apparent character encoding issue when reading csv on windows

[jhpoelen]

To reproduce:

desired_fields = c("source_taxon_name", "target_taxon_name", "target_specimen_frequency_of_occurrence_percent")
otherkeys = list("limit"=1000, "skip"=0)
Box2_1 <- get_interactions_in_area(bbox = c(-88.422, 23.914, -80.591, 25.910), interactiontype = c("eats"), showfield = desired_fields, returnobservations = T, otherkeys = otherkeys)

On ubuntu / linux, no warnings / errors are seen:

> Box2_1 <- rglobi::get_interactions_in_area(bbox = c(-88.422, 23.914, -80.591, 25.910), interactiontype = c("eats"), showfield = desired_fields, returnobservations = T, otherkeys = otherkeys)
200
200
> head(Box2_1)
  source_taxon_name target_taxon_name
1        Clupanodon          Annelida
2 Galeocerdo cuvier          Annelida
3 Galeocerdo cuvier       Cheloniidae
4 Galeocerdo cuvier    Chelonia mydas
5 Galeocerdo cuvier   Caretta caretta
6 Galeocerdo cuvier           Equidae
  target_specimen_frequency_of_occurrence_percent
1                                              NA
2                                              NA
3                                              NA
4                                              NA
5                                              NA
6                                              NA

However, on windows, warning are observed and no data is made available.

"Warning messages:
1: In read.table(file = file, header = header, sep = sep, quote = quote,  :
  invalid input found on input connection 'C:\Users\XXXX'
2: In read.table(file = file, header = header, sep = sep, quote = quote,  :
  incomplete final line found by readTableHeader on 'C:\Users\XXXX"

This seems consistent with encoding issues seen on windows as described by https://dss.iq.harvard.edu/blog/escaping-character-encoding-hell-r-windows and related stackoverflows posts like: https://stackoverflow.com/questions/16838613/cannot-read-unicode-csv-into-r .

[jhpoelen]

Some suggest to use the readr package at https://github.com/tidyverse/readr . This package is said to deal with these encoding issues.

[jhpoelen]

Please note that warning 2: In read.table(file = file, header = header, sep = sep, quote = quote, : incomplete final line found by readTableHeader on 'C:\Users\XXXX" was caused by the GloBI api not adding a newline to the end of the last line of data. This has been resolved in globalbioticinteractions/globalbioticinteractions@3a089c9 in the GloBI API.

This leaves the warning 1: In read.table(file = file, header = header, sep = sep, quote = quote, : invalid input found on input connection 'C:\Users\XXXX' .

[jhpoelen]

Confirmed fix on Windows. Included in https://github.com/ropensci/rglobi/releases/tag/v0.2.20 .