Suppressing "indexed..." messages when downloading large datasets with get_eurostat

[ferenci-tamas]

When downloading large datasets, get_eurostat emits messages like this:

indexed 0B in 0s, 0B/sindexed 2.15GB in 0s, 2.15GB/s

Is it possible to somehow disable this? I found no quiet, verbose or similar option for get_eurostat.

[pitkant]

Thank you for reporting. I think this is based on a dependency that the eurostat uses to download and/or handle datasets. Can you provide a reproducible example, which dataset were you trying to download? And what is your package version?

[ferenci-tamas]

Sure, sorry for not providing it earlier: eurostat::get_eurostat("hlth_co_proc2") produces this phenomenon for me.

[pitkant]

Thanks, and what package versions are you running on? Please share your sessionInfo() after you have loaded the eurostat package

[ferenci-tamas]

Sure, here is my sessionInfo:

R version 4.5.1 (2025-06-13 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 10 x64 (build 19045)

Matrix products: default
  LAPACK version 3.12.1

locale:
[1] LC_COLLATE=Hungarian_Hungary.utf8  LC_CTYPE=Hungarian_Hungary.utf8    LC_MONETARY=Hungarian_Hungary.utf8 LC_NUMERIC=C                       LC_TIME=Hungarian_Hungary.utf8    

time zone: Europe/Budapest
tzcode source: internal

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

other attached packages:
[1] eurostat_4.0.0

loaded via a namespace (and not attached):
 [1] httr2_1.1.2        htmlwidgets_1.6.4  devtools_2.4.5     remotes_2.5.0      tzdb_0.5.0         vctrs_0.6.5        tools_4.5.1        ISOweek_0.6-2      generics_0.1.4     curl_6.4.0        
[11] tibble_3.3.0       proxy_0.4-27       RefManageR_1.4.0   pkgconfig_2.0.3    KernSmooth_2.23-26 data.table_1.17.6  readxl_1.4.5       assertthat_0.2.1   lifecycle_1.0.4    compiler_4.5.1    
[21] stringr_1.5.1      httpuv_1.6.16      usethis_3.1.0      htmltools_0.5.8.1  class_7.3-23       urlchecker_1.0.1   pillar_1.10.2      later_1.4.2        tidyr_1.3.1        ellipsis_0.3.2    
[31] regions_0.1.8      classInt_0.4-11    cachem_1.1.0       sessioninfo_1.2.3  mime_0.13          countrycode_1.6.1  tidyselect_1.2.1   digest_0.6.37      stringi_1.8.7      dplyr_1.1.4       
[41] pander_0.6.6       purrr_1.0.4        bibtex_0.5.1       rprojroot_2.0.4    fastmap_1.2.0      here_1.0.1         cli_3.6.5          magrittr_2.0.3     pkgbuild_1.4.8     e1071_1.7-16      
[51] clipr_0.8.0        readr_2.1.5        promises_1.3.3     backports_1.5.0    rappdirs_0.3.3     lubridate_1.9.4    timechange_0.3.0   httr_1.4.7         cellranger_1.1.0   hms_1.1.3         
[61] memoise_2.0.1      shiny_1.10.0       miniUI_0.1.2       profvis_0.4.0      rlang_1.1.6        Rcpp_1.0.14        xtable_1.8-4       glue_1.8.0         xml2_1.3.8         pkgload_1.4.0     
[71] rstudioapi_0.17.1  jsonlite_2.0.0     R6_2.6.1           plyr_1.8.9         fs_1.6.6          

[pitkant]

Ok, thanks. I couldn't replicate the message in this specific case but I saw a similar message when downloading the TOC. I think emitting that specific message is related to readr::read_tsv using vroom::vroom in the background.

For the time being you can make readr messages silent with the following option:

options(readr.show_progress = FALSE)

Check the option state:

options("readr.show_progress")

But ultimately I think it would be worth it to go through all messages that are outputted by the eurostat package and make them optional through the verbose / quiet argument.

[ferenci-tamas]

I couldn't replicate the message in this specific case

One possible reason is that the message gets deleted from the console. You can nevertheless see it with capture.output(eurostat::get_eurostat("hlth_co_proc2")). Of course only if you run it first time in the R session when it is not yet cached.

(You might ask why this issue is relevant given this? The answer is that while it is deleted from the console, it'll still appear when rendered to a markdown document!)

But ultimately I think it would be worth it to go through all messages that are outputted by the eurostat package and make them optional through the verbose / quiet argument.

Exactly, that'd be nice, especially for the markdown scenario.

Thank you very much!

[pitkant]

I couldn't replicate the message in this specific case

One possible reason is that the message gets deleted from the console. You can nevertheless see it with capture.output(eurostat::get_eurostat("hlth_co_proc2")). Of course only if you run it first time in the R session when it is not yet cached.

(You might ask why this issue is relevant given this? The answer is that while it is deleted from the console, it'll still appear when rendered to a markdown document!)

But ultimately I think it would be worth it to go through all messages that are outputted by the eurostat package and make them optional through the verbose / quiet argument.

Exactly, that'd be nice, especially for the markdown scenario.

Thank you very much!

Thank you for the detailed response. I agree with this.

Just as a note to self: I was wondering for a bit why the progress text was included in a markdown document since the readr::read_tsv default is that "[the progress bar] will only display in an interactive session and not while knitting a document" but I noticed that it was hardcoded as TRUE in the eurostat functions set_eurostat_toc and get_eurostat_raw...

[ferenci-tamas]

Thank you very much @pitkant! Indeed, likely that is causing these to appear even in markdown documents.

[pitkant]

@ake123 could you have a look at this to add a verbose argument to different functions, with verbose = TRUE being the default and verbose = FALSE of course suppressing all the messages?