`"No encoding supplied"` message with `polite::scrape()`

[NONONOexe]

Hi,

When using polite::scrape() to fetch page content, I encounter the message: "No encoding supplied: defaulting to UTF-8".

# OK: no message
scraped_page_content <- polite::scrape(bow = polite_session, content = "text/html; charset=UTF-8")

# NG: message
scraped_page_content <- polite::scrape(bow = polite_session)

Why does this specification (the content argument) affect a GET request this way? (perhaps as discussed in Issue #25)

[dmi3kno]

It would be good to have a reproducible example on this. Can you show a website which reports this message (i.e. where encoding can not be guessed by httr from the context)

[NONONOexe]

I have confirmed this situation on the Geofabrik website. I believe it can be reproduced with the following code:

polite_session <- polite::bow(
  url = "https://download.geofabrik.de/"
)

scraped_page_content <- polite::scrape(
  bow = polite_session
)
## --> "No encoding supplied: defaulting to UTF-8."

scraped_page_content <- polite::scrape(
  bow = polite_session,
  content = "text/html; charset=UTF-8"
)
## --> **No message**

Please let me know if you need any additional information.