[FR] clean_duplicates with more control?

[apreshill]

I love this function, but just noticed that in "full markdown mode"

options(
  # build .Rmd to .html (via Pandoc); to build to Markdown, set this option to 'markdown'
  blogdown.method = 'markdown'
)

When I use clean_duplicates, it is deleting all the md files in favor of html (as documented) because my source files are not .Rmarkdown. Would it be possible to add an argument here to keep = markdown or keep = html? Or perhaps it could read whether this option is present in the .Rprofile?

[cderv]

Yes I think this function should be aware of blogdown.method to know what to do. So I would classify this has a bug from something we did not adapt when introducing markdown method.

blogdown/R/check.R

Lines 467 to 474 in 0cde6d6

	clean_duplicates = function(preview = TRUE) in_root({
	x = list_duplicates()
	x1 = with_ext(x, 'Rmd'); i1 = file_exists(x1)
	x2 = with_ext(x, 'Rmarkdown'); i2 = file_exists(x2)
	# if .Rmd exists, delete .md; if .Rmd does not exist or .Rmarkdown exists,
	# delete .html
	x = c(with_ext(x[i1], 'md'), x[i2 | !i1])
	x = x[file_exists(x)]