Quanteda: import files from Mac

[Ale123]

Hi!
On my Mac desktop I have a folder with many files in txt. I'm trying to use the "quanteda" package but I haven't understood how to upload them. With tm it was easy, with quanteda / readtext I didn't find a way to load files from desktop. Can anyone help me, please?
Thanks and best wishes to all!

[technocrat]

The vignette for readtext provides an example you can adapt

library(readtext)
DATA_DIR <- system.file("extdata/", package = "readtext")
(rt1 <- readtext(paste0(DATA_DIR, "/txt/UDHR/*")))
#> readtext object consisting of 13 documents and 0 docvars.
#> # Description: df[,2] [13 × 2]
#>   doc_id            text                         
#>   <chr>             <chr>                        
#> 1 UDHR_chinese.txt  "\"世界人权宣言\n联合国\"..."
#> 2 UDHR_czech.txt    "\"VŠEOBECNÁ \"..."          
#> 3 UDHR_danish.txt   "\"Den 10. de\"..."          
#> 4 UDHR_english.txt  "\"Universal \"..."          
#> 5 UDHR_french.txt   "\"Déclaratio\"..."          
#> 6 UDHR_georgian.txt "\"FLFVBFYBC \"..."          
#> # … with 7 more rows

^{Created on 2019-12-28 by the reprex package (v0.3.0)}

For DATA_DIR of *.txt files in your desktop (assuming that you have only *.txt files in your desktop.

DATA_DIR <- system.file("/Users/YOURS/Desktop")
(rt1 <- readtext(DATA_DIR))

[Ale123]

Thank you very much! I apologize: I have a folder, "texts", on my Desktop. I need the files.txt in the folder... Can it works DATA_DIR <- system.file("/Users/YOURS/Desktop/folder")?
Thanks!

[technocrat]

If the folder is named texts, you can use

DATA_DIR <- system.file("/Users/YOURS/Desktop/texts")
(rt1 <- readtext(DATA_DIR/*))

[Ale123]

@technocrat

(rt1 <- readtext(paste0(DATA_DIR, "/txt/UDHR/*")))

Thanks! The first is clear - DATA_DIR <- system.file("/Users/YOURS/Desktop/texts") - but it doesn't work - (rt1 <- readtext(DATA_DIR/*)) -

(rt1 <- readtext(DATA_DIR/))
Errore: unexpected '' in "(rt1 <- readtext(DATA_DIR/*"

[Ale123]

Example: I can load the same files folder using tm, in this way:

> library(tm)
Carico il pacchetto richiesto: NLP
> cname <- file.path("~", "Desktop", "texts")
> cname
[1] "~/Desktop/texts"
> dir(cname)
 [1] "BIN 2017 1.txt"         "BIN 2017 2.txt"         "BIN 2017 3.txt"         "BIN 2017 4.txt"         "BIN 2018 1.txt"         "BIN 2018 2.txt"        
>  docs <- VCorpus(DirSource(cname)) 
> summary(docs)
                       Length Class             Mode
BIN 2017 1.txt         2      PlainTextDocument list
BIN 2017 2.txt         2      PlainTextDocument list
BIN 2017 3.txt         2      PlainTextDocument list
BIN 2017 4.txt         2      PlainTextDocument list
BIN 2018 1.txt         2      PlainTextDocument list
>  inspect(docs)
<<VCorpus>>
Metadata:  corpus specific: 0, document level (indexed): 0
Content:  documents: 32

And then I can analyze it.

I would like to do the same thing using quanteda, but it doesn't work:

> library(readtext)
> DATA_DIR <- system.file("extdata/", package = "readtext")
> DATA_DIR <- system.file("~", "Desktop", "texts")
> (rt1 <- readtext(DATA_DIR/*))
Errore: unexpected '*' in "(rt1 <- readtext(DATA_DIR/*"
> library(readtext)
> DATA_DIR <- system.file("extdata/", package = "readtext")
> DATA_DIR <- system.file("~", "Desktop", "texts")
> (rt1 <- readtext(paste0(DATA_DIR, "/txt/texts/*")))
Error in list_files(file, ignore_missing, TRUE, verbosity) : 
  File '' does not exist.
> 

I looked for tutorials and examples, but the examples concern URL, not folders on disk: https://cran.r-project.org/web/packages/quanteda/vignettes/quickstart.html.

[technocrat]

Instead of ~ use the full path name: /Users/rc/Desktop/texts, for example