Add warning/error if at least one column name is missing

[stefan-mueller]

Describe the bug

If a column name of a data.frame is missing, constructing a corpus() still works. Yet, summary(), for example, results in an error due to undefined docvars. I was wondering whether we might add a test to the corpus construction stage to check that column names are not NA?

Reproducible code

library(quanteda)

df <- data.frame(text = letters, docvar1 = paste(letters, letters), num = 1:26, 
  stringsAsFactors = F)

colnames(df) <- c("text", "docvar1")

corp <- corpus(df)

corp
#> Corpus consisting of 26 documents and 1 docvar.

summary(corp)
#> Warning in nsentence.character(object, ...): nsentence() does not correctly
#> count sentences in all lower-cased text
#> Error in `[.data.frame`(x, !is_system & !is_text): undefined columns selected

Expected behavior

Already return an error (or at least a warning) when a corpus is constructed from an object with missing column names.

R version 3.5.0 (2018-04-23)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Sierra 10.12.6

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

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

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

other attached packages:
[1] reprex_0.1.2   quanteda_1.3.0

loaded via a namespace (and not attached):
 [1] Rcpp_0.12.17       compiler_3.5.0     pillar_1.2.1       formatR_1.5       
 [5] plyr_1.8.4         tools_3.5.0        stopwords_0.9.0    digest_0.6.15     
 [9] debugme_1.1.0      lubridate_1.7.4    evaluate_0.10.1    memoise_1.1.0     
[13] tibble_1.4.2       gtable_0.2.0       lattice_0.20-35    rlang_0.2.1       
[17] Matrix_1.2-14      fastmatch_1.1-0    rstudioapi_0.7     yaml_2.1.19       
[21] blogdown_0.6       xfun_0.1           withr_2.1.2        stringr_1.3.1     
[25] knitr_1.20         devtools_1.13.5    rprojroot_1.3-2    grid_3.5.0        
[29] data.table_1.11.2  R6_2.2.2           rmarkdown_1.9      bookdown_0.7      
[33] callr_2.0.3        whisker_0.3-2      ggplot2_2.2.1      spacyr_0.9.9      
[37] magrittr_1.5       backports_1.1.2    scales_0.5.0       htmltools_0.3.6   
[41] assertthat_0.2.0   colorspace_1.3-2   stringi_1.2.2      lazyeval_0.2.1    
[45] RcppParallel_4.4.0 munsell_0.4.3      crayon_1.3.4  

[koheiw]

It is easy to add checks, but I wonder when you get a data.frame with column names in natural usage of R?

[stefan-mueller]

I didn't encounter this problem personally, but received a request from another user why summary() failed. The problem occurred when colnames were set manually and when there are more columns in the data.frame than specified in colnames(). Maybe you just decide whether we need to add such checks or not. The problem probably does not occur frequently.