auk_split doesn't work

[auman-chan]

I met a problem that auk_split doesn't work . This function only exported files without any rows. Only I splited species before filter would it work.
image.

Are there any solution or suggestion? Thanks!

[mstrimas]

Can you provide the output of sessionInfo() and the code that generated the above output?

[auman-chan]

Well, here is the code I used, and I conducted the sample of EBD from the EBD download website

 library(auk)
 library(dplyr)

 #list.files("test")
ebd_file <- "test/ebd_US-AL-101_202204_202204_relApr-2022.txt"
 ebd_out <- "test/output.txt"
 prefix_spe <- "test/spe_"
 prefix_spe2 <- "test/spe2_"
 ebd_in <- auk_ebd(file = ebd_file) 

 data <- ebd_in %>% 
  auk_complete() %>% 
  auk_year(c(2012, 2022)) %>% 
   auk_duration(duration = c(0, 300)) %>% 
   auk_distance(distance = c(0, 5)) %>% 
   auk_protocol(protocol=c("Stationary","Area"))

 df <- auk_filter(data,file = ebd_out,
                overwrite = T) %>% read_ebd()
                                                                                                                                   
splist <- unique(df$scientific_name)[1:5]
 splist
[1] "Cardinalis cardinalis"    "Mimus polyglottos"        "Poecile carolinensis"    
[4] "Sitta pusilla"            "Thryothorus ludovicianus"

 spe_split <- auk_split(file = ebd_out,
                       species = splist,
                       prefix = prefix_spe,
                        overwrite = T)

 list.files("test")
 [1] "ebd_US-AL-101_202204_202204_relApr-2022.txt" "output.txt"                                 
 [3] "spe_Cardinalis_cardinalis.txt"               "spe_Mimus_polyglottos.txt"                  
 [5] "spe_Poecile_carolinensis.txt"                "spe_Sitta_pusilla.txt"                      
 [7] "spe_Thryothorus_ludovicianus.txt"            "spe2_Cardinalis_cardinalis.txt"             
 [9] "spe2_Mimus_polyglottos.txt"                  "spe2_Poecile_carolinensis.txt"              
[11] "spe2_Sitta_pusilla.txt"                      "spe2_Thryothorus_ludovicianus.txt"          
[13] "test.R"                                     
 file.size("test/spe_Mimus_polyglottos.txt")
[1] 693

The split file with 693 B sizes means it only contains column names.

Here is the information of my sesison:

R version 4.3.3 (2024-02-29)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 22.04.4 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.10.0 
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0

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

time zone: Asia/Shanghai
tzcode source: system (glibc)

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

other attached packages:
[1] dplyr_1.1.4 auk_0.7.1  

loaded via a namespace (and not attached):
 [1] crayon_1.5.2      vctrs_0.6.5       cli_3.6.2         rlang_1.1.4       stringi_1.8.4    
 [6] generics_0.1.3    assertthat_0.2.1  glue_1.7.0        bit_4.0.5         hms_1.1.3        
[11] readxl_1.4.3      writexl_1.4.2     fansi_1.0.6       cellranger_1.1.0  tibble_3.2.1     
[16] tzdb_0.4.0        lifecycle_1.0.4   stringr_1.5.1     compiler_4.3.3    pkgconfig_2.0.3  
[21] rstudioapi_0.15.0 R6_2.5.1          readr_2.1.5       tidyselect_1.2.1  utf8_1.2.4       
[26] parallel_4.3.3    vroom_1.6.5       pillar_1.9.0      magrittr_2.0.3    withr_3.0.0      
[31] tools_4.3.3       bit64_4.0.5   

[auman-chan]

As an alternative, I selected species with the function read_delim_chunked, and remove duplicate group checklists and roll up taxonomy by the function distinct() and filter(). But I hope this pipeline could be fixed.

[mstrimas]

I ran your exact code on the sample EBD file and it appears to be working fine. I have a Mac, so I also tried running it in a Ubuntu Docker container to emulate your environment and also am not having any issues. For example, this is the file for Cardinal that I'm getting
spe_Cardinalis_cardinalis.txt

It's hard to troubleshoot since I can replicate the issue... You might try running in a Docker container as well to test.

[auman-chan]

OK, I will have a try in Windows. Now I split species with the function read_delim_chunked and write_delim, and then imported them by read_ebd.

[auman-chan]

Additionally I have another question.The read_ebd or the auk_unique would only keep the distinct observations by the values of group_identifier, even though it is a missing value. It seems all group_identifier in NA are considered as duplicate and only keep one of them by read_ebd.

However, I don't think observations with the missing values in group_identifier are the duplicate, as they usually are in different locations and recorded in different time.

Is there any further meaning in this procedure?

[mstrimas]

I’m on vacation until June 25 so won’t be able to look into this in detail until then. However, auk_unique() shouldn’t impact rows that have NA for group_identifier. If that is happening, I’ll fix it when I return.

[auman-chan]

It seems that the bug is on the path of the ebd file.

When I read the file in other disks with the path like "/media/username/disk_name/ebird_data/ebd.txt"，the split function works. But when my file in the system disk and evoked by "/home/username/R//imp/ebd.txt", this function returns empty files.

[mstrimas]

That's strange, I'm not sure what that would be happening and I'm not able to reproduce the issue. If you figure it out how to fix it let me know and I can make the change.

[auman-chan]

I know what's wrong. My fiolder name included a whitespace, it could be identified in R (detected by file.exists) but could not in awk. It had better provide a check to avoid this situation, as the error from the shell didn't pass to R console.

On the other hand, I will further confirm the question of auk_unique().