Because I want to download ALL files, I gave as scan_rule +*.*

I the log I see and I have the impression all "unknown" "non-standard"
file-types are downloaded as text/html which is of course wrong. This probably
explains the reason why .ABC becomes first .html.tmp and later .html

In the log I also see that for some files the extention is changed from
Uppercase to Lowercase (example ZIP becomes zip)

Is there a way to avoid those two behaviours and download everything as-is ?