: Feel free to try the latest code. I have a hunch it will work now. — Reply to this email directly, view it on GitHub <#257 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATZIJKOXHOEO3RM2ABAFWWTZCTVNLAVCNFSM6AAAAABDJB6YLOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMJVHA2DCMZZGQ>. You are receiving this because you were mentioned.