Jennifer (Jenny) Bryan

Such a swap (e.g. penguins for iris) will affect readxl documentation and perhaps tests, at the very least. And it's possible that other packages use that file. So it's a...

A lightweight solution for this in "user space" is to pass a custom function to the new `.name_repair` argument in the dev version of readxl. ``` r devtools::load_all(".") #> Loading...

This will need to become a minimal, clean example using readxl, installed from CRAN or this repo, before I can consider evaluating it. Chances are this is a problem caused...

@renkun-ken Have a look at [`excel_sheets()`](https://readxl.tidyverse.org/reference/excel_sheets.html)

I'm not thinking about readxl at the moment, but I can imagine what's going on here because there's a similar phenomenon in Google Sheets (https://github.com/tidyverse/googlesheets4/issues/172). The long-term solution would be...

@arnyeinstein Here's what I get when I read the xlsx you provided, using your code: ``` r library(dplyr) library(readxl) df % filter(SUTcode > 1) %>% print(n = Inf) #> #...

@jessekrailler I advise that you format that cell as *text* in Excel, rather than numeric. If the scale of your problem makes such manual work impractical, then you could import...

The conversion of Excel numbers to text happens at the C level and there's no way of knowing how many decimal places to show, i.e. how many decimal places were...

Asking around, it seems in the medium-term, i.e. next release, I could try to improve readxl's automatic guessing of precision for a "number in a text column". There are ways...

@jessekrailler I experimented with your file. It seems that the gratuitous decimal places are stored in the case of a numeric cell to which you then apply a text format....