In the golang docs for encoding/csv, WriteAll is stated to Write the given CSV records, and then to call Flush on the underlying writer.
Now, I'm confused as to whether I should detect errors during WriteAll via its return value, or the Error() method. Let me explain why.
Write returns error. This makes me think I can detect errors based on its return value.Flush doesn't return error, and the docs explicitly say "To check if an error occurred during the Flush, call Error."WriteAll mentions nothing about error handling, but it returns error, and states it performs both Write and Flush. The docs include an example that goes like this:
w := csv.NewWriter(os.Stdout)
w.WriteAll(records) // calls Flush internally
if err := w.Error(); err != nil {
log.Fatalln("error writing csv:", err)
}
Error() itself states: "Error reports any error that has occurred during a previous Write or Flush." (emphasis mine).
I looked at the code and I saw some shenanigans I wasn't familiar with, so I'm turning the question to SO.
Which is it, then? Should I look at the returned error? Should I use w.Error()? Should I use both? What's the difference?
As far as i can see you can use both, i would propose to use the error from
err := w.WriteAll(records)
this is more common in go.
All errors returned are coming from the *bufio.Writer, and if it fails it will always return an error. So if w.WriteAll returns an error a call to Error() will also return an error.